INSTRUCTION
stringlengths 1
46.3k
| RESPONSE
stringlengths 75
80.2k
|
|---|---|
Get the number of rows in the Dataset.
Returns
-------
number_of_rows : int
The number of rows in the Dataset. | def num_data(self):
"""Get the number of rows in the Dataset.
Returns
-------
number_of_rows : int
The number of rows in the Dataset.
"""
if self.handle is not None:
ret = ctypes.c_int()
_safe_call(_LIB.LGBM_DatasetGetNumData(self.handle,
ctypes.byref(ret)))
return ret.value
else:
raise LightGBMError("Cannot get num_data before construct dataset") |
Get the number of columns (features) in the Dataset.
Returns
-------
number_of_columns : int
The number of columns (features) in the Dataset. | def num_feature(self):
"""Get the number of columns (features) in the Dataset.
Returns
-------
number_of_columns : int
The number of columns (features) in the Dataset.
"""
if self.handle is not None:
ret = ctypes.c_int()
_safe_call(_LIB.LGBM_DatasetGetNumFeature(self.handle,
ctypes.byref(ret)))
return ret.value
else:
raise LightGBMError("Cannot get num_feature before construct dataset") |
Get a chain of Dataset objects.
Starts with r, then goes to r.reference (if exists),
then to r.reference.reference, etc.
until we hit ``ref_limit`` or a reference loop.
Parameters
----------
ref_limit : int, optional (default=100)
The limit number of references.
Returns
-------
ref_chain : set of Dataset
Chain of references of the Datasets. | def get_ref_chain(self, ref_limit=100):
"""Get a chain of Dataset objects.
Starts with r, then goes to r.reference (if exists),
then to r.reference.reference, etc.
until we hit ``ref_limit`` or a reference loop.
Parameters
----------
ref_limit : int, optional (default=100)
The limit number of references.
Returns
-------
ref_chain : set of Dataset
Chain of references of the Datasets.
"""
head = self
ref_chain = set()
while len(ref_chain) < ref_limit:
if isinstance(head, Dataset):
ref_chain.add(head)
if (head.reference is not None) and (head.reference not in ref_chain):
head = head.reference
else:
break
else:
break
return ref_chain |
Save Dataset to a text file.
This format cannot be loaded back in by LightGBM, but is useful for debugging purposes.
Parameters
----------
filename : string
Name of the output file.
Returns
-------
self : Dataset
Returns self. | def dump_text(self, filename):
"""Save Dataset to a text file.
This format cannot be loaded back in by LightGBM, but is useful for debugging purposes.
Parameters
----------
filename : string
Name of the output file.
Returns
-------
self : Dataset
Returns self.
"""
_safe_call(_LIB.LGBM_DatasetDumpText(
self.construct().handle,
c_str(filename)))
return self |
Free Booster's Datasets.
Returns
-------
self : Booster
Booster without Datasets. | def free_dataset(self):
"""Free Booster's Datasets.
Returns
-------
self : Booster
Booster without Datasets.
"""
self.__dict__.pop('train_set', None)
self.__dict__.pop('valid_sets', None)
self.__num_dataset = 0
return self |
Set the network configuration.
Parameters
----------
machines : list, set or string
Names of machines.
local_listen_port : int, optional (default=12400)
TCP listen port for local machines.
listen_time_out : int, optional (default=120)
Socket time-out in minutes.
num_machines : int, optional (default=1)
The number of machines for parallel learning application.
Returns
-------
self : Booster
Booster with set network. | def set_network(self, machines, local_listen_port=12400,
listen_time_out=120, num_machines=1):
"""Set the network configuration.
Parameters
----------
machines : list, set or string
Names of machines.
local_listen_port : int, optional (default=12400)
TCP listen port for local machines.
listen_time_out : int, optional (default=120)
Socket time-out in minutes.
num_machines : int, optional (default=1)
The number of machines for parallel learning application.
Returns
-------
self : Booster
Booster with set network.
"""
_safe_call(_LIB.LGBM_NetworkInit(c_str(machines),
ctypes.c_int(local_listen_port),
ctypes.c_int(listen_time_out),
ctypes.c_int(num_machines)))
self.network = True
return self |
Add validation data.
Parameters
----------
data : Dataset
Validation data.
name : string
Name of validation data.
Returns
-------
self : Booster
Booster with set validation data. | def add_valid(self, data, name):
"""Add validation data.
Parameters
----------
data : Dataset
Validation data.
name : string
Name of validation data.
Returns
-------
self : Booster
Booster with set validation data.
"""
if not isinstance(data, Dataset):
raise TypeError('Validation data should be Dataset instance, met {}'
.format(type(data).__name__))
if data._predictor is not self.__init_predictor:
raise LightGBMError("Add validation data failed, "
"you should use same predictor for these data")
_safe_call(_LIB.LGBM_BoosterAddValidData(
self.handle,
data.construct().handle))
self.valid_sets.append(data)
self.name_valid_sets.append(name)
self.__num_dataset += 1
self.__inner_predict_buffer.append(None)
self.__is_predicted_cur_iter.append(False)
return self |
Reset parameters of Booster.
Parameters
----------
params : dict
New parameters for Booster.
Returns
-------
self : Booster
Booster with new parameters. | def reset_parameter(self, params):
"""Reset parameters of Booster.
Parameters
----------
params : dict
New parameters for Booster.
Returns
-------
self : Booster
Booster with new parameters.
"""
if any(metric_alias in params for metric_alias in ('metric', 'metrics', 'metric_types')):
self.__need_reload_eval_info = True
params_str = param_dict_to_str(params)
if params_str:
_safe_call(_LIB.LGBM_BoosterResetParameter(
self.handle,
c_str(params_str)))
self.params.update(params)
return self |
Update Booster for one iteration.
Parameters
----------
train_set : Dataset or None, optional (default=None)
Training data.
If None, last training data is used.
fobj : callable or None, optional (default=None)
Customized objective function.
For multi-class task, the score is group by class_id first, then group by row_id.
If you want to get i-th row score in j-th class, the access way is score[j * num_data + i]
and you should group grad and hess in this way as well.
Returns
-------
is_finished : bool
Whether the update was successfully finished. | def update(self, train_set=None, fobj=None):
"""Update Booster for one iteration.
Parameters
----------
train_set : Dataset or None, optional (default=None)
Training data.
If None, last training data is used.
fobj : callable or None, optional (default=None)
Customized objective function.
For multi-class task, the score is group by class_id first, then group by row_id.
If you want to get i-th row score in j-th class, the access way is score[j * num_data + i]
and you should group grad and hess in this way as well.
Returns
-------
is_finished : bool
Whether the update was successfully finished.
"""
# need reset training data
if train_set is not None and train_set is not self.train_set:
if not isinstance(train_set, Dataset):
raise TypeError('Training data should be Dataset instance, met {}'
.format(type(train_set).__name__))
if train_set._predictor is not self.__init_predictor:
raise LightGBMError("Replace training data failed, "
"you should use same predictor for these data")
self.train_set = train_set
_safe_call(_LIB.LGBM_BoosterResetTrainingData(
self.handle,
self.train_set.construct().handle))
self.__inner_predict_buffer[0] = None
is_finished = ctypes.c_int(0)
if fobj is None:
if self.__set_objective_to_none:
raise LightGBMError('Cannot update due to null objective function.')
_safe_call(_LIB.LGBM_BoosterUpdateOneIter(
self.handle,
ctypes.byref(is_finished)))
self.__is_predicted_cur_iter = [False for _ in range_(self.__num_dataset)]
return is_finished.value == 1
else:
if not self.__set_objective_to_none:
self.reset_parameter({"objective": "none"}).__set_objective_to_none = True
grad, hess = fobj(self.__inner_predict(0), self.train_set)
return self.__boost(grad, hess) |
Boost Booster for one iteration with customized gradient statistics.
Note
----
For multi-class task, the score is group by class_id first, then group by row_id.
If you want to get i-th row score in j-th class, the access way is score[j * num_data + i]
and you should group grad and hess in this way as well.
Parameters
----------
grad : 1-D numpy array or 1-D list
The first order derivative (gradient).
hess : 1-D numpy array or 1-D list
The second order derivative (Hessian).
Returns
-------
is_finished : bool
Whether the boost was successfully finished. | def __boost(self, grad, hess):
"""Boost Booster for one iteration with customized gradient statistics.
Note
----
For multi-class task, the score is group by class_id first, then group by row_id.
If you want to get i-th row score in j-th class, the access way is score[j * num_data + i]
and you should group grad and hess in this way as well.
Parameters
----------
grad : 1-D numpy array or 1-D list
The first order derivative (gradient).
hess : 1-D numpy array or 1-D list
The second order derivative (Hessian).
Returns
-------
is_finished : bool
Whether the boost was successfully finished.
"""
grad = list_to_1d_numpy(grad, name='gradient')
hess = list_to_1d_numpy(hess, name='hessian')
assert grad.flags.c_contiguous
assert hess.flags.c_contiguous
if len(grad) != len(hess):
raise ValueError("Lengths of gradient({}) and hessian({}) don't match"
.format(len(grad), len(hess)))
is_finished = ctypes.c_int(0)
_safe_call(_LIB.LGBM_BoosterUpdateOneIterCustom(
self.handle,
grad.ctypes.data_as(ctypes.POINTER(ctypes.c_float)),
hess.ctypes.data_as(ctypes.POINTER(ctypes.c_float)),
ctypes.byref(is_finished)))
self.__is_predicted_cur_iter = [False for _ in range_(self.__num_dataset)]
return is_finished.value == 1 |
Rollback one iteration.
Returns
-------
self : Booster
Booster with rolled back one iteration. | def rollback_one_iter(self):
"""Rollback one iteration.
Returns
-------
self : Booster
Booster with rolled back one iteration.
"""
_safe_call(_LIB.LGBM_BoosterRollbackOneIter(
self.handle))
self.__is_predicted_cur_iter = [False for _ in range_(self.__num_dataset)]
return self |
Get the index of the current iteration.
Returns
-------
cur_iter : int
The index of the current iteration. | def current_iteration(self):
"""Get the index of the current iteration.
Returns
-------
cur_iter : int
The index of the current iteration.
"""
out_cur_iter = ctypes.c_int(0)
_safe_call(_LIB.LGBM_BoosterGetCurrentIteration(
self.handle,
ctypes.byref(out_cur_iter)))
return out_cur_iter.value |
Get number of models per iteration.
Returns
-------
model_per_iter : int
The number of models per iteration. | def num_model_per_iteration(self):
"""Get number of models per iteration.
Returns
-------
model_per_iter : int
The number of models per iteration.
"""
model_per_iter = ctypes.c_int(0)
_safe_call(_LIB.LGBM_BoosterNumModelPerIteration(
self.handle,
ctypes.byref(model_per_iter)))
return model_per_iter.value |
Get number of weak sub-models.
Returns
-------
num_trees : int
The number of weak sub-models. | def num_trees(self):
"""Get number of weak sub-models.
Returns
-------
num_trees : int
The number of weak sub-models.
"""
num_trees = ctypes.c_int(0)
_safe_call(_LIB.LGBM_BoosterNumberOfTotalModel(
self.handle,
ctypes.byref(num_trees)))
return num_trees.value |
Evaluate for data.
Parameters
----------
data : Dataset
Data for the evaluating.
name : string
Name of the data.
feval : callable or None, optional (default=None)
Customized evaluation function.
Should accept two parameters: preds, train_data,
and return (eval_name, eval_result, is_higher_better) or list of such tuples.
For multi-class task, the preds is group by class_id first, then group by row_id.
If you want to get i-th row preds in j-th class, the access way is preds[j * num_data + i].
Returns
-------
result : list
List with evaluation results. | def eval(self, data, name, feval=None):
"""Evaluate for data.
Parameters
----------
data : Dataset
Data for the evaluating.
name : string
Name of the data.
feval : callable or None, optional (default=None)
Customized evaluation function.
Should accept two parameters: preds, train_data,
and return (eval_name, eval_result, is_higher_better) or list of such tuples.
For multi-class task, the preds is group by class_id first, then group by row_id.
If you want to get i-th row preds in j-th class, the access way is preds[j * num_data + i].
Returns
-------
result : list
List with evaluation results.
"""
if not isinstance(data, Dataset):
raise TypeError("Can only eval for Dataset instance")
data_idx = -1
if data is self.train_set:
data_idx = 0
else:
for i in range_(len(self.valid_sets)):
if data is self.valid_sets[i]:
data_idx = i + 1
break
# need to push new valid data
if data_idx == -1:
self.add_valid(data, name)
data_idx = self.__num_dataset - 1
return self.__inner_eval(name, data_idx, feval) |
Evaluate for validation data.
Parameters
----------
feval : callable or None, optional (default=None)
Customized evaluation function.
Should accept two parameters: preds, train_data,
and return (eval_name, eval_result, is_higher_better) or list of such tuples.
For multi-class task, the preds is group by class_id first, then group by row_id.
If you want to get i-th row preds in j-th class, the access way is preds[j * num_data + i].
Returns
-------
result : list
List with evaluation results. | def eval_valid(self, feval=None):
"""Evaluate for validation data.
Parameters
----------
feval : callable or None, optional (default=None)
Customized evaluation function.
Should accept two parameters: preds, train_data,
and return (eval_name, eval_result, is_higher_better) or list of such tuples.
For multi-class task, the preds is group by class_id first, then group by row_id.
If you want to get i-th row preds in j-th class, the access way is preds[j * num_data + i].
Returns
-------
result : list
List with evaluation results.
"""
return [item for i in range_(1, self.__num_dataset)
for item in self.__inner_eval(self.name_valid_sets[i - 1], i, feval)] |
Save Booster to file.
Parameters
----------
filename : string
Filename to save Booster.
num_iteration : int or None, optional (default=None)
Index of the iteration that should be saved.
If None, if the best iteration exists, it is saved; otherwise, all iterations are saved.
If <= 0, all iterations are saved.
start_iteration : int, optional (default=0)
Start index of the iteration that should be saved.
Returns
-------
self : Booster
Returns self. | def save_model(self, filename, num_iteration=None, start_iteration=0):
"""Save Booster to file.
Parameters
----------
filename : string
Filename to save Booster.
num_iteration : int or None, optional (default=None)
Index of the iteration that should be saved.
If None, if the best iteration exists, it is saved; otherwise, all iterations are saved.
If <= 0, all iterations are saved.
start_iteration : int, optional (default=0)
Start index of the iteration that should be saved.
Returns
-------
self : Booster
Returns self.
"""
if num_iteration is None:
num_iteration = self.best_iteration
_safe_call(_LIB.LGBM_BoosterSaveModel(
self.handle,
ctypes.c_int(start_iteration),
ctypes.c_int(num_iteration),
c_str(filename)))
_dump_pandas_categorical(self.pandas_categorical, filename)
return self |
Shuffle models.
Parameters
----------
start_iteration : int, optional (default=0)
The first iteration that will be shuffled.
end_iteration : int, optional (default=-1)
The last iteration that will be shuffled.
If <= 0, means the last available iteration.
Returns
-------
self : Booster
Booster with shuffled models. | def shuffle_models(self, start_iteration=0, end_iteration=-1):
"""Shuffle models.
Parameters
----------
start_iteration : int, optional (default=0)
The first iteration that will be shuffled.
end_iteration : int, optional (default=-1)
The last iteration that will be shuffled.
If <= 0, means the last available iteration.
Returns
-------
self : Booster
Booster with shuffled models.
"""
_safe_call(_LIB.LGBM_BoosterShuffleModels(
self.handle,
ctypes.c_int(start_iteration),
ctypes.c_int(end_iteration)))
return self |
Load Booster from a string.
Parameters
----------
model_str : string
Model will be loaded from this string.
verbose : bool, optional (default=True)
Whether to print messages while loading model.
Returns
-------
self : Booster
Loaded Booster object. | def model_from_string(self, model_str, verbose=True):
"""Load Booster from a string.
Parameters
----------
model_str : string
Model will be loaded from this string.
verbose : bool, optional (default=True)
Whether to print messages while loading model.
Returns
-------
self : Booster
Loaded Booster object.
"""
if self.handle is not None:
_safe_call(_LIB.LGBM_BoosterFree(self.handle))
self._free_buffer()
self.handle = ctypes.c_void_p()
out_num_iterations = ctypes.c_int(0)
_safe_call(_LIB.LGBM_BoosterLoadModelFromString(
c_str(model_str),
ctypes.byref(out_num_iterations),
ctypes.byref(self.handle)))
out_num_class = ctypes.c_int(0)
_safe_call(_LIB.LGBM_BoosterGetNumClasses(
self.handle,
ctypes.byref(out_num_class)))
if verbose:
print('Finished loading model, total used %d iterations' % int(out_num_iterations.value))
self.__num_class = out_num_class.value
self.pandas_categorical = _load_pandas_categorical(model_str=model_str)
return self |
Save Booster to string.
Parameters
----------
num_iteration : int or None, optional (default=None)
Index of the iteration that should be saved.
If None, if the best iteration exists, it is saved; otherwise, all iterations are saved.
If <= 0, all iterations are saved.
start_iteration : int, optional (default=0)
Start index of the iteration that should be saved.
Returns
-------
str_repr : string
String representation of Booster. | def model_to_string(self, num_iteration=None, start_iteration=0):
"""Save Booster to string.
Parameters
----------
num_iteration : int or None, optional (default=None)
Index of the iteration that should be saved.
If None, if the best iteration exists, it is saved; otherwise, all iterations are saved.
If <= 0, all iterations are saved.
start_iteration : int, optional (default=0)
Start index of the iteration that should be saved.
Returns
-------
str_repr : string
String representation of Booster.
"""
if num_iteration is None:
num_iteration = self.best_iteration
buffer_len = 1 << 20
tmp_out_len = ctypes.c_int64(0)
string_buffer = ctypes.create_string_buffer(buffer_len)
ptr_string_buffer = ctypes.c_char_p(*[ctypes.addressof(string_buffer)])
_safe_call(_LIB.LGBM_BoosterSaveModelToString(
self.handle,
ctypes.c_int(start_iteration),
ctypes.c_int(num_iteration),
ctypes.c_int64(buffer_len),
ctypes.byref(tmp_out_len),
ptr_string_buffer))
actual_len = tmp_out_len.value
# if buffer length is not long enough, re-allocate a buffer
if actual_len > buffer_len:
string_buffer = ctypes.create_string_buffer(actual_len)
ptr_string_buffer = ctypes.c_char_p(*[ctypes.addressof(string_buffer)])
_safe_call(_LIB.LGBM_BoosterSaveModelToString(
self.handle,
ctypes.c_int(start_iteration),
ctypes.c_int(num_iteration),
ctypes.c_int64(actual_len),
ctypes.byref(tmp_out_len),
ptr_string_buffer))
ret = string_buffer.value.decode()
ret += _dump_pandas_categorical(self.pandas_categorical)
return ret |
Dump Booster to JSON format.
Parameters
----------
num_iteration : int or None, optional (default=None)
Index of the iteration that should be dumped.
If None, if the best iteration exists, it is dumped; otherwise, all iterations are dumped.
If <= 0, all iterations are dumped.
start_iteration : int, optional (default=0)
Start index of the iteration that should be dumped.
Returns
-------
json_repr : dict
JSON format of Booster. | def dump_model(self, num_iteration=None, start_iteration=0):
"""Dump Booster to JSON format.
Parameters
----------
num_iteration : int or None, optional (default=None)
Index of the iteration that should be dumped.
If None, if the best iteration exists, it is dumped; otherwise, all iterations are dumped.
If <= 0, all iterations are dumped.
start_iteration : int, optional (default=0)
Start index of the iteration that should be dumped.
Returns
-------
json_repr : dict
JSON format of Booster.
"""
if num_iteration is None:
num_iteration = self.best_iteration
buffer_len = 1 << 20
tmp_out_len = ctypes.c_int64(0)
string_buffer = ctypes.create_string_buffer(buffer_len)
ptr_string_buffer = ctypes.c_char_p(*[ctypes.addressof(string_buffer)])
_safe_call(_LIB.LGBM_BoosterDumpModel(
self.handle,
ctypes.c_int(start_iteration),
ctypes.c_int(num_iteration),
ctypes.c_int64(buffer_len),
ctypes.byref(tmp_out_len),
ptr_string_buffer))
actual_len = tmp_out_len.value
# if buffer length is not long enough, reallocate a buffer
if actual_len > buffer_len:
string_buffer = ctypes.create_string_buffer(actual_len)
ptr_string_buffer = ctypes.c_char_p(*[ctypes.addressof(string_buffer)])
_safe_call(_LIB.LGBM_BoosterDumpModel(
self.handle,
ctypes.c_int(start_iteration),
ctypes.c_int(num_iteration),
ctypes.c_int64(actual_len),
ctypes.byref(tmp_out_len),
ptr_string_buffer))
ret = json.loads(string_buffer.value.decode())
ret['pandas_categorical'] = json.loads(json.dumps(self.pandas_categorical,
default=json_default_with_numpy))
return ret |
Make a prediction.
Parameters
----------
data : string, numpy array, pandas DataFrame, H2O DataTable's Frame or scipy.sparse
Data source for prediction.
If string, it represents the path to txt file.
num_iteration : int or None, optional (default=None)
Limit number of iterations in the prediction.
If None, if the best iteration exists, it is used; otherwise, all iterations are used.
If <= 0, all iterations are used (no limits).
raw_score : bool, optional (default=False)
Whether to predict raw scores.
pred_leaf : bool, optional (default=False)
Whether to predict leaf index.
pred_contrib : bool, optional (default=False)
Whether to predict feature contributions.
Note
----
If you want to get more explanations for your model's predictions using SHAP values,
like SHAP interaction values,
you can install the shap package (https://github.com/slundberg/shap).
