File size: 13,269 Bytes
d1ceb73
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
# Copyright 2024 The HuggingFace Inc. team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import importlib
from typing import TYPE_CHECKING, Any, Dict, List, Optional, Union

from packaging import version

from .base import HfQuantizer


if TYPE_CHECKING:
    from ..modeling_utils import PreTrainedModel

from ..utils import is_accelerate_available, is_bitsandbytes_available, is_torch_available, logging
from .quantizers_utils import get_module_from_name


if is_torch_available():
    import torch

    from ..pytorch_utils import Conv1D

logger = logging.get_logger(__name__)


class Bnb8BitHfQuantizer(HfQuantizer):
    """
    8-bit quantization from bitsandbytes quantization method:
        before loading: converts transformer layers into Linear8bitLt during loading: load 16bit weight and pass to the
        layer object after: quantizes individual weights in Linear8bitLt into 8bit at fitst .cuda() call
    saving:
        from state dict, as usual; saves weights and 'SCB' component
    loading:
        need to locate SCB component and pass to the Linear8bitLt object
    """

    use_keep_in_fp32_modules = True
    requires_parameters_quantization = True
    requires_calibration = False

    required_packages = ["bitsandbytes", "accelerate"]

    def __init__(self, quantization_config, **kwargs):
        super().__init__(quantization_config, **kwargs)

        if self.quantization_config.llm_int8_skip_modules is not None:
            self.modules_to_not_convert = self.quantization_config.llm_int8_skip_modules

    def validate_environment(self, *args, **kwargs):
        if not torch.cuda.is_available():
            raise RuntimeError("No GPU found. A GPU is needed for quantization.")

        if not is_accelerate_available():
            raise ImportError("Using `bitsandbytes` 8-bit quantization requires Accelerate: `pip install accelerate`")
        if not is_bitsandbytes_available():
            raise ImportError(
                "Using `bitsandbytes` 8-bit quantization requires the latest version of bitsandbytes: `pip install -U bitsandbytes`"
            )

        if kwargs.get("from_tf", False) or kwargs.get("from_flax", False):
            raise ValueError(
                "Converting into 4-bit or 8-bit weights from tf/flax weights is currently not supported, please make"
                " sure the weights are in PyTorch format."
            )

        device_map = kwargs.get("device_map", None)
        if (
            device_map is not None
            and isinstance(device_map, dict)
            and not self.quantization_config.llm_int8_enable_fp32_cpu_offload
        ):
            device_map_without_lm_head = {
                key: device_map[key] for key in device_map.keys() if key not in self.modules_to_not_convert
            }
            if "cpu" in device_map_without_lm_head.values() or "disk" in device_map_without_lm_head.values():
                raise ValueError(
                    "Some modules are dispatched on the CPU or the disk. Make sure you have enough GPU RAM to fit the "
                    "quantized model. If you want to dispatch the model on the CPU or the disk while keeping these modules "
                    "in 32-bit, you need to set `load_in_8bit_fp32_cpu_offload=True` and pass a custom `device_map` to "
                    "`from_pretrained`. Check "
                    "https://huggingface.co/docs/transformers/main/en/main_classes/quantization#offload-between-cpu-and-gpu "
                    "for more details. "
                )

        if version.parse(importlib.metadata.version("bitsandbytes")) < version.parse("0.37.2"):
            raise ValueError(
                "You have a version of `bitsandbytes` that is not compatible with 8bit inference and training"
                " make sure you have the latest version of `bitsandbytes` installed"
            )

    def adjust_max_memory(self, max_memory: Dict[str, Union[int, str]]) -> Dict[str, Union[int, str]]:
        # need more space for buffers that are created during quantization
        max_memory = {key: val * 0.90 for key, val in max_memory.items()}
        return max_memory