Note that unlike the shap package, with ``pred_contrib`` we return a matrix with an extra
column, where the last column is the expected value.
data_has_header : bool, optional (default=False)
Whether the data has header.
Used only if data is string.
is_reshape : bool, optional (default=True)
If True, result is reshaped to [nrow, ncol].
**kwargs
Other parameters for the prediction.
Returns
-------
result : numpy array
Prediction result. | def predict(self, data, num_iteration=None,
raw_score=False, pred_leaf=False, pred_contrib=False,
data_has_header=False, is_reshape=True, **kwargs):
"""Make a prediction.
Parameters
----------
data : string, numpy array, pandas DataFrame, H2O DataTable's Frame or scipy.sparse
Data source for prediction.
If string, it represents the path to txt file.
num_iteration : int or None, optional (default=None)
Limit number of iterations in the prediction.
If None, if the best iteration exists, it is used; otherwise, all iterations are used.
If <= 0, all iterations are used (no limits).
raw_score : bool, optional (default=False)
Whether to predict raw scores.
pred_leaf : bool, optional (default=False)
Whether to predict leaf index.
pred_contrib : bool, optional (default=False)
Whether to predict feature contributions.
Note
----
If you want to get more explanations for your model's predictions using SHAP values,
like SHAP interaction values,
you can install the shap package (https://github.com/slundberg/shap).
Note that unlike the shap package, with ``pred_contrib`` we return a matrix with an extra
column, where the last column is the expected value.
data_has_header : bool, optional (default=False)
Whether the data has header.
Used only if data is string.
is_reshape : bool, optional (default=True)
If True, result is reshaped to [nrow, ncol].
**kwargs
Other parameters for the prediction.
Returns
-------
result : numpy array
Prediction result.
"""
predictor = self._to_predictor(copy.deepcopy(kwargs))
if num_iteration is None:
num_iteration = self.best_iteration
return predictor.predict(data, num_iteration,
raw_score, pred_leaf, pred_contrib,
data_has_header, is_reshape) |
Refit the existing Booster by new data.
Parameters
----------
data : string, numpy array, pandas DataFrame, H2O DataTable's Frame or scipy.sparse
Data source for refit.
If string, it represents the path to txt file.
label : list, numpy 1-D array or pandas Series / one-column DataFrame
Label for refit.
decay_rate : float, optional (default=0.9)
Decay rate of refit,
will use ``leaf_output = decay_rate * old_leaf_output + (1.0 - decay_rate) * new_leaf_output`` to refit trees.
**kwargs
Other parameters for refit.
These parameters will be passed to ``predict`` method.
Returns
-------
result : Booster
Refitted Booster. | def refit(self, data, label, decay_rate=0.9, **kwargs):
"""Refit the existing Booster by new data.
Parameters
----------
data : string, numpy array, pandas DataFrame, H2O DataTable's Frame or scipy.sparse
Data source for refit.
If string, it represents the path to txt file.
label : list, numpy 1-D array or pandas Series / one-column DataFrame
Label for refit.
decay_rate : float, optional (default=0.9)
Decay rate of refit,
will use ``leaf_output = decay_rate * old_leaf_output + (1.0 - decay_rate) * new_leaf_output`` to refit trees.
**kwargs
Other parameters for refit.
These parameters will be passed to ``predict`` method.
Returns
-------
result : Booster
Refitted Booster.
"""
if self.__set_objective_to_none:
raise LightGBMError('Cannot refit due to null objective function.')
predictor = self._to_predictor(copy.deepcopy(kwargs))
leaf_preds = predictor.predict(data, -1, pred_leaf=True)
nrow, ncol = leaf_preds.shape
train_set = Dataset(data, label, silent=True)
new_booster = Booster(self.params, train_set, silent=True)
# Copy models
_safe_call(_LIB.LGBM_BoosterMerge(
new_booster.handle,
predictor.handle))
leaf_preds = leaf_preds.reshape(-1)
ptr_data, type_ptr_data, _ = c_int_array(leaf_preds)
_safe_call(_LIB.LGBM_BoosterRefit(
new_booster.handle,
ptr_data,
ctypes.c_int(nrow),
ctypes.c_int(ncol)))
new_booster.network = self.network
new_booster.__attr = self.__attr.copy()
return new_booster |
Get the output of a leaf.
Parameters
----------
tree_id : int
The index of the tree.
leaf_id : int
The index of the leaf in the tree.
Returns
-------
result : float
The output of the leaf. | def get_leaf_output(self, tree_id, leaf_id):
"""Get the output of a leaf.
Parameters
----------
tree_id : int
The index of the tree.
leaf_id : int
The index of the leaf in the tree.
Returns
-------
result : float
The output of the leaf.
"""
ret = ctypes.c_double(0)
_safe_call(_LIB.LGBM_BoosterGetLeafValue(
self.handle,
ctypes.c_int(tree_id),
ctypes.c_int(leaf_id),
ctypes.byref(ret)))
return ret.value |
Convert to predictor. | def _to_predictor(self, pred_parameter=None):
"""Convert to predictor."""
predictor = _InnerPredictor(booster_handle=self.handle, pred_parameter=pred_parameter)
predictor.pandas_categorical = self.pandas_categorical
return predictor |
Get number of features.
Returns
-------
num_feature : int
The number of features. | def num_feature(self):
"""Get number of features.
Returns
-------
num_feature : int
The number of features.
"""
out_num_feature = ctypes.c_int(0)
_safe_call(_LIB.LGBM_BoosterGetNumFeature(
self.handle,
ctypes.byref(out_num_feature)))
return out_num_feature.value |
Get names of features.
Returns
-------
result : list
List with names of features. | def feature_name(self):
"""Get names of features.
Returns
-------
result : list
List with names of features.
"""
num_feature = self.num_feature()
# Get name of features
tmp_out_len = ctypes.c_int(0)
string_buffers = [ctypes.create_string_buffer(255) for i in range_(num_feature)]
ptr_string_buffers = (ctypes.c_char_p * num_feature)(*map(ctypes.addressof, string_buffers))
_safe_call(_LIB.LGBM_BoosterGetFeatureNames(
self.handle,
ctypes.byref(tmp_out_len),
ptr_string_buffers))
if num_feature != tmp_out_len.value:
raise ValueError("Length of feature names doesn't equal with num_feature")
return [string_buffers[i].value.decode() for i in range_(num_feature)] |
Get feature importances.
Parameters
----------
importance_type : string, optional (default="split")
How the importance is calculated.
If "split", result contains numbers of times the feature is used in a model.
If "gain", result contains total gains of splits which use the feature.
iteration : int or None, optional (default=None)
Limit number of iterations in the feature importance calculation.
If None, if the best iteration exists, it is used; otherwise, all trees are used.
If <= 0, all trees are used (no limits).
Returns
-------
result : numpy array
Array with feature importances. | def feature_importance(self, importance_type='split', iteration=None):
"""Get feature importances.
Parameters
----------
importance_type : string, optional (default="split")
How the importance is calculated.
If "split", result contains numbers of times the feature is used in a model.
If "gain", result contains total gains of splits which use the feature.
iteration : int or None, optional (default=None)
Limit number of iterations in the feature importance calculation.
If None, if the best iteration exists, it is used; otherwise, all trees are used.
If <= 0, all trees are used (no limits).
Returns
-------
result : numpy array
Array with feature importances.
"""
if iteration is None:
iteration = self.best_iteration
if importance_type == "split":
importance_type_int = 0
elif importance_type == "gain":
importance_type_int = 1
else:
importance_type_int = -1
result = np.zeros(self.num_feature(), dtype=np.float64)
_safe_call(_LIB.LGBM_BoosterFeatureImportance(
self.handle,
ctypes.c_int(iteration),
ctypes.c_int(importance_type_int),
result.ctypes.data_as(ctypes.POINTER(ctypes.c_double))))
if importance_type_int == 0:
return result.astype(int)
else:
return result |
Get split value histogram for the specified feature.
Parameters
----------
feature : int or string
The feature name or index the histogram is calculated for.
If int, interpreted as index.
If string, interpreted as name.
Note
----
Categorical features are not supported.
bins : int, string or None, optional (default=None)
The maximum number of bins.
If None, or int and > number of unique split values and ``xgboost_style=True``,
the number of bins equals number of unique split values.
If string, it should be one from the list of the supported values by ``numpy.histogram()`` function.
xgboost_style : bool, optional (default=False)
Whether the returned result should be in the same form as it is in XGBoost.
If False, the returned value is tuple of 2 numpy arrays as it is in ``numpy.histogram()`` function.
If True, the returned value is matrix, in which the first column is the right edges of non-empty bins
and the second one is the histogram values.
Returns
-------
result_tuple : tuple of 2 numpy arrays
If ``xgboost_style=False``, the values of the histogram of used splitting values for the specified feature
and the bin edges.
result_array_like : numpy array or pandas DataFrame (if pandas is installed)
If ``xgboost_style=True``, the histogram of used splitting values for the specified feature. | def get_split_value_histogram(self, feature, bins=None, xgboost_style=False):
"""Get split value histogram for the specified feature.
Parameters
----------
feature : int or string
The feature name or index the histogram is calculated for.
If int, interpreted as index.
If string, interpreted as name.
Note
----
Categorical features are not supported.
bins : int, string or None, optional (default=None)
The maximum number of bins.
If None, or int and > number of unique split values and ``xgboost_style=True``,
the number of bins equals number of unique split values.
If string, it should be one from the list of the supported values by ``numpy.histogram()`` function.
xgboost_style : bool, optional (default=False)
Whether the returned result should be in the same form as it is in XGBoost.
If False, the returned value is tuple of 2 numpy arrays as it is in ``numpy.histogram()`` function.
If True, the returned value is matrix, in which the first column is the right edges of non-empty bins
and the second one is the histogram values.
Returns
-------
result_tuple : tuple of 2 numpy arrays
If ``xgboost_style=False``, the values of the histogram of used splitting values for the specified feature
and the bin edges.
result_array_like : numpy array or pandas DataFrame (if pandas is installed)
If ``xgboost_style=True``, the histogram of used splitting values for the specified feature.
"""
def add(root):
"""Recursively add thresholds."""
if 'split_index' in root: # non-leaf
if feature_names is not None and isinstance(feature, string_type):
split_feature = feature_names[root['split_feature']]
else:
split_feature = root['split_feature']
if split_feature == feature:
if isinstance(root['threshold'], string_type):
raise LightGBMError('Cannot compute split value histogram for the categorical feature')
else:
values.append(root['threshold'])
add(root['left_child'])
add(root['right_child'])
model = self.dump_model()
feature_names = model.get('feature_names')
tree_infos = model['tree_info']
values = []
for tree_info in tree_infos:
add(tree_info['tree_structure'])
if bins is None or isinstance(bins, integer_types) and xgboost_style:
n_unique = len(np.unique(values))
bins = max(min(n_unique, bins) if bins is not None else n_unique, 1)
hist, bin_edges = np.histogram(values, bins=bins)
if xgboost_style:
ret = np.column_stack((bin_edges[1:], hist))
ret = ret[ret[:, 1] > 0]
if PANDAS_INSTALLED:
return DataFrame(ret, columns=['SplitValue', 'Count'])
else:
return ret
else:
return hist, bin_edges |
Evaluate training or validation data. | def __inner_eval(self, data_name, data_idx, feval=None):
"""Evaluate training or validation data."""
if data_idx >= self.__num_dataset:
raise ValueError("Data_idx should be smaller than number of dataset")
self.__get_eval_info()
ret = []
if self.__num_inner_eval > 0:
result = np.zeros(self.__num_inner_eval, dtype=np.float64)
tmp_out_len = ctypes.c_int(0)
_safe_call(_LIB.LGBM_BoosterGetEval(
self.handle,
ctypes.c_int(data_idx),
ctypes.byref(tmp_out_len),
result.ctypes.data_as(ctypes.POINTER(ctypes.c_double))))
if tmp_out_len.value != self.__num_inner_eval:
raise ValueError("Wrong length of eval results")
for i in range_(self.__num_inner_eval):
ret.append((data_name, self.__name_inner_eval[i],
result[i], self.__higher_better_inner_eval[i]))
if feval is not None:
if data_idx == 0:
cur_data = self.train_set
else:
cur_data = self.valid_sets[data_idx - 1]
feval_ret = feval(self.__inner_predict(data_idx), cur_data)
if isinstance(feval_ret, list):
for eval_name, val, is_higher_better in feval_ret:
ret.append((data_name, eval_name, val, is_higher_better))
else:
eval_name, val, is_higher_better = feval_ret
ret.append((data_name, eval_name, val, is_higher_better))
return ret |
Predict for training and validation dataset. | def __inner_predict(self, data_idx):
"""Predict for training and validation dataset."""
if data_idx >= self.__num_dataset:
raise ValueError("Data_idx should be smaller than number of dataset")
if self.__inner_predict_buffer[data_idx] is None:
if data_idx == 0:
n_preds = self.train_set.num_data() * self.__num_class
else:
n_preds = self.valid_sets[data_idx - 1].num_data() * self.__num_class
self.__inner_predict_buffer[data_idx] = np.zeros(n_preds, dtype=np.float64)
# avoid to predict many time in one iteration
if not self.__is_predicted_cur_iter[data_idx]:
tmp_out_len = ctypes.c_int64(0)
data_ptr = self.__inner_predict_buffer[data_idx].ctypes.data_as(ctypes.POINTER(ctypes.c_double))
_safe_call(_LIB.LGBM_BoosterGetPredict(
self.handle,
ctypes.c_int(data_idx),
ctypes.byref(tmp_out_len),
data_ptr))
if tmp_out_len.value != len(self.__inner_predict_buffer[data_idx]):
raise ValueError("Wrong length of predict results for data %d" % (data_idx))
self.__is_predicted_cur_iter[data_idx] = True
return self.__inner_predict_buffer[data_idx] |
Get inner evaluation count and names. | def __get_eval_info(self):
"""Get inner evaluation count and names."""
if self.__need_reload_eval_info:
self.__need_reload_eval_info = False
out_num_eval = ctypes.c_int(0)
# Get num of inner evals
_safe_call(_LIB.LGBM_BoosterGetEvalCounts(
self.handle,
ctypes.byref(out_num_eval)))
self.__num_inner_eval = out_num_eval.value
if self.__num_inner_eval > 0:
# Get name of evals
tmp_out_len = ctypes.c_int(0)
string_buffers = [ctypes.create_string_buffer(255) for i in range_(self.__num_inner_eval)]
ptr_string_buffers = (ctypes.c_char_p * self.__num_inner_eval)(*map(ctypes.addressof, string_buffers))
_safe_call(_LIB.LGBM_BoosterGetEvalNames(
self.handle,
ctypes.byref(tmp_out_len),
ptr_string_buffers))
if self.__num_inner_eval != tmp_out_len.value:
raise ValueError("Length of eval names doesn't equal with num_evals")
self.__name_inner_eval = \
[string_buffers[i].value.decode() for i in range_(self.__num_inner_eval)]
self.__higher_better_inner_eval = \
[name.startswith(('auc', 'ndcg@', 'map@')) for name in self.__name_inner_eval] |
Find the path to LightGBM library files.
Returns
-------
lib_path: list of strings
List of all found library paths to LightGBM. | def find_lib_path():
"""Find the path to LightGBM library files.
Returns
-------
lib_path: list of strings
List of all found library paths to LightGBM.
"""
if os.environ.get('LIGHTGBM_BUILD_DOC', False):
# we don't need lib_lightgbm while building docs
return []
curr_path = os.path.dirname(os.path.abspath(os.path.expanduser(__file__)))
dll_path = [curr_path,
os.path.join(curr_path, '../../'),
os.path.join(curr_path, 'compile'),
os.path.join(curr_path, '../compile'),
os.path.join(curr_path, '../../lib/')]
if system() in ('Windows', 'Microsoft'):
dll_path.append(os.path.join(curr_path, '../compile/Release/'))
dll_path.append(os.path.join(curr_path, '../compile/windows/x64/DLL/'))
dll_path.append(os.path.join(curr_path, '../../Release/'))
dll_path.append(os.path.join(curr_path, '../../windows/x64/DLL/'))
dll_path = [os.path.join(p, 'lib_lightgbm.dll') for p in dll_path]
else:
dll_path = [os.path.join(p, 'lib_lightgbm.so') for p in dll_path]
lib_path = [p for p in dll_path if os.path.exists(p) and os.path.isfile(p)]
if not lib_path:
dll_path = [os.path.realpath(p) for p in dll_path]
raise Exception('Cannot find lightgbm library file in following paths:\n' + '\n'.join(dll_path))
return lib_path |
Set attributes to the Booster.
Parameters
----------
**kwargs
The attributes to set.
Setting a value to None deletes an attribute.
Returns
-------
self : Booster
Booster with set attributes. | def set_attr(self, **kwargs):
"""Set attributes to the Booster.
Parameters
----------
**kwargs
The attributes to set.
Setting a value to None deletes an attribute.
Returns
-------
self : Booster
Booster with set attributes.
"""
for key, value in kwargs.items():
if value is not None:
if not isinstance(value, string_type):
raise ValueError("Only string values are accepted")
self.__attr[key] = value
else:
self.__attr.pop(key, None)
return self |
Convert numpy classes to JSON serializable objects. | def json_default_with_numpy(obj):
"""Convert numpy classes to JSON serializable objects."""
if isinstance(obj, (np.integer, np.floating, np.bool_)):
return obj.item()
elif isinstance(obj, np.ndarray):
return obj.tolist()
else:
return obj |
Create a callback that prints the evaluation results.
Parameters
----------
period : int, optional (default=1)
The period to print the evaluation results.
show_stdv : bool, optional (default=True)
Whether to show stdv (if provided).
Returns
-------
callback : function
The callback that prints the evaluation results every ``period`` iteration(s). | def print_evaluation(period=1, show_stdv=True):
"""Create a callback that prints the evaluation results.
Parameters
----------
period : int, optional (default=1)
The period to print the evaluation results.
show_stdv : bool, optional (default=True)
Whether to show stdv (if provided).
Returns
-------
callback : function
The callback that prints the evaluation results every ``period`` iteration(s).
"""
def _callback(env):
if period > 0 and env.evaluation_result_list and (env.iteration + 1) % period == 0:
result = '\t'.join([_format_eval_result(x, show_stdv) for x in env.evaluation_result_list])
print('[%d]\t%s' % (env.iteration + 1, result))
_callback.order = 10
return _callback |
Format metric string. | def _format_eval_result(value, show_stdv=True):
"""Format metric string."""
if len(value) == 4:
return '%s\'s %s: %g' % (value[0], value[1], value[2])
elif len(value) == 5:
if show_stdv:
return '%s\'s %s: %g + %g' % (value[0], value[1], value[2], value[4])
else:
return '%s\'s %s: %g' % (value[0], value[1], value[2])
else:
raise ValueError("Wrong metric value") |
Create a callback that records the evaluation history into ``eval_result``.
Parameters
----------
eval_result : dict
A dictionary to store the evaluation results.
Returns
-------
callback : function
The callback that records the evaluation history into the passed dictionary. | def record_evaluation(eval_result):
"""Create a callback that records the evaluation history into ``eval_result``.
Parameters
----------
eval_result : dict
A dictionary to store the evaluation results.
Returns
-------
callback : function
The callback that records the evaluation history into the passed dictionary.
"""
if not isinstance(eval_result, dict):
raise TypeError('Eval_result should be a dictionary')
eval_result.clear()
def _init(env):
for data_name, _, _, _ in env.evaluation_result_list:
eval_result.setdefault(data_name, collections.defaultdict(list))
def _callback(env):
if not eval_result:
_init(env)
for data_name, eval_name, result, _ in env.evaluation_result_list:
eval_result[data_name][eval_name].append(result)
_callback.order = 20
return _callback |
Create a callback that resets the parameter after the first iteration.
Note
----
The initial parameter will still take in-effect on first iteration.
Parameters
----------
**kwargs : value should be list or function
List of parameters for each boosting round
or a customized function that calculates the parameter in terms of
current number of round (e.g. yields learning rate decay).
If list lst, parameter = lst[current_round].
If function func, parameter = func(current_round).
Returns
-------
callback : function
The callback that resets the parameter after the first iteration. | def reset_parameter(**kwargs):
"""Create a callback that resets the parameter after the first iteration.
Note
----
The initial parameter will still take in-effect on first iteration.
Parameters
----------
**kwargs : value should be list or function
List of parameters for each boosting round
or a customized function that calculates the parameter in terms of
current number of round (e.g. yields learning rate decay).
If list lst, parameter = lst[current_round].
If function func, parameter = func(current_round).
Returns
-------
callback : function
The callback that resets the parameter after the first iteration.
"""
def _callback(env):
new_parameters = {}
for key, value in kwargs.items():
if key in ['num_class', 'num_classes',
'boosting', 'boost', 'boosting_type',
'metric', 'metrics', 'metric_types']:
raise RuntimeError("cannot reset {} during training".format(repr(key)))
if isinstance(value, list):
if len(value) != env.end_iteration - env.begin_iteration:
raise ValueError("Length of list {} has to equal to 'num_boost_round'."
.format(repr(key)))
new_param = value[env.iteration - env.begin_iteration]
else:
new_param = value(env.iteration - env.begin_iteration)
if new_param != env.params.get(key, None):
new_parameters[key] = new_param
if new_parameters:
env.model.reset_parameter(new_parameters)
env.params.update(new_parameters)
_callback.before_iteration = True
_callback.order = 10
return _callback |
Create a callback that activates early stopping.
Note
----
Activates early stopping.
The model will train until the validation score stops improving.
Validation score needs to improve at least every ``early_stopping_rounds`` round(s)
to continue training.
Requires at least one validation data and one metric.
If there's more than one, will check all of them. But the training data is ignored anyway.
To check only the first metric set ``first_metric_only`` to True.
Parameters
----------
stopping_rounds : int
The possible number of rounds without the trend occurrence.
first_metric_only : bool, optional (default=False)
Whether to use only the first metric for early stopping.
verbose : bool, optional (default=True)
Whether to print message with early stopping information.
Returns
-------
callback : function
The callback that activates early stopping. | def early_stopping(stopping_rounds, first_metric_only=False, verbose=True):
"""Create a callback that activates early stopping.
Note
----
Activates early stopping.
The model will train until the validation score stops improving.
Validation score needs to improve at least every ``early_stopping_rounds`` round(s)
to continue training.
Requires at least one validation data and one metric.
If there's more than one, will check all of them. But the training data is ignored anyway.
To check only the first metric set ``first_metric_only`` to True.
Parameters
----------
stopping_rounds : int
The possible number of rounds without the trend occurrence.
first_metric_only : bool, optional (default=False)
Whether to use only the first metric for early stopping.
verbose : bool, optional (default=True)
Whether to print message with early stopping information.
Returns
-------
callback : function
The callback that activates early stopping.
"""
best_score = []
best_iter = []
best_score_list = []
cmp_op = []
enabled = [True]
def _init(env):
enabled[0] = not any((boost_alias in env.params
and env.params[boost_alias] == 'dart') for boost_alias in ('boosting',
'boosting_type',
'boost'))
if not enabled[0]:
warnings.warn('Early stopping is not available in dart mode')
return
if not env.evaluation_result_list:
raise ValueError('For early stopping, '
'at least one dataset and eval metric is required for evaluation')
if verbose:
msg = "Training until validation scores don't improve for {} rounds."
print(msg.format(stopping_rounds))
for eval_ret in env.evaluation_result_list:
best_iter.append(0)
best_score_list.append(None)
if eval_ret[3]:
best_score.append(float('-inf'))
cmp_op.append(gt)
else:
best_score.append(float('inf'))
cmp_op.append(lt)
def _callback(env):
if not cmp_op:
_init(env)
if not enabled[0]:
return
for i in range_(len(env.evaluation_result_list)):
score = env.evaluation_result_list[i][2]
if best_score_list[i] is None or cmp_op[i](score, best_score[i]):
best_score[i] = score
best_iter[i] = env.iteration
best_score_list[i] = env.evaluation_result_list
elif env.iteration - best_iter[i] >= stopping_rounds:
if verbose:
print('Early stopping, best iteration is:\n[%d]\t%s' % (
best_iter[i] + 1, '\t'.join([_format_eval_result(x) for x in best_score_list[i]])))
raise EarlyStopException(best_iter[i], best_score_list[i])
if env.iteration == env.end_iteration - 1:
if verbose:
print('Did not meet early stopping. Best iteration is:\n[%d]\t%s' % (
best_iter[i] + 1, '\t'.join([_format_eval_result(x) for x in best_score_list[i]])))
raise EarlyStopException(best_iter[i], best_score_list[i])
if first_metric_only: # the only first metric is used for early stopping
break
_callback.order = 30
return _callback |
Perform the training with given parameters.
Parameters
----------
params : dict
Parameters for training.
train_set : Dataset
Data to be trained on.
num_boost_round : int, optional (default=100)
Number of boosting iterations.
valid_sets : list of Datasets or None, optional (default=None)
List of data to be evaluated on during training.
valid_names : list of strings or None, optional (default=None)
Names of ``valid_sets``.
fobj : callable or None, optional (default=None)
Customized objective function.
feval : callable or None, optional (default=None)
Customized evaluation function.
Should accept two parameters: preds, train_data,
and return (eval_name, eval_result, is_higher_better) or list of such tuples.
For multi-class task, the preds is group by class_id first, then group by row_id.
If you want to get i-th row preds in j-th class, the access way is preds[j * num_data + i].
To ignore the default metric corresponding to the used objective,
set the ``metric`` parameter to the string ``"None"`` in ``params``.
init_model : string, Booster or None, optional (default=None)
Filename of LightGBM model or Booster instance used for continue training.
feature_name : list of strings or 'auto', optional (default="auto")
Feature names.
If 'auto' and data is pandas DataFrame, data columns names are used.
categorical_feature : list of strings or int, or 'auto', optional (default="auto")
Categorical features.
If list of int, interpreted as indices.
If list of strings, interpreted as feature names (need to specify ``feature_name`` as well).
If 'auto' and data is pandas DataFrame, pandas unordered categorical columns are used.
All values in categorical features should be less than int32 max value (2147483647).
Large values could be memory consuming. Consider using consecutive integers starting from zero.
All negative values in categorical features will be treated as missing values.
early_stopping_rounds : int or None, optional (default=None)
Activates early stopping. The model will train until the validation score stops improving.
Validation score needs to improve at least every ``early_stopping_rounds`` round(s)
to continue training.
Requires at least one validation data and one metric.
If there's more than one, will check all of them. But the training data is ignored anyway.
To check only the first metric you can pass in ``callbacks``
``early_stopping`` callback with ``first_metric_only=True``.
The index of iteration that has the best performance will be saved in the ``best_iteration`` field
if early stopping logic is enabled by setting ``early_stopping_rounds``.
evals_result: dict or None, optional (default=None)
This dictionary used to store all evaluation results of all the items in ``valid_sets``.
Example
-------
With a ``valid_sets`` = [valid_set, train_set],
``valid_names`` = ['eval', 'train']
and a ``params`` = {'metric': 'logloss'}
returns {'train': {'logloss': ['0.48253', '0.35953', ...]},
'eval': {'logloss': ['0.480385', '0.357756', ...]}}.
verbose_eval : bool or int, optional (default=True)
Requires at least one validation data.
If True, the eval metric on the valid set is printed at each boosting stage.
If int, the eval metric on the valid set is printed at every ``verbose_eval`` boosting stage.
The last boosting stage or the boosting stage found by using ``early_stopping_rounds`` is also printed.
Example
-------
With ``verbose_eval`` = 4 and at least one item in ``valid_sets``,
an evaluation metric is printed every 4 (instead of 1) boosting stages.
learning_rates : list, callable or None, optional (default=None)
List of learning rates for each boosting round
or a customized function that calculates ``learning_rate``
in terms of current number of round (e.g. yields learning rate decay).
keep_training_booster : bool, optional (default=False)
Whether the returned Booster will be used to keep training.
If False, the returned value will be converted into _InnerPredictor before returning.
You can still use _InnerPredictor as ``init_model`` for future continue training.
callbacks : list of callables or None, optional (default=None)
List of callback functions that are applied at each iteration.
See Callbacks in Python API for more information.
Returns
-------
booster : Booster
The trained Booster model. | def train(params, train_set, num_boost_round=100,
valid_sets=None, valid_names=None,
fobj=None, feval=None, init_model=None,
feature_name='auto', categorical_feature='auto',
early_stopping_rounds=None, evals_result=None,
verbose_eval=True, learning_rates=None,
keep_training_booster=False, callbacks=None):
"""Perform the training with given parameters.
Parameters
----------
params : dict
Parameters for training.
train_set : Dataset
Data to be trained on.
num_boost_round : int, optional (default=100)
Number of boosting iterations.
valid_sets : list of Datasets or None, optional (default=None)
List of data to be evaluated on during training.
valid_names : list of strings or None, optional (default=None)
Names of ``valid_sets``.
fobj : callable or None, optional (default=None)
Customized objective function.
feval : callable or None, optional (default=None)
Customized evaluation function.
Should accept two parameters: preds, train_data,
and return (eval_name, eval_result, is_higher_better) or list of such tuples.
For multi-class task, the preds is group by class_id first, then group by row_id.
If you want to get i-th row preds in j-th class, the access way is preds[j * num_data + i].
To ignore the default metric corresponding to the used objective,
set the ``metric`` parameter to the string ``"None"`` in ``params``.
init_model : string, Booster or None, optional (default=None)
Filename of LightGBM model or Booster instance used for continue training.
feature_name : list of strings or 'auto', optional (default="auto")
Feature names.
If 'auto' and data is pandas DataFrame, data columns names are used.
categorical_feature : list of strings or int, or 'auto', optional (default="auto")
Categorical features.
If list of int, interpreted as indices.
If list of strings, interpreted as feature names (need to specify ``feature_name`` as well).
If 'auto' and data is pandas DataFrame, pandas unordered categorical columns are used.
All values in categorical features should be less than int32 max value (2147483647).
Large values could be memory consuming. Consider using consecutive integers starting from zero.
All negative values in categorical features will be treated as missing values.
early_stopping_rounds : int or None, optional (default=None)
Activates early stopping. The model will train until the validation score stops improving.
Validation score needs to improve at least every ``early_stopping_rounds`` round(s)
to continue training.
Requires at least one validation data and one metric.
If there's more than one, will check all of them. But the training data is ignored anyway.
To check only the first metric you can pass in ``callbacks``
``early_stopping`` callback with ``first_metric_only=True``.
The index of iteration that has the best performance will be saved in the ``best_iteration`` field
if early stopping logic is enabled by setting ``early_stopping_rounds``.
evals_result: dict or None, optional (default=None)
This dictionary used to store all evaluation results of all the items in ``valid_sets``.
Example
-------
With a ``valid_sets`` = [valid_set, train_set],
``valid_names`` = ['eval', 'train']
and a ``params`` = {'metric': 'logloss'}
returns {'train': {'logloss': ['0.48253', '0.35953', ...]},
'eval': {'logloss': ['0.480385', '0.357756', ...]}}.
verbose_eval : bool or int, optional (default=True)
Requires at least one validation data.
If True, the eval metric on the valid set is printed at each boosting stage.
If int, the eval metric on the valid set is printed at every ``verbose_eval`` boosting stage.
The last boosting stage or the boosting stage found by using ``early_stopping_rounds`` is also printed.
Example
-------
With ``verbose_eval`` = 4 and at least one item in ``valid_sets``,
an evaluation metric is printed every 4 (instead of 1) boosting stages.
learning_rates : list, callable or None, optional (default=None)
List of learning rates for each boosting round
or a customized function that calculates ``learning_rate``
in terms of current number of round (e.g. yields learning rate decay).
keep_training_booster : bool, optional (default=False)
Whether the returned Booster will be used to keep training.
If False, the returned value will be converted into _InnerPredictor before returning.
You can still use _InnerPredictor as ``init_model`` for future continue training.
callbacks : list of callables or None, optional (default=None)
List of callback functions that are applied at each iteration.
See Callbacks in Python API for more information.
Returns
-------
booster : Booster
The trained Booster model.
"""
# create predictor first
params = copy.deepcopy(params)
if fobj is not None:
params['objective'] = 'none'
for alias in ["num_iterations", "num_iteration", "n_iter", "num_tree", "num_trees",
"num_round", "num_rounds", "num_boost_round", "n_estimators"]:
if alias in params:
num_boost_round = int(params.pop(alias))
warnings.warn("Found `{}` in params. Will use it instead of argument".format(alias))
break
for alias in ["early_stopping_round", "early_stopping_rounds", "early_stopping"]:
if alias in params and params[alias] is not None:
early_stopping_rounds = int(params.pop(alias))
warnings.warn("Found `{}` in params. Will use it instead of argument".format(alias))
break
if num_boost_round <= 0:
raise ValueError("num_boost_round should be greater than zero.")
if isinstance(init_model, string_type):
predictor = _InnerPredictor(model_file=init_model, pred_parameter=params)
elif isinstance(init_model, Booster):
predictor = init_model._to_predictor(dict(init_model.params, **params))
else:
predictor = None
init_iteration = predictor.num_total_iteration if predictor is not None else 0
# check dataset
if not isinstance(train_set, Dataset):
raise TypeError("Training only accepts Dataset object")
train_set._update_params(params) \
._set_predictor(predictor) \
.set_feature_name(feature_name) \
.set_categorical_feature(categorical_feature)
is_valid_contain_train = False
train_data_name = "training"
reduced_valid_sets = []
name_valid_sets = []
if valid_sets is not None:
if isinstance(valid_sets, Dataset):
valid_sets = [valid_sets]
if isinstance(valid_names, string_type):
valid_names = [valid_names]
for i, valid_data in enumerate(valid_sets):
# reduce cost for prediction training data
if valid_data is train_set:
is_valid_contain_train = True
if valid_names is not None:
train_data_name = valid_names[i]
continue
if not isinstance(valid_data, Dataset):
raise TypeError("Traninig only accepts Dataset object")
reduced_valid_sets.append(valid_data._update_params(params).set_reference(train_set))
if valid_names is not None and len(valid_names) > i:
name_valid_sets.append(valid_names[i])
else:
name_valid_sets.append('valid_' + str(i))
# process callbacks
if callbacks is None:
callbacks = set()
else:
for i, cb in enumerate(callbacks):
cb.__dict__.setdefault('order', i - len(callbacks))
callbacks = set(callbacks)
# Most of legacy advanced options becomes callbacks
if verbose_eval is True:
callbacks.add(callback.print_evaluation())
elif isinstance(verbose_eval, integer_types):
callbacks.add(callback.print_evaluation(verbose_eval))
if early_stopping_rounds is not None:
callbacks.add(callback.early_stopping(early_stopping_rounds, verbose=bool(verbose_eval)))
if learning_rates is not None:
callbacks.add(callback.reset_parameter(learning_rate=learning_rates))
if evals_result is not None:
callbacks.add(callback.record_evaluation(evals_result))
callbacks_before_iter = {cb for cb in callbacks if getattr(cb, 'before_iteration', False)}
callbacks_after_iter = callbacks - callbacks_before_iter
callbacks_before_iter = sorted(callbacks_before_iter, key=attrgetter('order'))
callbacks_after_iter = sorted(callbacks_after_iter, key=attrgetter('order'))
# construct booster
try:
booster = Booster(params=params, train_set=train_set)
if is_valid_contain_train:
booster.set_train_data_name(train_data_name)
for valid_set, name_valid_set in zip_(reduced_valid_sets, name_valid_sets):
booster.add_valid(valid_set, name_valid_set)
finally:
train_set._reverse_update_params()
for valid_set in reduced_valid_sets:
valid_set._reverse_update_params()
booster.best_iteration = 0
# start training
for i in range_(init_iteration, init_iteration + num_boost_round):
for cb in callbacks_before_iter:
cb(callback.CallbackEnv(model=booster,
params=params,
iteration=i,
begin_iteration=init_iteration,
end_iteration=init_iteration + num_boost_round,
evaluation_result_list=None))
booster.update(fobj=fobj)
evaluation_result_list = []
# check evaluation result.
if valid_sets is not None:
if is_valid_contain_train:
evaluation_result_list.extend(booster.eval_train(feval))
evaluation_result_list.extend(booster.eval_valid(feval))
try:
for cb in callbacks_after_iter:
cb(callback.CallbackEnv(model=booster,
params=params,
iteration=i,
begin_iteration=init_iteration,
end_iteration=init_iteration + num_boost_round,
evaluation_result_list=evaluation_result_list))
except callback.EarlyStopException as earlyStopException:
booster.best_iteration = earlyStopException.best_iteration + 1
evaluation_result_list = earlyStopException.best_score
break
booster.best_score = collections.defaultdict(dict)
for dataset_name, eval_name, score, _ in evaluation_result_list:
booster.best_score[dataset_name][eval_name] = score
if not keep_training_booster:
booster.model_from_string(booster.model_to_string(), False).free_dataset()
return booster |
Make a n-fold list of Booster from random indices. | def _make_n_folds(full_data, folds, nfold, params, seed, fpreproc=None, stratified=True,
shuffle=True, eval_train_metric=False):
"""Make a n-fold list of Booster from random indices."""
full_data = full_data.construct()
num_data = full_data.num_data()
if folds is not None:
if not hasattr(folds, '__iter__') and not hasattr(folds, 'split'):
raise AttributeError("folds should be a generator or iterator of (train_idx, test_idx) tuples "
"or scikit-learn splitter object with split method")
if hasattr(folds, 'split'):
group_info = full_data.get_group()
if group_info is not None:
group_info = group_info.astype(int)
flatted_group = np.repeat(range_(len(group_info)), repeats=group_info)
else:
flatted_group = np.zeros(num_data, dtype=int)
folds = folds.split(X=np.zeros(num_data), y=full_data.get_label(), groups=flatted_group)
else:
if 'objective' in params and params['objective'] == 'lambdarank':
if not SKLEARN_INSTALLED:
raise LightGBMError('Scikit-learn is required for lambdarank cv.')
# lambdarank task, split according to groups
group_info = full_data.get_group().astype(int)
flatted_group = np.repeat(range_(len(group_info)), repeats=group_info)
group_kfold = _LGBMGroupKFold(n_splits=nfold)
folds = group_kfold.split(X=np.zeros(num_data), groups=flatted_group)
elif stratified:
if not SKLEARN_INSTALLED:
raise LightGBMError('Scikit-learn is required for stratified cv.')
skf = _LGBMStratifiedKFold(n_splits=nfold, shuffle=shuffle, random_state=seed)
folds = skf.split(X=np.zeros(num_data), y=full_data.get_label())
else:
if shuffle:
randidx = np.random.RandomState(seed).permutation(num_data)
else:
randidx = np.arange(num_data)
kstep = int(num_data / nfold)
test_id = [randidx[i: i + kstep] for i in range_(0, num_data, kstep)]
train_id = [np.concatenate([test_id[i] for i in range_(nfold) if k != i]) for k in range_(nfold)]
folds = zip_(train_id, test_id)
ret = _CVBooster()
for train_idx, test_idx in folds:
train_set = full_data.subset(train_idx)
valid_set = full_data.subset(test_idx)
# run preprocessing on the data set if needed
if fpreproc is not None:
train_set, valid_set, tparam = fpreproc(train_set, valid_set, params.copy())
else:
tparam = params
cvbooster = Booster(tparam, train_set)
if eval_train_metric:
cvbooster.add_valid(train_set, 'train')
cvbooster.add_valid(valid_set, 'valid')
ret.append(cvbooster)
return ret |
Aggregate cross-validation results. | def _agg_cv_result(raw_results, eval_train_metric=False):
"""Aggregate cross-validation results."""
cvmap = collections.defaultdict(list)
metric_type = {}
for one_result in raw_results:
for one_line in one_result:
if eval_train_metric:
key = "{} {}".format(one_line[0], one_line[1])
else:
key = one_line[1]
metric_type[key] = one_line[3]
cvmap[key].append(one_line[2])
return [('cv_agg', k, np.mean(v), metric_type[k], np.std(v)) for k, v in cvmap.items()] |
Perform the cross-validation with given paramaters.
Parameters
----------
params : dict
Parameters for Booster.
train_set : Dataset
Data to be trained on.
num_boost_round : int, optional (default=100)
Number of boosting iterations.
folds : generator or iterator of (train_idx, test_idx) tuples, scikit-learn splitter object or None, optional (default=None)
If generator or iterator, it should yield the train and test indices for each fold.
If object, it should be one of the scikit-learn splitter classes
(https://scikit-learn.org/stable/modules/classes.html#splitter-classes)
and have ``split`` method.
This argument has highest priority over other data split arguments.
nfold : int, optional (default=5)
Number of folds in CV.
stratified : bool, optional (default=True)
Whether to perform stratified sampling.
shuffle : bool, optional (default=True)
Whether to shuffle before splitting data.
metrics : string, list of strings or None, optional (default=None)
Evaluation metrics to be monitored while CV.
If not None, the metric in ``params`` will be overridden.
fobj : callable or None, optional (default=None)
Custom objective function.
feval : callable or None, optional (default=None)
Customized evaluation function.
Should accept two parameters: preds, train_data,
and return (eval_name, eval_result, is_higher_better) or list of such tuples.
For multi-class task, the preds is group by class_id first, then group by row_id.
If you want to get i-th row preds in j-th class, the access way is preds[j * num_data + i].
To ignore the default metric corresponding to the used objective,
set ``metrics`` to the string ``"None"``.
init_model : string, Booster or None, optional (default=None)
Filename of LightGBM model or Booster instance used for continue training.
feature_name : list of strings or 'auto', optional (default="auto")
Feature names.
If 'auto' and data is pandas DataFrame, data columns names are used.
categorical_feature : list of strings or int, or 'auto', optional (default="auto")
Categorical features.
If list of int, interpreted as indices.
If list of strings, interpreted as feature names (need to specify ``feature_name`` as well).
If 'auto' and data is pandas DataFrame, pandas unordered categorical columns are used.
All values in categorical features should be less than int32 max value (2147483647).
Large values could be memory consuming. Consider using consecutive integers starting from zero.
All negative values in categorical features will be treated as missing values.
early_stopping_rounds : int or None, optional (default=None)
Activates early stopping.
CV score needs to improve at least every ``early_stopping_rounds`` round(s)
to continue.
Requires at least one metric. If there's more than one, will check all of them.
To check only the first metric you can pass in ``callbacks``
``early_stopping`` callback with ``first_metric_only=True``.
Last entry in evaluation history is the one from the best iteration.
fpreproc : callable or None, optional (default=None)
Preprocessing function that takes (dtrain, dtest, params)
and returns transformed versions of those.
verbose_eval : bool, int, or None, optional (default=None)
Whether to display the progress.
If None, progress will be displayed when np.ndarray is returned.
If True, progress will be displayed at every boosting stage.
If int, progress will be displayed at every given ``verbose_eval`` boosting stage.
show_stdv : bool, optional (default=True)
Whether to display the standard deviation in progress.
Results are not affected by this parameter, and always contain std.
seed : int, optional (default=0)
Seed used to generate the folds (passed to numpy.random.seed).
callbacks : list of callables or None, optional (default=None)
List of callback functions that are applied at each iteration.
See Callbacks in Python API for more information.
eval_train_metric : bool, optional (default=False)
Whether to display the train metric in progress.
The score of the metric is calculated again after each training step, so there is some impact on performance.
Returns
-------
eval_hist : dict
Evaluation history.
The dictionary has the following format:
{'metric1-mean': [values], 'metric1-stdv': [values],
'metric2-mean': [values], 'metric2-stdv': [values],
...}. | def cv(params, train_set, num_boost_round=100,
folds=None, nfold=5, stratified=True, shuffle=True,
metrics=None, fobj=None, feval=None, init_model=None,
feature_name='auto', categorical_feature='auto',
early_stopping_rounds=None, fpreproc=None,
verbose_eval=None, show_stdv=True, seed=0,
callbacks=None, eval_train_metric=False):
"""Perform the cross-validation with given paramaters.
Parameters
----------
params : dict
Parameters for Booster.
train_set : Dataset
Data to be trained on.
num_boost_round : int, optional (default=100)
Number of boosting iterations.
folds : generator or iterator of (train_idx, test_idx) tuples, scikit-learn splitter object or None, optional (default=None)
If generator or iterator, it should yield the train and test indices for each fold.
If object, it should be one of the scikit-learn splitter classes
(https://scikit-learn.org/stable/modules/classes.html#splitter-classes)
and have ``split`` method.
This argument has highest priority over other data split arguments.
nfold : int, optional (default=5)
Number of folds in CV.
stratified : bool, optional (default=True)
Whether to perform stratified sampling.
shuffle : bool, optional (default=True)
Whether to shuffle before splitting data.
metrics : string, list of strings or None, optional (default=None)
Evaluation metrics to be monitored while CV.
If not None, the metric in ``params`` will be overridden.
fobj : callable or None, optional (default=None)
Custom objective function.
feval : callable or None, optional (default=None)
Customized evaluation function.
Should accept two parameters: preds, train_data,
and return (eval_name, eval_result, is_higher_better) or list of such tuples.
For multi-class task, the preds is group by class_id first, then group by row_id.
If you want to get i-th row preds in j-th class, the access way is preds[j * num_data + i].
To ignore the default metric corresponding to the used objective,
set ``metrics`` to the string ``"None"``.
init_model : string, Booster or None, optional (default=None)
Filename of LightGBM model or Booster instance used for continue training.
feature_name : list of strings or 'auto', optional (default="auto")
Feature names.
If 'auto' and data is pandas DataFrame, data columns names are used.
categorical_feature : list of strings or int, or 'auto', optional (default="auto")
Categorical features.
If list of int, interpreted as indices.
If list of strings, interpreted as feature names (need to specify ``feature_name`` as well).
If 'auto' and data is pandas DataFrame, pandas unordered categorical columns are used.
All values in categorical features should be less than int32 max value (2147483647).
Large values could be memory consuming. Consider using consecutive integers starting from zero.
All negative values in categorical features will be treated as missing values.
early_stopping_rounds : int or None, optional (default=None)
Activates early stopping.
CV score needs to improve at least every ``early_stopping_rounds`` round(s)
to continue.
Requires at least one metric. If there's more than one, will check all of them.
To check only the first metric you can pass in ``callbacks``
``early_stopping`` callback with ``first_metric_only=True``.
Last entry in evaluation history is the one from the best iteration.
fpreproc : callable or None, optional (default=None)
Preprocessing function that takes (dtrain, dtest, params)
and returns transformed versions of those.
verbose_eval : bool, int, or None, optional (default=None)
Whether to display the progress.
If None, progress will be displayed when np.ndarray is returned.
If True, progress will be displayed at every boosting stage.
If int, progress will be displayed at every given ``verbose_eval`` boosting stage.
show_stdv : bool, optional (default=True)
Whether to display the standard deviation in progress.
Results are not affected by this parameter, and always contain std.
seed : int, optional (default=0)
Seed used to generate the folds (passed to numpy.random.seed).
callbacks : list of callables or None, optional (default=None)
List of callback functions that are applied at each iteration.
See Callbacks in Python API for more information.
eval_train_metric : bool, optional (default=False)
Whether to display the train metric in progress.
The score of the metric is calculated again after each training step, so there is some impact on performance.
Returns
-------
eval_hist : dict
Evaluation history.
The dictionary has the following format:
{'metric1-mean': [values], 'metric1-stdv': [values],
'metric2-mean': [values], 'metric2-stdv': [values],
...}.
"""
if not isinstance(train_set, Dataset):
raise TypeError("Traninig only accepts Dataset object")
params = copy.deepcopy(params)
if fobj is not None:
params['objective'] = 'none'
for alias in ["num_iterations", "num_iteration", "n_iter", "num_tree", "num_trees",
"num_round", "num_rounds", "num_boost_round", "n_estimators"]:
if alias in params:
warnings.warn("Found `{}` in params. Will use it instead of argument".format(alias))
num_boost_round = params.pop(alias)
break
for alias in ["early_stopping_round", "early_stopping_rounds", "early_stopping"]:
if alias in params:
warnings.warn("Found `{}` in params. Will use it instead of argument".format(alias))
early_stopping_rounds = params.pop(alias)
break
if num_boost_round <= 0:
raise ValueError("num_boost_round should be greater than zero.")
if isinstance(init_model, string_type):
predictor = _InnerPredictor(model_file=init_model, pred_parameter=params)
elif isinstance(init_model, Booster):
predictor = init_model._to_predictor(dict(init_model.params, **params))
else:
predictor = None
train_set._update_params(params) \
._set_predictor(predictor) \
.set_feature_name(feature_name) \
.set_categorical_feature(categorical_feature)
if metrics is not None:
params['metric'] = metrics
results = collections.defaultdict(list)
cvfolds = _make_n_folds(train_set, folds=folds, nfold=nfold,
params=params, seed=seed, fpreproc=fpreproc,
stratified=stratified, shuffle=shuffle,
eval_train_metric=eval_train_metric)
# setup callbacks
if callbacks is None:
callbacks = set()
else:
for i, cb in enumerate(callbacks):
cb.__dict__.setdefault('order', i - len(callbacks))
callbacks = set(callbacks)
if early_stopping_rounds is not None:
callbacks.add(callback.early_stopping(early_stopping_rounds, verbose=False))
if verbose_eval is True:
callbacks.add(callback.print_evaluation(show_stdv=show_stdv))
elif isinstance(verbose_eval, integer_types):
callbacks.add(callback.print_evaluation(verbose_eval, show_stdv=show_stdv))
callbacks_before_iter = {cb for cb in callbacks if getattr(cb, 'before_iteration', False)}
callbacks_after_iter = callbacks - callbacks_before_iter
callbacks_before_iter = sorted(callbacks_before_iter, key=attrgetter('order'))
callbacks_after_iter = sorted(callbacks_after_iter, key=attrgetter('order'))
for i in range_(num_boost_round):
for cb in callbacks_before_iter:
cb(callback.CallbackEnv(model=cvfolds,
params=params,
iteration=i,
begin_iteration=0,
end_iteration=num_boost_round,
evaluation_result_list=None))
cvfolds.update(fobj=fobj)
res = _agg_cv_result(cvfolds.eval_valid(feval), eval_train_metric)
for _, key, mean, _, std in res:
results[key + '-mean'].append(mean)
results[key + '-stdv'].append(std)
try:
for cb in callbacks_after_iter:
cb(callback.CallbackEnv(model=cvfolds,
params=params,
iteration=i,
begin_iteration=0,
end_iteration=num_boost_round,
evaluation_result_list=res))
except callback.EarlyStopException as earlyStopException:
cvfolds.best_iteration = earlyStopException.best_iteration + 1
for k in results:
results[k] = results[k][:cvfolds.best_iteration]
break
return dict(results) |
Logarithmic loss with non-necessarily-binary labels. | def log_loss(preds, labels):
"""Logarithmic loss with non-necessarily-binary labels."""
log_likelihood = np.sum(labels * np.log(preds)) / len(preds)
return -log_likelihood |
Measure performance of an objective.
Parameters
----------
objective : string 'binary' or 'xentropy'
Objective function.
label_type : string 'binary' or 'probability'
Type of the label.
data : dict
Data for training.
Returns
-------
result : dict
Experiment summary stats. | def experiment(objective, label_type, data):
"""Measure performance of an objective.
Parameters
----------
objective : string 'binary' or 'xentropy'
Objective function.
label_type : string 'binary' or 'probability'
Type of the label.
data : dict
Data for training.
Returns
-------
result : dict
Experiment summary stats.
"""
np.random.seed(0)
nrounds = 5
lgb_data = data['lgb_with_' + label_type + '_labels']
params = {
'objective': objective,
'feature_fraction': 1,
'bagging_fraction': 1,
'verbose': -1
}
time_zero = time.time()
gbm = lgb.train(params, lgb_data, num_boost_round=nrounds)
y_fitted = gbm.predict(data['X'])
y_true = data[label_type + '_labels']
duration = time.time() - time_zero
return {
'time': duration,
'correlation': np.corrcoef(y_fitted, y_true)[0, 1],
'logloss': log_loss(y_fitted, y_true)
} |
Check object is not tuple or does not have 2 elements. | def _check_not_tuple_of_2_elements(obj, obj_name='obj'):
"""Check object is not tuple or does not have 2 elements."""
if not isinstance(obj, tuple) or len(obj) != 2:
raise TypeError('%s must be a tuple of 2 elements.' % obj_name) |
Plot model's feature importances.
Parameters
----------
booster : Booster or LGBMModel
Booster or LGBMModel instance which feature importance should be plotted.
ax : matplotlib.axes.Axes or None, optional (default=None)
Target axes instance.
If None, new figure and axes will be created.
height : float, optional (default=0.2)
Bar height, passed to ``ax.barh()``.
xlim : tuple of 2 elements or None, optional (default=None)
Tuple passed to ``ax.xlim()``.
ylim : tuple of 2 elements or None, optional (default=None)
Tuple passed to ``ax.ylim()``.
title : string or None, optional (default="Feature importance")
Axes title.
If None, title is disabled.
xlabel : string or None, optional (default="Feature importance")
X-axis title label.
If None, title is disabled.
ylabel : string or None, optional (default="Features")
Y-axis title label.
If None, title is disabled.
importance_type : string, optional (default="split")
How the importance is calculated.
If "split", result contains numbers of times the feature is used in a model.
If "gain", result contains total gains of splits which use the feature.
max_num_features : int or None, optional (default=None)
Max number of top features displayed on plot.
If None or <1, all features will be displayed.
ignore_zero : bool, optional (default=True)
Whether to ignore features with zero importance.
figsize : tuple of 2 elements or None, optional (default=None)
Figure size.
grid : bool, optional (default=True)
Whether to add a grid for axes.
precision : int or None, optional (default=None)
Used to restrict the display of floating point values to a certain precision.
**kwargs
Other parameters passed to ``ax.barh()``.
Returns
-------
ax : matplotlib.axes.Axes
The plot with model's feature importances. | def plot_importance(booster, ax=None, height=0.2,
xlim=None, ylim=None, title='Feature importance',
xlabel='Feature importance', ylabel='Features',
importance_type='split', max_num_features=None,
ignore_zero=True, figsize=None, grid=True,
precision=None, **kwargs):
"""Plot model's feature importances.
Parameters
----------
booster : Booster or LGBMModel
Booster or LGBMModel instance which feature importance should be plotted.
ax : matplotlib.axes.Axes or None, optional (default=None)
Target axes instance.
If None, new figure and axes will be created.
height : float, optional (default=0.2)
Bar height, passed to ``ax.barh()``.
xlim : tuple of 2 elements or None, optional (default=None)
Tuple passed to ``ax.xlim()``.
ylim : tuple of 2 elements or None, optional (default=None)
Tuple passed to ``ax.ylim()``.
title : string or None, optional (default="Feature importance")
Axes title.
If None, title is disabled.
xlabel : string or None, optional (default="Feature importance")
X-axis title label.
If None, title is disabled.
ylabel : string or None, optional (default="Features")
Y-axis title label.
If None, title is disabled.
importance_type : string, optional (default="split")
How the importance is calculated.
If "split", result contains numbers of times the feature is used in a model.
If "gain", result contains total gains of splits which use the feature.
max_num_features : int or None, optional (default=None)
Max number of top features displayed on plot.
If None or <1, all features will be displayed.
ignore_zero : bool, optional (default=True)
Whether to ignore features with zero importance.
figsize : tuple of 2 elements or None, optional (default=None)
Figure size.
grid : bool, optional (default=True)
Whether to add a grid for axes.
precision : int or None, optional (default=None)
Used to restrict the display of floating point values to a certain precision.
**kwargs
Other parameters passed to ``ax.barh()``.
Returns
-------
ax : matplotlib.axes.Axes
The plot with model's feature importances.
"""
if MATPLOTLIB_INSTALLED:
import matplotlib.pyplot as plt
else:
raise ImportError('You must install matplotlib to plot importance.')
if isinstance(booster, LGBMModel):
booster = booster.booster_
elif not isinstance(booster, Booster):
raise TypeError('booster must be Booster or LGBMModel.')
importance = booster.feature_importance(importance_type=importance_type)
feature_name = booster.feature_name()
if not len(importance):
raise ValueError("Booster's feature_importance is empty.")
tuples = sorted(zip_(feature_name, importance), key=lambda x: x[1])
if ignore_zero:
tuples = [x for x in tuples if x[1] > 0]
if max_num_features is not None and max_num_features > 0:
tuples = tuples[-max_num_features:]
labels, values = zip_(*tuples)
if ax is None:
if figsize is not None:
_check_not_tuple_of_2_elements(figsize, 'figsize')
_, ax = plt.subplots(1, 1, figsize=figsize)
ylocs = np.arange(len(values))
ax.barh(ylocs, values, align='center', height=height, **kwargs)
for x, y in zip_(values, ylocs):
ax.text(x + 1, y,
_float2str(x, precision) if importance_type == 'gain' else x,
va='center')
ax.set_yticks(ylocs)
ax.set_yticklabels(labels)
if xlim is not None:
_check_not_tuple_of_2_elements(xlim, 'xlim')
else:
xlim = (0, max(values) * 1.1)
ax.set_xlim(xlim)
if ylim is not None:
_check_not_tuple_of_2_elements(ylim, 'ylim')
else:
ylim = (-1, len(values))
ax.set_ylim(ylim)
if title is not None:
ax.set_title(title)
if xlabel is not None:
ax.set_xlabel(xlabel)
if ylabel is not None:
ax.set_ylabel(ylabel)
ax.grid(grid)
return ax |
Plot one metric during training.
Parameters
----------
booster : dict or LGBMModel
Dictionary returned from ``lightgbm.train()`` or LGBMModel instance.
metric : string or None, optional (default=None)
The metric name to plot.
Only one metric supported because different metrics have various scales.
If None, first metric picked from dictionary (according to hashcode).
dataset_names : list of strings or None, optional (default=None)
List of the dataset names which are used to calculate metric to plot.
If None, all datasets are used.
ax : matplotlib.axes.Axes or None, optional (default=None)
Target axes instance.
If None, new figure and axes will be created.
xlim : tuple of 2 elements or None, optional (default=None)
Tuple passed to ``ax.xlim()``.
ylim : tuple of 2 elements or None, optional (default=None)
Tuple passed to ``ax.ylim()``.
title : string or None, optional (default="Metric during training")
Axes title.
If None, title is disabled.
xlabel : string or None, optional (default="Iterations")
X-axis title label.
If None, title is disabled.
ylabel : string or None, optional (default="auto")
Y-axis title label.
If 'auto', metric name is used.
If None, title is disabled.
figsize : tuple of 2 elements or None, optional (default=None)
Figure size.
grid : bool, optional (default=True)
Whether to add a grid for axes.
Returns
-------
ax : matplotlib.axes.Axes
The plot with metric's history over the training. | def plot_metric(booster, metric=None, dataset_names=None,
ax=None, xlim=None, ylim=None,
title='Metric during training',
xlabel='Iterations', ylabel='auto',
figsize=None, grid=True):
"""Plot one metric during training.
Parameters
----------
booster : dict or LGBMModel
Dictionary returned from ``lightgbm.train()`` or LGBMModel instance.
metric : string or None, optional (default=None)
The metric name to plot.
Only one metric supported because different metrics have various scales.
If None, first metric picked from dictionary (according to hashcode).
dataset_names : list of strings or None, optional (default=None)
List of the dataset names which are used to calculate metric to plot.
If None, all datasets are used.
ax : matplotlib.axes.Axes or None, optional (default=None)
Target axes instance.
If None, new figure and axes will be created.
xlim : tuple of 2 elements or None, optional (default=None)
Tuple passed to ``ax.xlim()``.
ylim : tuple of 2 elements or None, optional (default=None)
Tuple passed to ``ax.ylim()``.
title : string or None, optional (default="Metric during training")
Axes title.
If None, title is disabled.
xlabel : string or None, optional (default="Iterations")
X-axis title label.
If None, title is disabled.
ylabel : string or None, optional (default="auto")
Y-axis title label.
If 'auto', metric name is used.
If None, title is disabled.
figsize : tuple of 2 elements or None, optional (default=None)
Figure size.
grid : bool, optional (default=True)
Whether to add a grid for axes.
Returns
-------
ax : matplotlib.axes.Axes
The plot with metric's history over the training.
"""
if MATPLOTLIB_INSTALLED:
import matplotlib.pyplot as plt
else:
raise ImportError('You must install matplotlib to plot metric.')
if isinstance(booster, LGBMModel):
eval_results = deepcopy(booster.evals_result_)
elif isinstance(booster, dict):
eval_results = deepcopy(booster)
else:
raise TypeError('booster must be dict or LGBMModel.')
num_data = len(eval_results)
if not num_data:
raise ValueError('eval results cannot be empty.')
if ax is None:
if figsize is not None:
_check_not_tuple_of_2_elements(figsize, 'figsize')
_, ax = plt.subplots(1, 1, figsize=figsize)
if dataset_names is None:
dataset_names = iter(eval_results.keys())
elif not isinstance(dataset_names, (list, tuple, set)) or not dataset_names:
raise ValueError('dataset_names should be iterable and cannot be empty')
else:
dataset_names = iter(dataset_names)
name = next(dataset_names) # take one as sample
metrics_for_one = eval_results[name]
num_metric = len(metrics_for_one)
if metric is None:
if num_metric > 1:
msg = """more than one metric available, picking one to plot."""
warnings.warn(msg, stacklevel=2)
metric, results = metrics_for_one.popitem()
else:
if metric not in metrics_for_one:
raise KeyError('No given metric in eval results.')
results = metrics_for_one[metric]
num_iteration, max_result, min_result = len(results), max(results), min(results)
x_ = range_(num_iteration)
ax.plot(x_, results, label=name)
for name in dataset_names:
metrics_for_one = eval_results[name]
results = metrics_for_one[metric]
max_result, min_result = max(max(results), max_result), min(min(results), min_result)
ax.plot(x_, results, label=name)
ax.legend(loc='best')
if xlim is not None:
_check_not_tuple_of_2_elements(xlim, 'xlim')
else:
xlim = (0, num_iteration)
ax.set_xlim(xlim)
if ylim is not None:
_check_not_tuple_of_2_elements(ylim, 'ylim')
else:
range_result = max_result - min_result
ylim = (min_result - range_result * 0.2, max_result + range_result * 0.2)
ax.set_ylim(ylim)
if ylabel == 'auto':
ylabel = metric
if title is not None:
ax.set_title(title)
if xlabel is not None:
ax.set_xlabel(xlabel)
if ylabel is not None:
ax.set_ylabel(ylabel)
ax.grid(grid)
return ax |
Convert specified tree to graphviz instance.
See:
- https://graphviz.readthedocs.io/en/stable/api.html#digraph | def _to_graphviz(tree_info, show_info, feature_names, precision=None, **kwargs):
"""Convert specified tree to graphviz instance.
See:
- https://graphviz.readthedocs.io/en/stable/api.html#digraph
"""
if GRAPHVIZ_INSTALLED:
from graphviz import Digraph
else:
raise ImportError('You must install graphviz to plot tree.')
def add(root, parent=None, decision=None):
"""Recursively add node or edge."""
if 'split_index' in root: # non-leaf
name = 'split{0}'.format(root['split_index'])
if feature_names is not None:
label = 'split_feature_name: {0}'.format(feature_names[root['split_feature']])
else:
label = 'split_feature_index: {0}'.format(root['split_feature'])
label += r'\nthreshold: {0}'.format(_float2str(root['threshold'], precision))
for info in show_info:
if info in {'split_gain', 'internal_value'}:
label += r'\n{0}: {1}'.format(info, _float2str(root[info], precision))
elif info == 'internal_count':
label += r'\n{0}: {1}'.format(info, root[info])
graph.node(name, label=label)
if root['decision_type'] == '<=':
l_dec, r_dec = '<=', '>'
elif root['decision_type'] == '==':
l_dec, r_dec = 'is', "isn't"
else:
raise ValueError('Invalid decision type in tree model.')
add(root['left_child'], name, l_dec)
add(root['right_child'], name, r_dec)
else: # leaf
name = 'leaf{0}'.format(root['leaf_index'])
label = 'leaf_index: {0}'.format(root['leaf_index'])
label += r'\nleaf_value: {0}'.format(_float2str(root['leaf_value'], precision))
if 'leaf_count' in show_info:
label += r'\nleaf_count: {0}'.format(root['leaf_count'])
graph.node(name, label=label)
if parent is not None:
graph.edge(parent, name, decision)
graph = Digraph(**kwargs)
add(tree_info['tree_structure'])
return graph |
Create a digraph representation of specified tree.
Note
----
For more information please visit
https://graphviz.readthedocs.io/en/stable/api.html#digraph.
Parameters
----------
booster : Booster or LGBMModel
Booster or LGBMModel instance to be converted.
tree_index : int, optional (default=0)
The index of a target tree to convert.
show_info : list of strings or None, optional (default=None)
What information should be shown in nodes.
Possible values of list items: 'split_gain', 'internal_value', 'internal_count', 'leaf_count'.
precision : int or None, optional (default=None)
Used to restrict the display of floating point values to a certain precision.
**kwargs
Other parameters passed to ``Digraph`` constructor.
Check https://graphviz.readthedocs.io/en/stable/api.html#digraph for the full list of supported parameters.
Returns
-------
graph : graphviz.Digraph
The digraph representation of specified tree. | def create_tree_digraph(booster, tree_index=0, show_info=None, precision=None,
old_name=None, old_comment=None, old_filename=None, old_directory=None,
old_format=None, old_engine=None, old_encoding=None, old_graph_attr=None,
old_node_attr=None, old_edge_attr=None, old_body=None, old_strict=False, **kwargs):
"""Create a digraph representation of specified tree.
Note
----
For more information please visit
https://graphviz.readthedocs.io/en/stable/api.html#digraph.
Parameters
----------
booster : Booster or LGBMModel
Booster or LGBMModel instance to be converted.
tree_index : int, optional (default=0)
The index of a target tree to convert.
show_info : list of strings or None, optional (default=None)
What information should be shown in nodes.
Possible values of list items: 'split_gain', 'internal_value', 'internal_count', 'leaf_count'.
precision : int or None, optional (default=None)
Used to restrict the display of floating point values to a certain precision.
**kwargs
Other parameters passed to ``Digraph`` constructor.
Check https://graphviz.readthedocs.io/en/stable/api.html#digraph for the full list of supported parameters.
Returns
-------
graph : graphviz.Digraph
The digraph representation of specified tree.
"""
if isinstance(booster, LGBMModel):
booster = booster.booster_
elif not isinstance(booster, Booster):
raise TypeError('booster must be Booster or LGBMModel.')
for param_name in ['old_name', 'old_comment', 'old_filename', 'old_directory',
'old_format', 'old_engine', 'old_encoding', 'old_graph_attr',
'old_node_attr', 'old_edge_attr', 'old_body']:
param = locals().get(param_name)
if param is not None:
warnings.warn('{0} parameter is deprecated and will be removed in 2.4 version.\n'
'Please use **kwargs to pass {1} parameter.'.format(param_name, param_name[4:]),
LGBMDeprecationWarning)
if param_name[4:] not in kwargs:
kwargs[param_name[4:]] = param
if locals().get('strict'):
warnings.warn('old_strict parameter is deprecated and will be removed in 2.4 version.\n'
'Please use **kwargs to pass strict parameter.',
LGBMDeprecationWarning)
if 'strict' not in kwargs:
kwargs['strict'] = True
model = booster.dump_model()
tree_infos = model['tree_info']
if 'feature_names' in model:
feature_names = model['feature_names']
else:
feature_names = None
if tree_index < len(tree_infos):
tree_info = tree_infos[tree_index]
else:
raise IndexError('tree_index is out of range.')
if show_info is None:
show_info = []
graph = _to_graphviz(tree_info, show_info, feature_names, precision, **kwargs)
return graph |
Plot specified tree.
Note
----
It is preferable to use ``create_tree_digraph()`` because of its lossless quality
and returned objects can be also rendered and displayed directly inside a Jupyter notebook.
Parameters
----------
booster : Booster or LGBMModel
Booster or LGBMModel instance to be plotted.
ax : matplotlib.axes.Axes or None, optional (default=None)
Target axes instance.
If None, new figure and axes will be created.
tree_index : int, optional (default=0)
The index of a target tree to plot.
figsize : tuple of 2 elements or None, optional (default=None)
Figure size.
show_info : list of strings or None, optional (default=None)
What information should be shown in nodes.
Possible values of list items: 'split_gain', 'internal_value', 'internal_count', 'leaf_count'.
precision : int or None, optional (default=None)
Used to restrict the display of floating point values to a certain precision.
**kwargs
Other parameters passed to ``Digraph`` constructor.
Check https://graphviz.readthedocs.io/en/stable/api.html#digraph for the full list of supported parameters.
Returns
-------
ax : matplotlib.axes.Axes
The plot with single tree. | def plot_tree(booster, ax=None, tree_index=0, figsize=None,
old_graph_attr=None, old_node_attr=None, old_edge_attr=None,
show_info=None, precision=None, **kwargs):
"""Plot specified tree.
Note
----
It is preferable to use ``create_tree_digraph()`` because of its lossless quality
and returned objects can be also rendered and displayed directly inside a Jupyter notebook.
Parameters
----------
booster : Booster or LGBMModel
Booster or LGBMModel instance to be plotted.
ax : matplotlib.axes.Axes or None, optional (default=None)
Target axes instance.
If None, new figure and axes will be created.
tree_index : int, optional (default=0)
The index of a target tree to plot.
figsize : tuple of 2 elements or None, optional (default=None)
Figure size.
show_info : list of strings or None, optional (default=None)
What information should be shown in nodes.
Possible values of list items: 'split_gain', 'internal_value', 'internal_count', 'leaf_count'.
precision : int or None, optional (default=None)
Used to restrict the display of floating point values to a certain precision.
**kwargs
Other parameters passed to ``Digraph`` constructor.
Check https://graphviz.readthedocs.io/en/stable/api.html#digraph for the full list of supported parameters.
Returns
-------
ax : matplotlib.axes.Axes
The plot with single tree.
"""
if MATPLOTLIB_INSTALLED:
import matplotlib.pyplot as plt
import matplotlib.image as image
else:
raise ImportError('You must install matplotlib to plot tree.')
for param_name in ['old_graph_attr', 'old_node_attr', 'old_edge_attr']:
param = locals().get(param_name)
if param is not None:
warnings.warn('{0} parameter is deprecated and will be removed in 2.4 version.\n'
'Please use **kwargs to pass {1} parameter.'.format(param_name, param_name[4:]),
LGBMDeprecationWarning)
if param_name[4:] not in kwargs:
kwargs[param_name[4:]] = param
if ax is None:
if figsize is not None:
_check_not_tuple_of_2_elements(figsize, 'figsize')
_, ax = plt.subplots(1, 1, figsize=figsize)
graph = create_tree_digraph(booster=booster, tree_index=tree_index,
show_info=show_info, precision=precision, **kwargs)
s = BytesIO()
s.write(graph.pipe(format='png'))
s.seek(0)
img = image.imread(s)
ax.imshow(img)
ax.axis('off')
return ax |
Return the -std=c++[0x/11/14] compiler flag.
The c++14 is preferred over c++0x/11 (when it is available). | def cpp_flag(compiler):
"""Return the -std=c++[0x/11/14] compiler flag.
The c++14 is preferred over c++0x/11 (when it is available).
"""
standards = ['-std=c++14', '-std=c++11', '-std=c++0x']
for standard in standards:
if has_flag(compiler, [standard]):
return standard
raise RuntimeError(
'Unsupported compiler -- at least C++0x support '
'is needed!'
) |
query is a 1d numpy array corresponding to the vector to which you want to
find the closest vector
vectors is a 2d numpy array corresponding to the vectors you want to consider
ban_set is a set of indicies within vectors you want to ignore for nearest match
cossims is a 1d numpy array of size len(vectors), which can be passed for efficiency
returns the index of the closest match to query within vectors | def find_nearest_neighbor(query, vectors, ban_set, cossims=None):
"""
query is a 1d numpy array corresponding to the vector to which you want to
find the closest vector
vectors is a 2d numpy array corresponding to the vectors you want to consider
ban_set is a set of indicies within vectors you want to ignore for nearest match
cossims is a 1d numpy array of size len(vectors), which can be passed for efficiency
returns the index of the closest match to query within vectors
"""
if cossims is None:
cossims = np.matmul(vectors, query, out=cossims)
else:
np.matmul(vectors, query, out=cossims)
rank = len(cossims) - 1
result_i = np.argpartition(cossims, rank)[rank]
while result_i in ban_set:
rank -= 1
result_i = np.argpartition(cossims, rank)[rank]
return result_i |
Train a supervised model and return a model object.
input must be a filepath. The input text does not need to be tokenized
as per the tokenize function, but it must be preprocessed and encoded
as UTF-8. You might want to consult standard preprocessing scripts such
as tokenizer.perl mentioned here: http://www.statmt.org/wmt07/baseline.html
The input file must must contain at least one label per line. For an
example consult the example datasets which are part of the fastText
repository such as the dataset pulled by classification-example.sh. | def train_supervised(
input,
lr=0.1,
dim=100,
ws=5,
epoch=5,
minCount=1,
minCountLabel=0,
minn=0,
maxn=0,
neg=5,
wordNgrams=1,
loss="softmax",
bucket=2000000,
thread=multiprocessing.cpu_count() - 1,
lrUpdateRate=100,
t=1e-4,
label="__label__",
verbose=2,
pretrainedVectors="",
):
"""
Train a supervised model and return a model object.
input must be a filepath. The input text does not need to be tokenized
as per the tokenize function, but it must be preprocessed and encoded
as UTF-8. You might want to consult standard preprocessing scripts such
as tokenizer.perl mentioned here: http://www.statmt.org/wmt07/baseline.html
The input file must must contain at least one label per line. For an
example consult the example datasets which are part of the fastText
repository such as the dataset pulled by classification-example.sh.
"""
model = "supervised"
a = _build_args(locals())
ft = _FastText()
fasttext.train(ft.f, a)
return ft |
Get the vector representation of word. | def get_word_vector(self, word):
"""Get the vector representation of word."""
dim = self.get_dimension()
b = fasttext.Vector(dim)
self.f.getWordVector(b, word)
return np.array(b) |
Given a string, get a single vector represenation. This function
assumes to be given a single line of text. We split words on
whitespace (space, newline, tab, vertical tab) and the control
characters carriage return, formfeed and the null character. | def get_sentence_vector(self, text):
"""
Given a string, get a single vector represenation. This function
assumes to be given a single line of text. We split words on
whitespace (space, newline, tab, vertical tab) and the control
characters carriage return, formfeed and the null character.
"""
if text.find('\n') != -1:
raise ValueError(
"predict processes one line at a time (remove \'\\n\')"
)
text += "\n"
dim = self.get_dimension()
b = fasttext.Vector(dim)
self.f.getSentenceVector(b, text)
return np.array(b) |
Given a word, get the subwords and their indicies. | def get_subwords(self, word, on_unicode_error='strict'):
"""
Given a word, get the subwords and their indicies.
"""
pair = self.f.getSubwords(word, on_unicode_error)
return pair[0], np.array(pair[1]) |
Given an index, get the corresponding vector of the Input Matrix. | def get_input_vector(self, ind):
"""
Given an index, get the corresponding vector of the Input Matrix.
"""
dim = self.get_dimension()
b = fasttext.Vector(dim)
self.f.getInputVector(b, ind)
return np.array(b) |
Given a string, get a list of labels and a list of
corresponding probabilities. k controls the number
of returned labels. A choice of 5, will return the 5
most probable labels. By default this returns only
the most likely label and probability. threshold filters
the returned labels by a threshold on probability. A
choice of 0.5 will return labels with at least 0.5
probability. k and threshold will be applied together to
determine the returned labels.
This function assumes to be given
a single line of text. We split words on whitespace (space,
newline, tab, vertical tab) and the control characters carriage
return, formfeed and the null character.
If the model is not supervised, this function will throw a ValueError.
If given a list of strings, it will return a list of results as usually
received for a single line of text. | def predict(self, text, k=1, threshold=0.0, on_unicode_error='strict'):
"""
Given a string, get a list of labels and a list of
corresponding probabilities. k controls the number
of returned labels. A choice of 5, will return the 5
most probable labels. By default this returns only
the most likely label and probability. threshold filters
the returned labels by a threshold on probability. A
choice of 0.5 will return labels with at least 0.5
probability. k and threshold will be applied together to
determine the returned labels.
This function assumes to be given
a single line of text. We split words on whitespace (space,
newline, tab, vertical tab) and the control characters carriage
return, formfeed and the null character.
If the model is not supervised, this function will throw a ValueError.
If given a list of strings, it will return a list of results as usually
received for a single line of text.
"""
def check(entry):
if entry.find('\n') != -1:
raise ValueError(
"predict processes one line at a time (remove \'\\n\')"
)
entry += "\n"
return entry
if type(text) == list:
text = [check(entry) for entry in text]
predictions = self.f.multilinePredict(text, k, threshold, on_unicode_error)
dt = np.dtype([('probability', 'float64'), ('label', 'object')])
result_as_pair = np.array(predictions, dtype=dt)
return result_as_pair['label'].tolist(), result_as_pair['probability']
else:
text = check(text)
predictions = self.f.predict(text, k, threshold, on_unicode_error)
probs, labels = zip(*predictions)
return labels, np.array(probs, copy=False) |
Get a copy of the full input matrix of a Model. This only
works if the model is not quantized. | def get_input_matrix(self):
"""
Get a copy of the full input matrix of a Model. This only
works if the model is not quantized.
"""
if self.f.isQuant():
raise ValueError("Can't get quantized Matrix")
return np.array(self.f.getInputMatrix()) |
Get a copy of the full output matrix of a Model. This only
works if the model is not quantized. | def get_output_matrix(self):
"""
Get a copy of the full output matrix of a Model. This only
works if the model is not quantized.
"""
if self.f.isQuant():
raise ValueError("Can't get quantized Matrix")
return np.array(self.f.getOutputMatrix()) |
Get the entire list of words of the dictionary optionally
including the frequency of the individual words. This
does not include any subwords. For that please consult
the function get_subwords. | def get_words(self, include_freq=False, on_unicode_error='strict'):
"""
Get the entire list of words of the dictionary optionally
including the frequency of the individual words. This
does not include any subwords. For that please consult
the function get_subwords.
"""
pair = self.f.getVocab(on_unicode_error)
if include_freq:
return (pair[0], np.array(pair[1]))
else:
return pair[0] |
Get the entire list of labels of the dictionary optionally
including the frequency of the individual labels. Unsupervised
models use words as labels, which is why get_labels
will call and return get_words for this type of
model. | def get_labels(self, include_freq=False, on_unicode_error='strict'):
"""
Get the entire list of labels of the dictionary optionally
including the frequency of the individual labels. Unsupervised
models use words as labels, which is why get_labels
will call and return get_words for this type of
model.
"""
a = self.f.getArgs()
if a.model == model_name.supervised:
pair = self.f.getLabels(on_unicode_error)
if include_freq:
return (pair[0], np.array(pair[1]))
else:
return pair[0]
else:
return self.get_words(include_freq) |
Split a line of text into words and labels. Labels must start with
the prefix used to create the model (__label__ by default). | def get_line(self, text, on_unicode_error='strict'):
"""
Split a line of text into words and labels. Labels must start with
the prefix used to create the model (__label__ by default).
"""
def check(entry):
if entry.find('\n') != -1:
raise ValueError(
"get_line processes one line at a time (remove \'\\n\')"
)
entry += "\n"
return entry
if type(text) == list:
text = [check(entry) for entry in text]
return self.f.multilineGetLine(text, on_unicode_error)
else:
text = check(text)
return self.f.getLine(text, on_unicode_error) |
Quantize the model reducing the size of the model and
it's memory footprint. | def quantize(
self,
input=None,
qout=False,
cutoff=0,
retrain=False,
epoch=None,
lr=None,
thread=None,
verbose=None,
dsub=2,
qnorm=False
):
"""
Quantize the model reducing the size of the model and
it's memory footprint.
"""
a = self.f.getArgs()
if not epoch:
epoch = a.epoch
if not lr:
lr = a.lr
if not thread:
thread = a.thread
if not verbose:
verbose = a.verbose
if retrain and not input:
raise ValueError("Need input file path if retraining")
if input is None:
input = ""
self.f.quantize(
input, qout, cutoff, retrain, epoch, lr, thread, verbose, dsub,
qnorm
) |
Parameters
----------
inputs : ``PackedSequence``, required.
A batch first ``PackedSequence`` to run the stacked LSTM over.
initial_state : Tuple[torch.Tensor, torch.Tensor], optional, (default = None)
A tuple (state, memory) representing the initial hidden state and memory
of the LSTM. Each tensor has shape (num_layers, batch_size, output_dimension * 2).
Returns
-------
output_sequence : PackedSequence
The encoded sequence of shape (batch_size, sequence_length, hidden_size * 2)
final_states: torch.Tensor
The per-layer final (state, memory) states of the LSTM, each with shape
(num_layers * 2, batch_size, hidden_size * 2). | def forward(self, # pylint: disable=arguments-differ
inputs: PackedSequence,
initial_state: Optional[Tuple[torch.Tensor, torch.Tensor]] = None
) -> Tuple[PackedSequence, Tuple[torch.Tensor, torch.Tensor]]:
"""
Parameters
----------
inputs : ``PackedSequence``, required.
A batch first ``PackedSequence`` to run the stacked LSTM over.
initial_state : Tuple[torch.Tensor, torch.Tensor], optional, (default = None)
A tuple (state, memory) representing the initial hidden state and memory
of the LSTM. Each tensor has shape (num_layers, batch_size, output_dimension * 2).
Returns
-------
output_sequence : PackedSequence
The encoded sequence of shape (batch_size, sequence_length, hidden_size * 2)
final_states: torch.Tensor
The per-layer final (state, memory) states of the LSTM, each with shape
(num_layers * 2, batch_size, hidden_size * 2).
"""
if not initial_state:
hidden_states = [None] * len(self.lstm_layers)
elif initial_state[0].size()[0] != len(self.lstm_layers):
raise ConfigurationError("Initial states were passed to forward() but the number of "
"initial states does not match the number of layers.")
else:
hidden_states = list(zip(initial_state[0].split(1, 0),
initial_state[1].split(1, 0)))
output_sequence = inputs
final_h = []
final_c = []
for i, state in enumerate(hidden_states):
forward_layer = getattr(self, 'forward_layer_{}'.format(i))
backward_layer = getattr(self, 'backward_layer_{}'.format(i))
# The state is duplicated to mirror the Pytorch API for LSTMs.
forward_output, final_forward_state = forward_layer(output_sequence, state)
backward_output, final_backward_state = backward_layer(output_sequence, state)
forward_output, lengths = pad_packed_sequence(forward_output, batch_first=True)
backward_output, _ = pad_packed_sequence(backward_output, batch_first=True)
output_sequence = torch.cat([forward_output, backward_output], -1)
# Apply layer wise dropout on each output sequence apart from the
# first (input) and last
if i < (self.num_layers - 1):
output_sequence = self.layer_dropout(output_sequence)
output_sequence = pack_padded_sequence(output_sequence, lengths, batch_first=True)
final_h.extend([final_forward_state[0], final_backward_state[0]])
final_c.extend([final_forward_state[1], final_backward_state[1]])
final_h = torch.cat(final_h, dim=0)
final_c = torch.cat(final_c, dim=0)
final_state_tuple = (final_h, final_c)
return output_sequence, final_state_tuple |
Converts a List of pairs (regex, params) into an RegularizerApplicator.
This list should look like
[["regex1", {"type": "l2", "alpha": 0.01}], ["regex2", "l1"]]
where each parameter receives the penalty corresponding to the first regex
that matches its name (which may be no regex and hence no penalty).
The values can either be strings, in which case they correspond to the names
of regularizers, or dictionaries, in which case they must contain the "type"
key, corresponding to the name of a regularizer. In addition, they may contain
auxiliary named parameters which will be fed to the regularizer itself.
To determine valid auxiliary parameters, please refer to the torch.nn.init documentation.
Parameters
----------
params : ``Params``, required.
A Params object containing a "regularizers" key.
Returns
-------
A RegularizerApplicator containing the specified Regularizers,
or ``None`` if no Regularizers are specified. | def from_params(cls, params: Iterable[Tuple[str, Params]] = ()) -> Optional['RegularizerApplicator']:
"""
Converts a List of pairs (regex, params) into an RegularizerApplicator.
This list should look like
[["regex1", {"type": "l2", "alpha": 0.01}], ["regex2", "l1"]]
where each parameter receives the penalty corresponding to the first regex
that matches its name (which may be no regex and hence no penalty).
The values can either be strings, in which case they correspond to the names
of regularizers, or dictionaries, in which case they must contain the "type"
key, corresponding to the name of a regularizer. In addition, they may contain
auxiliary named parameters which will be fed to the regularizer itself.
To determine valid auxiliary parameters, please refer to the torch.nn.init documentation.
Parameters
----------
params : ``Params``, required.
A Params object containing a "regularizers" key.
Returns
-------
A RegularizerApplicator containing the specified Regularizers,
or ``None`` if no Regularizers are specified.
"""
if not params:
return None
instantiated_regularizers = []
for parameter_regex, regularizer_params in params:
if isinstance(regularizer_params, str):
regularizer = Regularizer.by_name(regularizer_params)()
else:
regularizer_type = Regularizer.by_name(regularizer_params.pop("type"))
regularizer = regularizer_type(**regularizer_params) # type: ignore
instantiated_regularizers.append((parameter_regex, regularizer))
return RegularizerApplicator(instantiated_regularizers) |
Sanitize turns PyTorch and Numpy types into basic Python types so they
can be serialized into JSON. | def sanitize(x: Any) -> Any: # pylint: disable=invalid-name,too-many-return-statements
"""
Sanitize turns PyTorch and Numpy types into basic Python types so they
can be serialized into JSON.
"""
if isinstance(x, (str, float, int, bool)):
# x is already serializable
return x
elif isinstance(x, torch.Tensor):
# tensor needs to be converted to a list (and moved to cpu if necessary)
return x.cpu().tolist()
elif isinstance(x, numpy.ndarray):
# array needs to be converted to a list
return x.tolist()
elif isinstance(x, numpy.number): # pylint: disable=no-member
# NumPy numbers need to be converted to Python numbers
return x.item()
elif isinstance(x, dict):
# Dicts need their values sanitized
return {key: sanitize(value) for key, value in x.items()}
elif isinstance(x, (spacy.tokens.Token, allennlp.data.Token)):
# Tokens get sanitized to just their text.
return x.text
elif isinstance(x, (list, tuple)):
# Lists and Tuples need their values sanitized
return [sanitize(x_i) for x_i in x]
elif x is None:
return "None"
elif hasattr(x, 'to_json'):
return x.to_json()
else:
raise ValueError(f"Cannot sanitize {x} of type {type(x)}. "
"If this is your own custom class, add a `to_json(self)` method "
"that returns a JSON-like object.") |
List default first if it exists | def list_available(cls) -> List[str]:
"""List default first if it exists"""
keys = list(Registrable._registry[cls].keys())
default = cls.default_implementation
if default is None:
return keys
elif default not in keys:
message = "Default implementation %s is not registered" % default
raise ConfigurationError(message)
else:
return [default] + [k for k in keys if k != default] |
Takes a list and groups it into sublists of size ``count``, using ``default_value`` to pad the
list at the end if the list is not divisable by ``count``.
For example:
>>> group_by_count([1, 2, 3, 4, 5, 6, 7], 3, 0)
[[1, 2, 3], [4, 5, 6], [7, 0, 0]]
This is a short method, but it's complicated and hard to remember as a one-liner, so we just
make a function out of it. | def group_by_count(iterable: List[Any], count: int, default_value: Any) -> List[List[Any]]:
"""
Takes a list and groups it into sublists of size ``count``, using ``default_value`` to pad the
list at the end if the list is not divisable by ``count``.
For example:
>>> group_by_count([1, 2, 3, 4, 5, 6, 7], 3, 0)
[[1, 2, 3], [4, 5, 6], [7, 0, 0]]
This is a short method, but it's complicated and hard to remember as a one-liner, so we just
make a function out of it.
"""
return [list(l) for l in zip_longest(*[iter(iterable)] * count, fillvalue=default_value)] |
Takes an iterator and batches the individual instances into lists of the
specified size. The last list may be smaller if there are instances left over. | def lazy_groups_of(iterator: Iterator[A], group_size: int) -> Iterator[List[A]]:
"""
Takes an iterator and batches the individual instances into lists of the
specified size. The last list may be smaller if there are instances left over.
"""
return iter(lambda: list(islice(iterator, 0, group_size)), []) |
Take a list of objects and pads it to the desired length, returning the padded list. The
original list is not modified.
Parameters
----------
sequence : List
A list of objects to be padded.
desired_length : int
Maximum length of each sequence. Longer sequences are truncated to this length, and
shorter ones are padded to it.
default_value: Callable, default=lambda: 0
Callable that outputs a default value (of any type) to use as padding values. This is
a lambda to avoid using the same object when the default value is more complex, like a
list.
padding_on_right : bool, default=True
When we add padding tokens (or truncate the sequence), should we do it on the right or
the left?
Returns
-------
padded_sequence : List | def pad_sequence_to_length(sequence: List,
desired_length: int,
default_value: Callable[[], Any] = lambda: 0,
padding_on_right: bool = True) -> List:
"""
Take a list of objects and pads it to the desired length, returning the padded list. The
original list is not modified.
Parameters
----------
sequence : List
A list of objects to be padded.
desired_length : int
Maximum length of each sequence. Longer sequences are truncated to this length, and
shorter ones are padded to it.
default_value: Callable, default=lambda: 0
Callable that outputs a default value (of any type) to use as padding values. This is
a lambda to avoid using the same object when the default value is more complex, like a
list.
padding_on_right : bool, default=True
When we add padding tokens (or truncate the sequence), should we do it on the right or
the left?
Returns
-------
padded_sequence : List
"""
# Truncates the sequence to the desired length.
if padding_on_right:
padded_sequence = sequence[:desired_length]
else:
padded_sequence = sequence[-desired_length:]
# Continues to pad with default_value() until we reach the desired length.
for _ in range(desired_length - len(padded_sequence)):
if padding_on_right:
padded_sequence.append(default_value())
else:
padded_sequence.insert(0, default_value())
return padded_sequence |
Returns a new dictionary with noise added to every key in ``dictionary``. The noise is
uniformly distributed within ``noise_param`` percent of the value for every value in the
dictionary. | def add_noise_to_dict_values(dictionary: Dict[A, float], noise_param: float) -> Dict[A, float]:
"""
Returns a new dictionary with noise added to every key in ``dictionary``. The noise is
uniformly distributed within ``noise_param`` percent of the value for every value in the
dictionary.
"""
new_dict = {}
for key, value in dictionary.items():
noise_value = value * noise_param
noise = random.uniform(-noise_value, noise_value)
new_dict[key] = value + noise
return new_dict |
Sets random seeds for reproducible experiments. This may not work as expected
if you use this from within a python project in which you have already imported Pytorch.
If you use the scripts/run_model.py entry point to training models with this library,
your experiments should be reasonably reproducible. If you are using this from your own
project, you will want to call this function before importing Pytorch. Complete determinism
is very difficult to achieve with libraries doing optimized linear algebra due to massively
parallel execution, which is exacerbated by using GPUs.
Parameters
----------
params: Params object or dict, required.
A ``Params`` object or dict holding the json parameters. | def prepare_environment(params: Params):
"""
Sets random seeds for reproducible experiments. This may not work as expected
if you use this from within a python project in which you have already imported Pytorch.
If you use the scripts/run_model.py entry point to training models with this library,
your experiments should be reasonably reproducible. If you are using this from your own
project, you will want to call this function before importing Pytorch. Complete determinism
is very difficult to achieve with libraries doing optimized linear algebra due to massively
parallel execution, which is exacerbated by using GPUs.
Parameters
----------
params: Params object or dict, required.
A ``Params`` object or dict holding the json parameters.
"""
seed = params.pop_int("random_seed", 13370)
numpy_seed = params.pop_int("numpy_seed", 1337)
torch_seed = params.pop_int("pytorch_seed", 133)
if seed is not None:
random.seed(seed)
if numpy_seed is not None:
numpy.random.seed(numpy_seed)
if torch_seed is not None:
torch.manual_seed(torch_seed)
# Seed all GPUs with the same seed if available.
if torch.cuda.is_available():
torch.cuda.manual_seed_all(torch_seed)
log_pytorch_version_info() |
Matches a namespace pattern against a namespace string. For example, ``*tags`` matches
``passage_tags`` and ``question_tags`` and ``tokens`` matches ``tokens`` but not
``stemmed_tokens``. | def namespace_match(pattern: str, namespace: str):
"""
Matches a namespace pattern against a namespace string. For example, ``*tags`` matches
``passage_tags`` and ``question_tags`` and ``tokens`` matches ``tokens`` but not
``stemmed_tokens``.
"""
if pattern[0] == '*' and namespace.endswith(pattern[1:]):
return True
elif pattern == namespace:
return True
return False |
This function configures 3 global logging attributes - streaming stdout and stderr
to a file as well as the terminal, setting the formatting for the python logging
library and setting the interval frequency for the Tqdm progress bar.
Note that this function does not set the logging level, which is set in ``allennlp/run.py``.
Parameters
----------
serialization_dir : ``str``, required.
The directory to stream logs to.
file_friendly_logging : ``bool``, required.
Whether logs should clean the output to prevent carriage returns
(used to update progress bars on a single terminal line). This
option is typically only used if you are running in an environment
without a terminal.
Returns
-------
``logging.FileHandler``
A logging file handler that can later be closed and removed from the global logger. | def prepare_global_logging(serialization_dir: str, file_friendly_logging: bool) -> logging.FileHandler:
"""
This function configures 3 global logging attributes - streaming stdout and stderr
to a file as well as the terminal, setting the formatting for the python logging
library and setting the interval frequency for the Tqdm progress bar.
Note that this function does not set the logging level, which is set in ``allennlp/run.py``.
Parameters
----------
serialization_dir : ``str``, required.
The directory to stream logs to.
file_friendly_logging : ``bool``, required.
Whether logs should clean the output to prevent carriage returns
(used to update progress bars on a single terminal line). This
option is typically only used if you are running in an environment
without a terminal.
Returns
-------
``logging.FileHandler``
A logging file handler that can later be closed and removed from the global logger.
"""
# If we don't have a terminal as stdout,
# force tqdm to be nicer.
if not sys.stdout.isatty():
file_friendly_logging = True
Tqdm.set_slower_interval(file_friendly_logging)
std_out_file = os.path.join(serialization_dir, "stdout.log")
sys.stdout = TeeLogger(std_out_file, # type: ignore
sys.stdout,
file_friendly_logging)
sys.stderr = TeeLogger(os.path.join(serialization_dir, "stderr.log"), # type: ignore
sys.stderr,
file_friendly_logging)
stdout_handler = logging.FileHandler(std_out_file)
stdout_handler.setFormatter(logging.Formatter('%(asctime)s - %(levelname)s - %(name)s - %(message)s'))
logging.getLogger().addHandler(stdout_handler)
return stdout_handler |
This function closes any open file handles and logs set up by `prepare_global_logging`.
Parameters
----------
stdout_handler : ``logging.FileHandler``, required.
The file handler returned from `prepare_global_logging`, attached to the global logger. | def cleanup_global_logging(stdout_handler: logging.FileHandler) -> None:
"""
This function closes any open file handles and logs set up by `prepare_global_logging`.
Parameters
----------
stdout_handler : ``logging.FileHandler``, required.
The file handler returned from `prepare_global_logging`, attached to the global logger.
"""
stdout_handler.close()
logging.getLogger().removeHandler(stdout_handler)
if isinstance(sys.stdout, TeeLogger):
sys.stdout = sys.stdout.cleanup()
if isinstance(sys.stderr, TeeLogger):
sys.stderr = sys.stderr.cleanup() |
In order to avoid loading spacy models a whole bunch of times, we'll save references to them,
keyed by the options we used to create the spacy model, so any particular configuration only
gets loaded once. | def get_spacy_model(spacy_model_name: str, pos_tags: bool, parse: bool, ner: bool) -> SpacyModelType:
"""
In order to avoid loading spacy models a whole bunch of times, we'll save references to them,
keyed by the options we used to create the spacy model, so any particular configuration only
gets loaded once.
"""
options = (spacy_model_name, pos_tags, parse, ner)
if options not in LOADED_SPACY_MODELS:
disable = ['vectors', 'textcat']
if not pos_tags:
disable.append('tagger')
if not parse:
disable.append('parser')
if not ner:
disable.append('ner')
try:
spacy_model = spacy.load(spacy_model_name, disable=disable)
except OSError:
logger.warning(f"Spacy models '{spacy_model_name}' not found. Downloading and installing.")
spacy_download(spacy_model_name)
# NOTE(mattg): The following four lines are a workaround suggested by Ines for spacy
# 2.1.0, which removed the linking that was done in spacy 2.0. importlib doesn't find
# packages that were installed in the same python session, so the way `spacy_download`
# works in 2.1.0 is broken for this use case. These four lines can probably be removed
# at some point in the future, once spacy has figured out a better way to handle this.
# See https://github.com/explosion/spaCy/issues/3435.
from spacy.cli import link
from spacy.util import get_package_path
package_path = get_package_path(spacy_model_name)
link(spacy_model_name, spacy_model_name, model_path=package_path)
spacy_model = spacy.load(spacy_model_name, disable=disable)
LOADED_SPACY_MODELS[options] = spacy_model
return LOADED_SPACY_MODELS[options] |
Import all submodules under the given package.
Primarily useful so that people using AllenNLP as a library
can specify their own custom packages and have their custom
classes get loaded and registered. | def import_submodules(package_name: str) -> None:
"""
Import all submodules under the given package.
Primarily useful so that people using AllenNLP as a library
can specify their own custom packages and have their custom
classes get loaded and registered.
"""
importlib.invalidate_caches()
# For some reason, python doesn't always add this by default to your path, but you pretty much
# always want it when using `--include-package`. And if it's already there, adding it again at
# the end won't hurt anything.
sys.path.append('.')
# Import at top level
module = importlib.import_module(package_name)
path = getattr(module, '__path__', [])
path_string = '' if not path else path[0]
# walk_packages only finds immediate children, so need to recurse.
for module_finder, name, _ in pkgutil.walk_packages(path):
# Sometimes when you import third-party libraries that are on your path,
# `pkgutil.walk_packages` returns those too, so we need to skip them.
if path_string and module_finder.path != path_string:
continue
subpackage = f"{package_name}.{name}"
import_submodules(subpackage) |
Get peak memory usage for this process, as measured by
max-resident-set size:
https://unix.stackexchange.com/questions/30940/getrusage-system-call-what-is-maximum-resident-set-size
Only works on OSX and Linux, returns 0.0 otherwise. | def peak_memory_mb() -> float:
"""
Get peak memory usage for this process, as measured by
max-resident-set size:
https://unix.stackexchange.com/questions/30940/getrusage-system-call-what-is-maximum-resident-set-size
Only works on OSX and Linux, returns 0.0 otherwise.
"""
if resource is None or sys.platform not in ('linux', 'darwin'):
return 0.0
# TODO(joelgrus): For whatever, our pinned version 0.521 of mypy does not like
# next line, but later versions (e.g. 0.530) are fine with it. Once we get that
# figured out, remove the type: ignore.
peak = resource.getrusage(resource.RUSAGE_SELF).ru_maxrss # type: ignore
if sys.platform == 'darwin':
# On OSX the result is in bytes.
return peak / 1_000_000
else:
# On Linux the result is in kilobytes.
return peak / 1_000 |
Get the current GPU memory usage.
Based on https://discuss.pytorch.org/t/access-gpu-memory-usage-in-pytorch/3192/4
Returns
-------
``Dict[int, int]``
Keys are device ids as integers.
Values are memory usage as integers in MB.
Returns an empty ``dict`` if GPUs are not available. | def gpu_memory_mb() -> Dict[int, int]:
"""
Get the current GPU memory usage.
Based on https://discuss.pytorch.org/t/access-gpu-memory-usage-in-pytorch/3192/4
Returns
-------
``Dict[int, int]``
Keys are device ids as integers.
Values are memory usage as integers in MB.
Returns an empty ``dict`` if GPUs are not available.
"""
# pylint: disable=bare-except
try:
result = subprocess.check_output(['nvidia-smi', '--query-gpu=memory.used',
'--format=csv,nounits,noheader'],
encoding='utf-8')
gpu_memory = [int(x) for x in result.strip().split('\n')]
return {gpu: memory for gpu, memory in enumerate(gpu_memory)}
except FileNotFoundError:
# `nvidia-smi` doesn't exist, assume that means no GPU.
return {}
except:
# Catch *all* exceptions, because this memory check is a nice-to-have
# and we'd never want a training run to fail because of it.
logger.exception("unable to check gpu_memory_mb(), continuing")
return {} |
An Iterable may be a list or a generator.
This ensures we get a list without making an unnecessary copy. | def ensure_list(iterable: Iterable[A]) -> List[A]:
"""
An Iterable may be a list or a generator.
This ensures we get a list without making an unnecessary copy.
"""
if isinstance(iterable, list):
return iterable
else:
return list(iterable) |
Takes an action index, updates checklist and returns an updated state. | def update(self, action: torch.Tensor) -> 'ChecklistStatelet':
"""
Takes an action index, updates checklist and returns an updated state.
"""
checklist_addition = (self.terminal_actions == action).float()
new_checklist = self.checklist + checklist_addition
new_checklist_state = ChecklistStatelet(terminal_actions=self.terminal_actions,
checklist_target=self.checklist_target,
checklist_mask=self.checklist_mask,
checklist=new_checklist,
terminal_indices_dict=self.terminal_indices_dict)
return new_checklist_state |
Finds the production rule matching the filter function in the given type's valid action
list, and removes it. If there is more than one matching function, we crash. | def _remove_action_from_type(valid_actions: Dict[str, List[str]],
type_: str,
filter_function: Callable[[str], bool]) -> None:
"""
Finds the production rule matching the filter function in the given type's valid action
list, and removes it. If there is more than one matching function, we crash.
"""
action_list = valid_actions[type_]
matching_action_index = [i for i, action in enumerate(action_list) if filter_function(action)]
assert len(matching_action_index) == 1, "Filter function didn't find one action"
action_list.pop(matching_action_index[0]) |
Parameters
----------
inputs : ``torch.FloatTensor``, required.
A tensor of shape (batch_size, num_timesteps, input_size)
to apply the LSTM over.
batch_lengths : ``List[int]``, required.
A list of length batch_size containing the lengths of the sequences in batch.
initial_state : ``Tuple[torch.Tensor, torch.Tensor]``, optional, (default = None)
A tuple (state, memory) representing the initial hidden state and memory
of the LSTM. The ``state`` has shape (1, batch_size, hidden_size) and the
``memory`` has shape (1, batch_size, cell_size).
Returns
-------
output_accumulator : ``torch.FloatTensor``
The outputs of the LSTM for each timestep. A tensor of shape
(batch_size, max_timesteps, hidden_size) where for a given batch
element, all outputs past the sequence length for that batch are
zero tensors.
final_state : ``Tuple[``torch.FloatTensor, torch.FloatTensor]``
A tuple (state, memory) representing the initial hidden state and memory
of the LSTM. The ``state`` has shape (1, batch_size, hidden_size) and the
``memory`` has shape (1, batch_size, cell_size). | def forward(self, # pylint: disable=arguments-differ
inputs: torch.FloatTensor,
batch_lengths: List[int],
initial_state: Optional[Tuple[torch.Tensor, torch.Tensor]] = None):
"""
Parameters
----------
inputs : ``torch.FloatTensor``, required.
A tensor of shape (batch_size, num_timesteps, input_size)
to apply the LSTM over.
batch_lengths : ``List[int]``, required.
A list of length batch_size containing the lengths of the sequences in batch.
initial_state : ``Tuple[torch.Tensor, torch.Tensor]``, optional, (default = None)
A tuple (state, memory) representing the initial hidden state and memory
of the LSTM. The ``state`` has shape (1, batch_size, hidden_size) and the
``memory`` has shape (1, batch_size, cell_size).
Returns
-------
output_accumulator : ``torch.FloatTensor``
The outputs of the LSTM for each timestep. A tensor of shape
(batch_size, max_timesteps, hidden_size) where for a given batch
element, all outputs past the sequence length for that batch are
zero tensors.
final_state : ``Tuple[``torch.FloatTensor, torch.FloatTensor]``
A tuple (state, memory) representing the initial hidden state and memory
of the LSTM. The ``state`` has shape (1, batch_size, hidden_size) and the
``memory`` has shape (1, batch_size, cell_size).
"""
batch_size = inputs.size()[0]
total_timesteps = inputs.size()[1]
output_accumulator = inputs.new_zeros(batch_size, total_timesteps, self.hidden_size)
if initial_state is None:
full_batch_previous_memory = inputs.new_zeros(batch_size, self.cell_size)
full_batch_previous_state = inputs.new_zeros(batch_size, self.hidden_size)
else:
full_batch_previous_state = initial_state[0].squeeze(0)
full_batch_previous_memory = initial_state[1].squeeze(0)
current_length_index = batch_size - 1 if self.go_forward else 0
if self.recurrent_dropout_probability > 0.0 and self.training:
dropout_mask = get_dropout_mask(self.recurrent_dropout_probability,
full_batch_previous_state)
else:
dropout_mask = None
for timestep in range(total_timesteps):
# The index depends on which end we start.
index = timestep if self.go_forward else total_timesteps - timestep - 1
# What we are doing here is finding the index into the batch dimension
# which we need to use for this timestep, because the sequences have
# variable length, so once the index is greater than the length of this
# particular batch sequence, we no longer need to do the computation for
# this sequence. The key thing to recognise here is that the batch inputs
# must be _ordered_ by length from longest (first in batch) to shortest
# (last) so initially, we are going forwards with every sequence and as we
# pass the index at which the shortest elements of the batch finish,
# we stop picking them up for the computation.
if self.go_forward:
while batch_lengths[current_length_index] <= index:
current_length_index -= 1
# If we're going backwards, we are _picking up_ more indices.
else:
# First conditional: Are we already at the maximum number of elements in the batch?
# Second conditional: Does the next shortest sequence beyond the current batch
# index require computation use this timestep?
while current_length_index < (len(batch_lengths) - 1) and \
batch_lengths[current_length_index + 1] > index:
current_length_index += 1
# Actually get the slices of the batch which we
# need for the computation at this timestep.
# shape (batch_size, cell_size)
previous_memory = full_batch_previous_memory[0: current_length_index + 1].clone()
# Shape (batch_size, hidden_size)
previous_state = full_batch_previous_state[0: current_length_index + 1].clone()
# Shape (batch_size, input_size)
timestep_input = inputs[0: current_length_index + 1, index]
# Do the projections for all the gates all at once.
# Both have shape (batch_size, 4 * cell_size)
projected_input = self.input_linearity(timestep_input)
projected_state = self.state_linearity(previous_state)
# Main LSTM equations using relevant chunks of the big linear
# projections of the hidden state and inputs.
input_gate = torch.sigmoid(projected_input[:, (0 * self.cell_size):(1 * self.cell_size)] +
projected_state[:, (0 * self.cell_size):(1 * self.cell_size)])
forget_gate = torch.sigmoid(projected_input[:, (1 * self.cell_size):(2 * self.cell_size)] +
projected_state[:, (1 * self.cell_size):(2 * self.cell_size)])
memory_init = torch.tanh(projected_input[:, (2 * self.cell_size):(3 * self.cell_size)] +
projected_state[:, (2 * self.cell_size):(3 * self.cell_size)])
output_gate = torch.sigmoid(projected_input[:, (3 * self.cell_size):(4 * self.cell_size)] +
projected_state[:, (3 * self.cell_size):(4 * self.cell_size)])
memory = input_gate * memory_init + forget_gate * previous_memory
# Here is the non-standard part of this LSTM cell; first, we clip the
# memory cell, then we project the output of the timestep to a smaller size
# and again clip it.
if self.memory_cell_clip_value:
# pylint: disable=invalid-unary-operand-type
memory = torch.clamp(memory, -self.memory_cell_clip_value, self.memory_cell_clip_value)
# shape (current_length_index, cell_size)
pre_projection_timestep_output = output_gate * torch.tanh(memory)
# shape (current_length_index, hidden_size)
timestep_output = self.state_projection(pre_projection_timestep_output)
if self.state_projection_clip_value:
# pylint: disable=invalid-unary-operand-type
timestep_output = torch.clamp(timestep_output,
-self.state_projection_clip_value,
self.state_projection_clip_value)
# Only do dropout if the dropout prob is > 0.0 and we are in training mode.
if dropout_mask is not None:
timestep_output = timestep_output * dropout_mask[0: current_length_index + 1]
# We've been doing computation with less than the full batch, so here we create a new
# variable for the the whole batch at this timestep and insert the result for the
# relevant elements of the batch into it.
full_batch_previous_memory = full_batch_previous_memory.clone()
full_batch_previous_state = full_batch_previous_state.clone()
full_batch_previous_memory[0:current_length_index + 1] = memory
full_batch_previous_state[0:current_length_index + 1] = timestep_output
output_accumulator[0:current_length_index + 1, index] = timestep_output
# Mimic the pytorch API by returning state in the following shape:
# (num_layers * num_directions, batch_size, ...). As this
# LSTM cell cannot be stacked, the first dimension here is just 1.
final_state = (full_batch_previous_state.unsqueeze(0),
full_batch_previous_memory.unsqueeze(0))
return output_accumulator, final_state |
Determine the URL corresponding to Python object
This code is from
https://github.com/numpy/numpy/blob/master/doc/source/conf.py#L290
and https://github.com/Lasagne/Lasagne/pull/262 | def linkcode_resolve(domain, info):
"""
Determine the URL corresponding to Python object
This code is from
https://github.com/numpy/numpy/blob/master/doc/source/conf.py#L290
and https://github.com/Lasagne/Lasagne/pull/262
"""
if domain != 'py':
return None
modname = info['module']
fullname = info['fullname']
submod = sys.modules.get(modname)
if submod is None:
return None
obj = submod
for part in fullname.split('.'):
try:
obj = getattr(obj, part)
except:
return None
try:
fn = inspect.getsourcefile(obj)
except:
fn = None
if not fn:
return None
try:
source, lineno = inspect.getsourcelines(obj)
except:
lineno = None
if lineno:
linespec = "#L%d-L%d" % (lineno, lineno + len(source) - 1)
else:
linespec = ""
filename = info['module'].replace('.', '/')
return "http://github.com/allenai/allennlp/blob/master/%s.py%s" % (filename, linespec) |
This method returns the indices of each entity's neighbors. A tensor
is accepted as a parameter for copying purposes.
Parameters
----------
worlds : ``List[WikiTablesWorld]``
num_entities : ``int``
tensor : ``torch.Tensor``
Used for copying the constructed list onto the right device.
Returns
-------
A ``torch.LongTensor`` with shape ``(batch_size, num_entities, num_neighbors)``. It is padded
with -1 instead of 0, since 0 is a valid neighbor index. | def _get_neighbor_indices(worlds: List[WikiTablesWorld],
num_entities: int,
tensor: torch.Tensor) -> torch.LongTensor:
"""
This method returns the indices of each entity's neighbors. A tensor
is accepted as a parameter for copying purposes.
Parameters
----------
worlds : ``List[WikiTablesWorld]``
num_entities : ``int``
tensor : ``torch.Tensor``
Used for copying the constructed list onto the right device.
Returns
-------
A ``torch.LongTensor`` with shape ``(batch_size, num_entities, num_neighbors)``. It is padded
with -1 instead of 0, since 0 is a valid neighbor index.
"""
num_neighbors = 0
for world in worlds:
for entity in world.table_graph.entities:
if len(world.table_graph.neighbors[entity]) > num_neighbors:
num_neighbors = len(world.table_graph.neighbors[entity])
batch_neighbors = []
for world in worlds:
# Each batch instance has its own world, which has a corresponding table.
entities = world.table_graph.entities
entity2index = {entity: i for i, entity in enumerate(entities)}
entity2neighbors = world.table_graph.neighbors
neighbor_indexes = []
for entity in entities:
entity_neighbors = [entity2index[n] for n in entity2neighbors[entity]]
# Pad with -1 instead of 0, since 0 represents a neighbor index.
padded = pad_sequence_to_length(entity_neighbors, num_neighbors, lambda: -1)
neighbor_indexes.append(padded)
neighbor_indexes = pad_sequence_to_length(neighbor_indexes,
num_entities,
lambda: [-1] * num_neighbors)
batch_neighbors.append(neighbor_indexes)
return tensor.new_tensor(batch_neighbors, dtype=torch.long) |
Encodes the question and table, computes a linking between the two, and constructs an
initial RnnStatelet and LambdaGrammarStatelet for each batch instance to pass to the
decoder.
We take ``outputs`` as a parameter here and `modify` it, adding things that we want to
visualize in a demo. | def _get_initial_rnn_and_grammar_state(self,
question: Dict[str, torch.LongTensor],
table: Dict[str, torch.LongTensor],
world: List[WikiTablesWorld],
actions: List[List[ProductionRule]],
outputs: Dict[str, Any]) -> Tuple[List[RnnStatelet],
List[LambdaGrammarStatelet]]:
"""
Encodes the question and table, computes a linking between the two, and constructs an
initial RnnStatelet and LambdaGrammarStatelet for each batch instance to pass to the
decoder.
We take ``outputs`` as a parameter here and `modify` it, adding things that we want to
visualize in a demo.
"""
table_text = table['text']
# (batch_size, question_length, embedding_dim)
embedded_question = self._question_embedder(question)
question_mask = util.get_text_field_mask(question).float()
# (batch_size, num_entities, num_entity_tokens, embedding_dim)
embedded_table = self._question_embedder(table_text, num_wrapping_dims=1)
table_mask = util.get_text_field_mask(table_text, num_wrapping_dims=1).float()
batch_size, num_entities, num_entity_tokens, _ = embedded_table.size()
num_question_tokens = embedded_question.size(1)
# (batch_size, num_entities, embedding_dim)
encoded_table = self._entity_encoder(embedded_table, table_mask)
# (batch_size, num_entities, num_neighbors)
neighbor_indices = self._get_neighbor_indices(world, num_entities, encoded_table)
# Neighbor_indices is padded with -1 since 0 is a potential neighbor index.
# Thus, the absolute value needs to be taken in the index_select, and 1 needs to
# be added for the mask since that method expects 0 for padding.
# (batch_size, num_entities, num_neighbors, embedding_dim)
embedded_neighbors = util.batched_index_select(encoded_table, torch.abs(neighbor_indices))
neighbor_mask = util.get_text_field_mask({'ignored': neighbor_indices + 1},
num_wrapping_dims=1).float()
# Encoder initialized to easily obtain a masked average.
neighbor_encoder = TimeDistributed(BagOfEmbeddingsEncoder(self._embedding_dim, averaged=True))
# (batch_size, num_entities, embedding_dim)
embedded_neighbors = neighbor_encoder(embedded_neighbors, neighbor_mask)
# entity_types: tensor with shape (batch_size, num_entities), where each entry is the
# entity's type id.
# entity_type_dict: Dict[int, int], mapping flattened_entity_index -> type_index
# These encode the same information, but for efficiency reasons later it's nice
# to have one version as a tensor and one that's accessible on the cpu.
entity_types, entity_type_dict = self._get_type_vector(world, num_entities, encoded_table)
entity_type_embeddings = self._entity_type_encoder_embedding(entity_types)
projected_neighbor_embeddings = self._neighbor_params(embedded_neighbors.float())
# (batch_size, num_entities, embedding_dim)
entity_embeddings = torch.tanh(entity_type_embeddings + projected_neighbor_embeddings)
# Compute entity and question word similarity. We tried using cosine distance here, but
# because this similarity is the main mechanism that the model can use to push apart logit
# scores for certain actions (like "n -> 1" and "n -> -1"), this needs to have a larger
# output range than [-1, 1].
question_entity_similarity = torch.bmm(embedded_table.view(batch_size,
num_entities * num_entity_tokens,
self._embedding_dim),
torch.transpose(embedded_question, 1, 2))
question_entity_similarity = question_entity_similarity.view(batch_size,
num_entities,
num_entity_tokens,
num_question_tokens)
# (batch_size, num_entities, num_question_tokens)
question_entity_similarity_max_score, _ = torch.max(question_entity_similarity, 2)
# (batch_size, num_entities, num_question_tokens, num_features)
linking_features = table['linking']
linking_scores = question_entity_similarity_max_score
if self._use_neighbor_similarity_for_linking:
# The linking score is computed as a linear projection of two terms. The first is the
# maximum similarity score over the entity's words and the question token. The second
# is the maximum similarity over the words in the entity's neighbors and the question
# token.
#
# The second term, projected_question_neighbor_similarity, is useful when a column
# needs to be selected. For example, the question token might have no similarity with
# the column name, but is similar with the cells in the column.
#
# Note that projected_question_neighbor_similarity is intended to capture the same
# information as the related_column feature.
#
# Also note that this block needs to be _before_ the `linking_params` block, because
# we're overwriting `linking_scores`, not adding to it.
# (batch_size, num_entities, num_neighbors, num_question_tokens)
question_neighbor_similarity = util.batched_index_select(question_entity_similarity_max_score,
torch.abs(neighbor_indices))
# (batch_size, num_entities, num_question_tokens)
question_neighbor_similarity_max_score, _ = torch.max(question_neighbor_similarity, 2)
projected_question_entity_similarity = self._question_entity_params(
question_entity_similarity_max_score.unsqueeze(-1)).squeeze(-1)
projected_question_neighbor_similarity = self._question_neighbor_params(
question_neighbor_similarity_max_score.unsqueeze(-1)).squeeze(-1)
linking_scores = projected_question_entity_similarity + projected_question_neighbor_similarity
feature_scores = None
if self._linking_params is not None:
feature_scores = self._linking_params(linking_features).squeeze(3)
linking_scores = linking_scores + feature_scores
# (batch_size, num_question_tokens, num_entities)
linking_probabilities = self._get_linking_probabilities(world, linking_scores.transpose(1, 2),
question_mask, entity_type_dict)
# (batch_size, num_question_tokens, embedding_dim)
link_embedding = util.weighted_sum(entity_embeddings, linking_probabilities)
encoder_input = torch.cat([link_embedding, embedded_question], 2)
# (batch_size, question_length, encoder_output_dim)
encoder_outputs = self._dropout(self._encoder(encoder_input, question_mask))
# This will be our initial hidden state and memory cell for the decoder LSTM.
final_encoder_output = util.get_final_encoder_states(encoder_outputs,
question_mask,
self._encoder.is_bidirectional())
memory_cell = encoder_outputs.new_zeros(batch_size, self._encoder.get_output_dim())
# To make grouping states together in the decoder easier, we convert the batch dimension in
# all of our tensors into an outer list. For instance, the encoder outputs have shape
# `(batch_size, question_length, encoder_output_dim)`. We need to convert this into a list
# of `batch_size` tensors, each of shape `(question_length, encoder_output_dim)`. Then we
# won't have to do any index selects, or anything, we'll just do some `torch.cat()`s.
encoder_output_list = [encoder_outputs[i] for i in range(batch_size)]
question_mask_list = [question_mask[i] for i in range(batch_size)]
initial_rnn_state = []
for i in range(batch_size):
initial_rnn_state.append(RnnStatelet(final_encoder_output[i],
memory_cell[i],
self._first_action_embedding,
self._first_attended_question,
encoder_output_list,
question_mask_list))
initial_grammar_state = [self._create_grammar_state(world[i],
actions[i],
linking_scores[i],
entity_types[i])
for i in range(batch_size)]
if not self.training:
# We add a few things to the outputs that will be returned from `forward` at evaluation
# time, for visualization in a demo.
outputs['linking_scores'] = linking_scores
if feature_scores is not None:
outputs['feature_scores'] = feature_scores
outputs['similarity_scores'] = question_entity_similarity_max_score
return initial_rnn_state, initial_grammar_state |
Produces a tensor with shape ``(batch_size, num_entities)`` that encodes each entity's
type. In addition, a map from a flattened entity index to type is returned to combine
entity type operations into one method.
Parameters
----------
worlds : ``List[WikiTablesWorld]``
num_entities : ``int``
tensor : ``torch.Tensor``
Used for copying the constructed list onto the right device.
Returns
-------
A ``torch.LongTensor`` with shape ``(batch_size, num_entities)``.
entity_types : ``Dict[int, int]``
This is a mapping from ((batch_index * num_entities) + entity_index) to entity type id. | def _get_type_vector(worlds: List[WikiTablesWorld],
num_entities: int,
tensor: torch.Tensor) -> Tuple[torch.LongTensor, Dict[int, int]]:
"""
Produces a tensor with shape ``(batch_size, num_entities)`` that encodes each entity's
type. In addition, a map from a flattened entity index to type is returned to combine
entity type operations into one method.
Parameters
----------
worlds : ``List[WikiTablesWorld]``
num_entities : ``int``
tensor : ``torch.Tensor``
Used for copying the constructed list onto the right device.
Returns
-------
A ``torch.LongTensor`` with shape ``(batch_size, num_entities)``.
entity_types : ``Dict[int, int]``
This is a mapping from ((batch_index * num_entities) + entity_index) to entity type id.
"""
entity_types = {}
batch_types = []
for batch_index, world in enumerate(worlds):
types = []
for entity_index, entity in enumerate(world.table_graph.entities):
# We need numbers to be first, then cells, then parts, then row, because our
# entities are going to be sorted. We do a split by type and then a merge later,
# and it relies on this sorting.
if entity.startswith('fb:cell'):
entity_type = 1
elif entity.startswith('fb:part'):
entity_type = 2
elif entity.startswith('fb:row'):
entity_type = 3
else:
entity_type = 0
types.append(entity_type)
# For easier lookups later, we're actually using a _flattened_ version
# of (batch_index, entity_index) for the key, because this is how the
# linking scores are stored.
flattened_entity_index = batch_index * num_entities + entity_index
entity_types[flattened_entity_index] = entity_type
padded = pad_sequence_to_length(types, num_entities, lambda: 0)
batch_types.append(padded)
return tensor.new_tensor(batch_types, dtype=torch.long), entity_types |
Produces the probability of an entity given a question word and type. The logic below
separates the entities by type since the softmax normalization term sums over entities
of a single type.
Parameters
----------
worlds : ``List[WikiTablesWorld]``
linking_scores : ``torch.FloatTensor``
Has shape (batch_size, num_question_tokens, num_entities).
question_mask: ``torch.LongTensor``
Has shape (batch_size, num_question_tokens).
entity_type_dict : ``Dict[int, int]``
This is a mapping from ((batch_index * num_entities) + entity_index) to entity type id.
Returns
-------
batch_probabilities : ``torch.FloatTensor``
Has shape ``(batch_size, num_question_tokens, num_entities)``.
Contains all the probabilities for an entity given a question word. | def _get_linking_probabilities(self,
worlds: List[WikiTablesWorld],
linking_scores: torch.FloatTensor,
question_mask: torch.LongTensor,
entity_type_dict: Dict[int, int]) -> torch.FloatTensor:
"""
Produces the probability of an entity given a question word and type. The logic below
separates the entities by type since the softmax normalization term sums over entities
of a single type.
Parameters
----------
worlds : ``List[WikiTablesWorld]``
linking_scores : ``torch.FloatTensor``
Has shape (batch_size, num_question_tokens, num_entities).
question_mask: ``torch.LongTensor``
Has shape (batch_size, num_question_tokens).
entity_type_dict : ``Dict[int, int]``
This is a mapping from ((batch_index * num_entities) + entity_index) to entity type id.
Returns
-------
batch_probabilities : ``torch.FloatTensor``
Has shape ``(batch_size, num_question_tokens, num_entities)``.
Contains all the probabilities for an entity given a question word.
"""
_, num_question_tokens, num_entities = linking_scores.size()
batch_probabilities = []
for batch_index, world in enumerate(worlds):
all_probabilities = []
num_entities_in_instance = 0
# NOTE: The way that we're doing this here relies on the fact that entities are
# implicitly sorted by their types when we sort them by name, and that numbers come
# before "fb:cell", and "fb:cell" comes before "fb:row". This is not a great
# assumption, and could easily break later, but it should work for now.
for type_index in range(self._num_entity_types):
# This index of 0 is for the null entity for each type, representing the case where a
# word doesn't link to any entity.
entity_indices = [0]
entities = world.table_graph.entities
for entity_index, _ in enumerate(entities):
if entity_type_dict[batch_index * num_entities + entity_index] == type_index:
entity_indices.append(entity_index)
if len(entity_indices) == 1:
# No entities of this type; move along...
continue
# We're subtracting one here because of the null entity we added above.
num_entities_in_instance += len(entity_indices) - 1
# We separate the scores by type, since normalization is done per type. There's an
# extra "null" entity per type, also, so we have `num_entities_per_type + 1`. We're
# selecting from a (num_question_tokens, num_entities) linking tensor on _dimension 1_,
# so we get back something of shape (num_question_tokens,) for each index we're
# selecting. All of the selected indices together then make a tensor of shape
# (num_question_tokens, num_entities_per_type + 1).
indices = linking_scores.new_tensor(entity_indices, dtype=torch.long)
entity_scores = linking_scores[batch_index].index_select(1, indices)
# We used index 0 for the null entity, so this will actually have some values in it.
# But we want the null entity's score to be 0, so we set that here.
entity_scores[:, 0] = 0
# No need for a mask here, as this is done per batch instance, with no padding.
type_probabilities = torch.nn.functional.softmax(entity_scores, dim=1)
all_probabilities.append(type_probabilities[:, 1:])
# We need to add padding here if we don't have the right number of entities.
if num_entities_in_instance != num_entities:
zeros = linking_scores.new_zeros(num_question_tokens,
num_entities - num_entities_in_instance)
all_probabilities.append(zeros)
# (num_question_tokens, num_entities)
probabilities = torch.cat(all_probabilities, dim=1)
batch_probabilities.append(probabilities)
batch_probabilities = torch.stack(batch_probabilities, dim=0)
return batch_probabilities * question_mask.unsqueeze(-1).float() |
We track three metrics here:
1. dpd_acc, which is the percentage of the time that our best output action sequence is
in the set of action sequences provided by DPD. This is an easy-to-compute lower bound
on denotation accuracy for the set of examples where we actually have DPD output. We
only score dpd_acc on that subset.
2. denotation_acc, which is the percentage of examples where we get the correct
denotation. This is the typical "accuracy" metric, and it is what you should usually
report in an experimental result. You need to be careful, though, that you're
computing this on the full data, and not just the subset that has DPD output (make sure
you pass "keep_if_no_dpd=True" to the dataset reader, which we do for validation data,
but not training data).
3. lf_percent, which is the percentage of time that decoding actually produces a
finished logical form. We might not produce a valid logical form if the decoder gets
into a repetitive loop, or we're trying to produce a super long logical form and run
out of time steps, or something. | def get_metrics(self, reset: bool = False) -> Dict[str, float]:
"""
We track three metrics here:
1. dpd_acc, which is the percentage of the time that our best output action sequence is
in the set of action sequences provided by DPD. This is an easy-to-compute lower bound
on denotation accuracy for the set of examples where we actually have DPD output. We
only score dpd_acc on that subset.
2. denotation_acc, which is the percentage of examples where we get the correct
denotation. This is the typical "accuracy" metric, and it is what you should usually
report in an experimental result. You need to be careful, though, that you're
computing this on the full data, and not just the subset that has DPD output (make sure
you pass "keep_if_no_dpd=True" to the dataset reader, which we do for validation data,
but not training data).
3. lf_percent, which is the percentage of time that decoding actually produces a
finished logical form. We might not produce a valid logical form if the decoder gets
into a repetitive loop, or we're trying to produce a super long logical form and run
out of time steps, or something.
"""
return {
'dpd_acc': self._action_sequence_accuracy.get_metric(reset),
'denotation_acc': self._denotation_accuracy.get_metric(reset),
'lf_percent': self._has_logical_form.get_metric(reset),
} |
This method creates the LambdaGrammarStatelet object that's used for decoding. Part of
creating that is creating the `valid_actions` dictionary, which contains embedded
representations of all of the valid actions. So, we create that here as well.
The way we represent the valid expansions is a little complicated: we use a
dictionary of `action types`, where the key is the action type (like "global", "linked", or
whatever your model is expecting), and the value is a tuple representing all actions of
that type. The tuple is (input tensor, output tensor, action id). The input tensor has
the representation that is used when `selecting` actions, for all actions of this type.
The output tensor has the representation that is used when feeding the action to the next
step of the decoder (this could just be the same as the input tensor). The action ids are
a list of indices into the main action list for each batch instance.
The inputs to this method are for a `single instance in the batch`; none of the tensors we
create here are batched. We grab the global action ids from the input
``ProductionRules``, and we use those to embed the valid actions for every
non-terminal type. We use the input ``linking_scores`` for non-global actions.
Parameters
----------
world : ``WikiTablesWorld``
From the input to ``forward`` for a single batch instance.
possible_actions : ``List[ProductionRule]``
From the input to ``forward`` for a single batch instance.
linking_scores : ``torch.Tensor``
Assumed to have shape ``(num_entities, num_question_tokens)`` (i.e., there is no batch
dimension).
entity_types : ``torch.Tensor``
Assumed to have shape ``(num_entities,)`` (i.e., there is no batch dimension). | def _create_grammar_state(self,
world: WikiTablesWorld,
possible_actions: List[ProductionRule],
linking_scores: torch.Tensor,
entity_types: torch.Tensor) -> LambdaGrammarStatelet:
"""
This method creates the LambdaGrammarStatelet object that's used for decoding. Part of
creating that is creating the `valid_actions` dictionary, which contains embedded
representations of all of the valid actions. So, we create that here as well.
The way we represent the valid expansions is a little complicated: we use a
dictionary of `action types`, where the key is the action type (like "global", "linked", or
whatever your model is expecting), and the value is a tuple representing all actions of
that type. The tuple is (input tensor, output tensor, action id). The input tensor has
the representation that is used when `selecting` actions, for all actions of this type.
The output tensor has the representation that is used when feeding the action to the next
step of the decoder (this could just be the same as the input tensor). The action ids are
a list of indices into the main action list for each batch instance.
The inputs to this method are for a `single instance in the batch`; none of the tensors we
create here are batched. We grab the global action ids from the input
``ProductionRules``, and we use those to embed the valid actions for every
non-terminal type. We use the input ``linking_scores`` for non-global actions.
Parameters
----------
world : ``WikiTablesWorld``
From the input to ``forward`` for a single batch instance.
possible_actions : ``List[ProductionRule]``
From the input to ``forward`` for a single batch instance.
linking_scores : ``torch.Tensor``
Assumed to have shape ``(num_entities, num_question_tokens)`` (i.e., there is no batch
dimension).
entity_types : ``torch.Tensor``
Assumed to have shape ``(num_entities,)`` (i.e., there is no batch dimension).
"""
# TODO(mattg): Move the "valid_actions" construction to another method.
action_map = {}
for action_index, action in enumerate(possible_actions):
action_string = action[0]
action_map[action_string] = action_index
entity_map = {}
for entity_index, entity in enumerate(world.table_graph.entities):
entity_map[entity] = entity_index
valid_actions = world.get_valid_actions()
translated_valid_actions: Dict[str, Dict[str, Tuple[torch.Tensor, torch.Tensor, List[int]]]] = {}
for key, action_strings in valid_actions.items():
translated_valid_actions[key] = {}
# `key` here is a non-terminal from the grammar, and `action_strings` are all the valid
# productions of that non-terminal. We'll first split those productions by global vs.
# linked action.
action_indices = [action_map[action_string] for action_string in action_strings]
production_rule_arrays = [(possible_actions[index], index) for index in action_indices]
global_actions = []
linked_actions = []
for production_rule_array, action_index in production_rule_arrays:
if production_rule_array[1]:
global_actions.append((production_rule_array[2], action_index))
else:
linked_actions.append((production_rule_array[0], action_index))
# Then we get the embedded representations of the global actions.
global_action_tensors, global_action_ids = zip(*global_actions)
global_action_tensor = torch.cat(global_action_tensors, dim=0)
global_input_embeddings = self._action_embedder(global_action_tensor)
if self._add_action_bias:
global_action_biases = self._action_biases(global_action_tensor)
global_input_embeddings = torch.cat([global_input_embeddings, global_action_biases], dim=-1)
global_output_embeddings = self._output_action_embedder(global_action_tensor)
translated_valid_actions[key]['global'] = (global_input_embeddings,
global_output_embeddings,
list(global_action_ids))
# Then the representations of the linked actions.
if linked_actions:
linked_rules, linked_action_ids = zip(*linked_actions)
entities = [rule.split(' -> ')[1] for rule in linked_rules]
entity_ids = [entity_map[entity] for entity in entities]
# (num_linked_actions, num_question_tokens)
entity_linking_scores = linking_scores[entity_ids]
# (num_linked_actions,)
entity_type_tensor = entity_types[entity_ids]
# (num_linked_actions, entity_type_embedding_dim)
entity_type_embeddings = self._entity_type_decoder_embedding(entity_type_tensor)
translated_valid_actions[key]['linked'] = (entity_linking_scores,
entity_type_embeddings,
list(linked_action_ids))
# Lastly, we need to also create embedded representations of context-specific actions. In
# this case, those are only variable productions, like "r -> x". Note that our language
# only permits one lambda at a time, so we don't need to worry about how nested lambdas
# might impact this.
context_actions = {}
for action_id, action in enumerate(possible_actions):
if action[0].endswith(" -> x"):
input_embedding = self._action_embedder(action[2])
if self._add_action_bias:
input_bias = self._action_biases(action[2])
input_embedding = torch.cat([input_embedding, input_bias], dim=-1)
output_embedding = self._output_action_embedder(action[2])
context_actions[action[0]] = (input_embedding, output_embedding, action_id)
return LambdaGrammarStatelet([START_SYMBOL],
{},
translated_valid_actions,
context_actions,
type_declaration.is_nonterminal) |
Does common things for validation time: computing logical form accuracy (which is expensive
and unnecessary during training), adding visualization info to the output dictionary, etc.
This doesn't return anything; instead it `modifies` the given ``outputs`` dictionary, and
calls metrics on ``self``. | def _compute_validation_outputs(self,
actions: List[List[ProductionRule]],
best_final_states: Mapping[int, Sequence[GrammarBasedState]],
world: List[WikiTablesWorld],
example_lisp_string: List[str],
metadata: List[Dict[str, Any]],
outputs: Dict[str, Any]) -> None:
"""
Does common things for validation time: computing logical form accuracy (which is expensive
and unnecessary during training), adding visualization info to the output dictionary, etc.
This doesn't return anything; instead it `modifies` the given ``outputs`` dictionary, and
calls metrics on ``self``.
"""
batch_size = len(actions)
action_mapping = {}
for batch_index, batch_actions in enumerate(actions):
for action_index, action in enumerate(batch_actions):
action_mapping[(batch_index, action_index)] = action[0]
outputs['action_mapping'] = action_mapping
outputs['best_action_sequence'] = []
outputs['debug_info'] = []
outputs['entities'] = []
outputs['logical_form'] = []
for i in range(batch_size):
# Decoding may not have terminated with any completed logical forms, if `num_steps`
# isn't long enough (or if the model is not trained enough and gets into an
# infinite action loop).
if i in best_final_states:
best_action_indices = best_final_states[i][0].action_history[0]
action_strings = [action_mapping[(i, action_index)] for action_index in best_action_indices]
try:
logical_form = world[i].get_logical_form(action_strings, add_var_function=False)
self._has_logical_form(1.0)
except ParsingError:
self._has_logical_form(0.0)
logical_form = 'Error producing logical form'
if example_lisp_string:
denotation_correct = self._executor.evaluate_logical_form(logical_form,
example_lisp_string[i])
self._denotation_accuracy(1.0 if denotation_correct else 0.0)
outputs['best_action_sequence'].append(action_strings)
outputs['logical_form'].append(logical_form)
outputs['debug_info'].append(best_final_states[i][0].debug_info[0]) # type: ignore
outputs['entities'].append(world[i].table_graph.entities)
else:
outputs['logical_form'].append('')
self._has_logical_form(0.0)
self._denotation_accuracy(0.0)
if metadata is not None:
outputs["question_tokens"] = [x["question_tokens"] for x in metadata]
outputs["original_table"] = [x["original_table"] for x in metadata] |
This method overrides ``Model.decode``, which gets called after ``Model.forward``, at test
time, to finalize predictions. This is (confusingly) a separate notion from the "decoder"
in "encoder/decoder", where that decoder logic lives in the ``TransitionFunction``.
This method trims the output predictions to the first end symbol, replaces indices with
corresponding tokens, and adds a field called ``predicted_tokens`` to the ``output_dict``. | def decode(self, output_dict: Dict[str, torch.Tensor]) -> Dict[str, torch.Tensor]:
"""
This method overrides ``Model.decode``, which gets called after ``Model.forward``, at test
time, to finalize predictions. This is (confusingly) a separate notion from the "decoder"
in "encoder/decoder", where that decoder logic lives in the ``TransitionFunction``.
This method trims the output predictions to the first end symbol, replaces indices with
corresponding tokens, and adds a field called ``predicted_tokens`` to the ``output_dict``.
"""
action_mapping = output_dict['action_mapping']
best_actions = output_dict["best_action_sequence"]
debug_infos = output_dict['debug_info']
batch_action_info = []
for batch_index, (predicted_actions, debug_info) in enumerate(zip(best_actions, debug_infos)):
instance_action_info = []
for predicted_action, action_debug_info in zip(predicted_actions, debug_info):
action_info = {}
action_info['predicted_action'] = predicted_action
considered_actions = action_debug_info['considered_actions']
probabilities = action_debug_info['probabilities']
actions = []
for action, probability in zip(considered_actions, probabilities):
if action != -1:
actions.append((action_mapping[(batch_index, action)], probability))
actions.sort()
considered_actions, probabilities = zip(*actions)
action_info['considered_actions'] = considered_actions
action_info['action_probabilities'] = probabilities
action_info['question_attention'] = action_debug_info.get('question_attention', [])
instance_action_info.append(action_info)
batch_action_info.append(instance_action_info)
output_dict["predicted_actions"] = batch_action_info
return output_dict |
Gets the logits of desired terminal actions yet to be produced by the decoder, and
returns them for the decoder to add to the prior action logits, biasing the model towards
predicting missing linked actions. | def _get_linked_logits_addition(checklist_state: ChecklistStatelet,
action_ids: List[int],
action_logits: torch.Tensor) -> torch.Tensor:
"""
Gets the logits of desired terminal actions yet to be produced by the decoder, and
returns them for the decoder to add to the prior action logits, biasing the model towards
predicting missing linked actions.
"""
# Our basic approach here will be to figure out which actions we want to bias, by doing
# some fancy indexing work, then multiply the action embeddings by a mask for those
# actions, and return the sum of the result.
# Shape: (num_terminal_actions, 1). This is 1 if we still want to predict something on the
# checklist, and 0 otherwise.
checklist_balance = checklist_state.get_balance().clamp(min=0)
# (num_terminal_actions, 1)
actions_in_agenda = checklist_state.terminal_actions
# (1, num_current_actions)
action_id_tensor = checklist_balance.new(action_ids).long().unsqueeze(0)
# Shape: (num_terminal_actions, num_current_actions). Will have a value of 1 if the
# terminal action i is our current action j, and a value of 0 otherwise. Because both sets
# of actions are free of duplicates, there will be at most one non-zero value per current
# action, and per terminal action.
current_agenda_actions = (actions_in_agenda == action_id_tensor).float()
# Shape: (num_current_actions,). With the inner multiplication, we remove any current
# agenda actions that are not in our checklist balance, then we sum over the terminal
# action dimension, which will have a sum of at most one. So this will be a 0/1 tensor,
# where a 1 means to encourage the current action in that position.
actions_to_encourage = torch.sum(current_agenda_actions * checklist_balance, dim=0)
# Shape: (num_current_actions,). This is the sum of the action embeddings that we want
# the model to prefer.
logit_addition = action_logits * actions_to_encourage
return logit_addition |
Given a query (which is typically the decoder hidden state), compute an attention over the
output of the question encoder, and return a weighted sum of the question representations
given this attention. We also return the attention weights themselves.
This is a simple computation, but we have it as a separate method so that the ``forward``
method on the main parser module can call it on the initial hidden state, to simplify the
logic in ``take_step``. | def attend_on_question(self,
query: torch.Tensor,
encoder_outputs: torch.Tensor,
encoder_output_mask: torch.Tensor) -> Tuple[torch.Tensor, torch.Tensor]:
"""
Given a query (which is typically the decoder hidden state), compute an attention over the
output of the question encoder, and return a weighted sum of the question representations
given this attention. We also return the attention weights themselves.
This is a simple computation, but we have it as a separate method so that the ``forward``
method on the main parser module can call it on the initial hidden state, to simplify the
logic in ``take_step``.
"""
# (group_size, question_length)
question_attention_weights = self._input_attention(query,
encoder_outputs,
encoder_output_mask)
# (group_size, encoder_output_dim)
attended_question = util.weighted_sum(encoder_outputs, question_attention_weights)
return attended_question, question_attention_weights |
Walk over action space to collect completed paths of at most ``self._max_path_length`` steps. | def _walk(self) -> None:
"""
Walk over action space to collect completed paths of at most ``self._max_path_length`` steps.
"""
# Buffer of NTs to expand, previous actions
incomplete_paths = [([str(type_)], [f"{START_SYMBOL} -> {type_}"]) for type_ in
self._world.get_valid_starting_types()]
self._completed_paths = []
actions = self._world.get_valid_actions()
# Keeps track of `MultiMatchNamedBasicTypes` to substitute them with appropriate types.
multi_match_substitutions = self._world.get_multi_match_mapping()
# Overview: We keep track of the buffer of non-terminals to expand, and the action history
# for each incomplete path. At every iteration in the while loop below, we iterate over all
# incomplete paths, expand one non-terminal from the buffer in a depth-first fashion, get
# all possible next actions triggered by that non-terminal and add to the paths. Then, we
# check the expanded paths, to see if they are 1) complete, in which case they are
# added to completed_paths, 2) longer than max_path_length, in which case they are
# discarded, or 3) neither, in which case they are used to form the incomplete_paths for the
# next iteration of this while loop.
# While the non-terminal expansion is done in a depth-first fashion, note that the search over
# the action space itself is breadth-first.
while incomplete_paths:
next_paths = []
for nonterminal_buffer, history in incomplete_paths:
# Taking the last non-terminal added to the buffer. We're going depth-first.
nonterminal = nonterminal_buffer.pop()
next_actions = []
if nonterminal in multi_match_substitutions:
for current_nonterminal in [nonterminal] + multi_match_substitutions[nonterminal]:
if current_nonterminal in actions:
next_actions.extend(actions[current_nonterminal])
elif nonterminal not in actions:
# This happens when the nonterminal corresponds to a type that does not exist in
# the context. For example, in the variable free variant of the WikiTables
# world, there are nonterminals for specific column types (like date). Say we
# produced a path containing "filter_date_greater" already, and we do not have
# an columns of type "date", then this condition would be triggered. We should
# just discard those paths.
continue
else:
next_actions.extend(actions[nonterminal])
# Iterating over all possible next actions.
for action in next_actions:
new_history = history + [action]
new_nonterminal_buffer = nonterminal_buffer[:]
# Since we expand the last action added to the buffer, the left child should be
# added after the right child.
for right_side_part in reversed(self._get_right_side_parts(action)):
if types.is_nonterminal(right_side_part):
new_nonterminal_buffer.append(right_side_part)
next_paths.append((new_nonterminal_buffer, new_history))
incomplete_paths = []
for nonterminal_buffer, path in next_paths:
# An empty buffer means that we've completed this path.
if not nonterminal_buffer:
# Indexing completed paths by the nonterminals they contain.
next_path_index = len(self._completed_paths)
for action in path:
for value in self._get_right_side_parts(action):
if not types.is_nonterminal(value):
self._terminal_path_index[action].add(next_path_index)
self._completed_paths.append(path)
# We're adding to incomplete_paths for the next iteration, only those paths that are
# shorter than the max_path_length. The remaining paths will be discarded.
elif len(path) <= self._max_path_length:
incomplete_paths.append((nonterminal_buffer, path)) |
Check that all the instances have the same types. | def _check_types(self) -> None:
"""
Check that all the instances have the same types.
"""
all_instance_fields_and_types: List[Dict[str, str]] = [{k: v.__class__.__name__
for k, v in x.fields.items()}
for x in self.instances]
# Check all the field names and Field types are the same for every instance.
if not all([all_instance_fields_and_types[0] == x for x in all_instance_fields_and_types]):
raise ConfigurationError("You cannot construct a Batch with non-homogeneous Instances.") |
This method converts this ``Batch`` into a set of pytorch Tensors that can be passed
through a model. In order for the tensors to be valid tensors, all ``Instances`` in this
batch need to be padded to the same lengths wherever padding is necessary, so we do that
first, then we combine all of the tensors for each field in each instance into a set of
batched tensors for each field.
Parameters
----------
padding_lengths : ``Dict[str, Dict[str, int]]``
If a key is present in this dictionary with a non-``None`` value, we will pad to that
length instead of the length calculated from the data. This lets you, e.g., set a
maximum value for sentence length if you want to throw out long sequences.
Entries in this dictionary are keyed first by field name (e.g., "question"), then by
padding key (e.g., "num_tokens").
verbose : ``bool``, optional (default=``False``)
Should we output logging information when we're doing this padding? If the batch is
large, this is nice to have, because padding a large batch could take a long time.
But if you're doing this inside of a data generator, having all of this output per
batch is a bit obnoxious (and really slow).
Returns
-------
tensors : ``Dict[str, DataArray]``
A dictionary of tensors, keyed by field name, suitable for passing as input to a model.
This is a `batch` of instances, so, e.g., if the instances have a "question" field and
an "answer" field, the "question" fields for all of the instances will be grouped
together into a single tensor, and the "answer" fields for all instances will be
similarly grouped in a parallel set of tensors, for batched computation. Additionally,
for complex ``Fields``, the value of the dictionary key is not necessarily a single
tensor. For example, with the ``TextField``, the output is a dictionary mapping
``TokenIndexer`` keys to tensors. The number of elements in this sub-dictionary
therefore corresponds to the number of ``TokenIndexers`` used to index the
``TextField``. Each ``Field`` class is responsible for batching its own output. | def as_tensor_dict(self,
padding_lengths: Dict[str, Dict[str, int]] = None,
verbose: bool = False) -> Dict[str, Union[torch.Tensor, Dict[str, torch.Tensor]]]:
# This complex return type is actually predefined elsewhere as a DataArray,
# but we can't use it because mypy doesn't like it.
"""
This method converts this ``Batch`` into a set of pytorch Tensors that can be passed
through a model. In order for the tensors to be valid tensors, all ``Instances`` in this
batch need to be padded to the same lengths wherever padding is necessary, so we do that
first, then we combine all of the tensors for each field in each instance into a set of
batched tensors for each field.
Parameters
----------
padding_lengths : ``Dict[str, Dict[str, int]]``
If a key is present in this dictionary with a non-``None`` value, we will pad to that
length instead of the length calculated from the data. This lets you, e.g., set a
maximum value for sentence length if you want to throw out long sequences.
Entries in this dictionary are keyed first by field name (e.g., "question"), then by
padding key (e.g., "num_tokens").
verbose : ``bool``, optional (default=``False``)
Should we output logging information when we're doing this padding? If the batch is
large, this is nice to have, because padding a large batch could take a long time.
But if you're doing this inside of a data generator, having all of this output per
batch is a bit obnoxious (and really slow).
Returns
-------
tensors : ``Dict[str, DataArray]``
A dictionary of tensors, keyed by field name, suitable for passing as input to a model.
This is a `batch` of instances, so, e.g., if the instances have a "question" field and
an "answer" field, the "question" fields for all of the instances will be grouped
together into a single tensor, and the "answer" fields for all instances will be
similarly grouped in a parallel set of tensors, for batched computation. Additionally,
for complex ``Fields``, the value of the dictionary key is not necessarily a single
tensor. For example, with the ``TextField``, the output is a dictionary mapping
``TokenIndexer`` keys to tensors. The number of elements in this sub-dictionary
therefore corresponds to the number of ``TokenIndexers`` used to index the
``TextField``. Each ``Field`` class is responsible for batching its own output.
"""
if padding_lengths is None:
padding_lengths = defaultdict(dict)
# First we need to decide _how much_ to pad. To do that, we find the max length for all
# relevant padding decisions from the instances themselves. Then we check whether we were
# given a max length for a particular field and padding key. If we were, we use that
# instead of the instance-based one.
if verbose:
logger.info("Padding batch of size %d to lengths %s", len(self.instances), str(padding_lengths))
logger.info("Getting max lengths from instances")
instance_padding_lengths = self.get_padding_lengths()
if verbose:
logger.info("Instance max lengths: %s", str(instance_padding_lengths))
lengths_to_use: Dict[str, Dict[str, int]] = defaultdict(dict)
for field_name, instance_field_lengths in instance_padding_lengths.items():
for padding_key in instance_field_lengths.keys():
if padding_lengths[field_name].get(padding_key) is not None:
lengths_to_use[field_name][padding_key] = padding_lengths[field_name][padding_key]
else:
lengths_to_use[field_name][padding_key] = instance_field_lengths[padding_key]
# Now we actually pad the instances to tensors.
field_tensors: Dict[str, list] = defaultdict(list)
if verbose:
logger.info("Now actually padding instances to length: %s", str(lengths_to_use))
for instance in self.instances:
for field, tensors in instance.as_tensor_dict(lengths_to_use).items():
field_tensors[field].append(tensors)
# Finally, we combine the tensors that we got for each instance into one big tensor (or set
# of tensors) per field. The `Field` classes themselves have the logic for batching the
# tensors together, so we grab a dictionary of field_name -> field class from the first
# instance in the batch.
field_classes = self.instances[0].fields
final_fields = {}
for field_name, field_tensor_list in field_tensors.items():
final_fields[field_name] = field_classes[field_name].batch_tensors(field_tensor_list)
return final_fields |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.