    def update_torch_dtype(self, torch_dtype: "torch.dtype") -> "torch.dtype":
        if torch_dtype is None:
            # We force the `dtype` to be float16, this is a requirement from `bitsandbytes`
            logger.info(
                "Overriding torch_dtype=%s with `torch_dtype=torch.float16` due to "
                "requirements of `bitsandbytes` to enable model loading in 8-bit or 4-bit. "
                "Pass your own torch_dtype to specify the dtype of the remaining non-linear layers or pass"
                " torch_dtype=torch.float16 to remove this warning.",
                torch_dtype,
            )
            torch_dtype = torch.float16
        return torch_dtype

    def update_device_map(self, device_map):
        if device_map is None:
            device_map = {"": torch.cuda.current_device()}
            logger.info(
                "The device_map was not initialized. "
                "Setting device_map to {'':torch.cuda.current_device()}. "
                "If you want to use the model for inference, please set device_map ='auto' "
            )
        return device_map

    def adjust_target_dtype(self, target_dtype: "torch.dtype") -> "torch.dtype":
        if target_dtype != torch.int8:
            logger.info("target_dtype {target_dtype} is replaced by `torch.int8` for 8-bit BnB quantization")
        return torch.int8

    def check_quantized_param(
        self,
        model: "PreTrainedModel",
        param_value: "torch.Tensor",
        param_name: str,
        state_dict: Dict[str, Any],
        **kwargs,
    ):
        import bitsandbytes as bnb

        module, tensor_name = get_module_from_name(model, param_name)
        if isinstance(module._parameters.get(tensor_name, None), bnb.nn.Int8Params):
            if self.pre_quantized:
                if param_name.replace("weight", "SCB") not in state_dict.keys():
                    raise ValueError("Missing quantization component `SCB`")
                if param_value.dtype != torch.int8:
                    raise ValueError(
                        f"Incompatible dtype `{param_value.dtype}` when loading 8-bit prequantized weight. Expected `torch.int8`."
                    )
            return True
        return False

    def create_quantized_param(
        self,
        model: "PreTrainedModel",
        param_value: "torch.Tensor",
        param_name: str,
        target_device: "torch.device",
        state_dict: Dict[str, Any],
        unexpected_keys: Optional[List[str]] = None,
    ):
        """
        combines logic from _load_state_dict_into_meta_model and .integrations.bitsandbytes.py::set_module_quantized_tensor_to_device()
        needs aux items from state dicts, if found - removes them from unexpected_keys
        """
        import bitsandbytes as bnb

        fp16_statistics_key = param_name.replace("weight", "SCB")
        fp16_weights_format_key = param_name.replace("weight", "weight_format")

        fp16_statistics = state_dict.get(fp16_statistics_key, None)
        fp16_weights_format = state_dict.get(fp16_weights_format_key, None)

        module, tensor_name = get_module_from_name(model, param_name)
        if tensor_name not in module._parameters:
            raise ValueError(f"{module} does not have a parameter or a buffer named {tensor_name}.")

        old_value = getattr(module, tensor_name)

        if not isinstance(module._parameters[tensor_name], bnb.nn.Int8Params):
            raise ValueError(f"Parameter `{tensor_name}` should only be a `bnb.nn.Int8Params` instance.")
        if (
            old_value.device == torch.device("meta")
            and target_device not in ["meta", torch.device("meta")]
            and param_value is None
        ):
            raise ValueError(f"{tensor_name} is on the meta device, we need a `value` to put in on {target_device}.")

        new_value = param_value.to("cpu")
        if self.pre_quantized and not self.is_serializable:
            raise ValueError(
                "Detected int8 weights but the version of bitsandbytes is not compatible with int8 serialization. "
                "Make sure to download the latest `bitsandbytes` version. `pip install --upgrade bitsandbytes`."
            )

        # Support models using `Conv1D` in place of `nn.Linear` (e.g. openai-community/gpt2) by transposing the weight matrix prior to quantization.
        # Since weights are saved in the correct "orientation", we skip transposing when loading.
        if issubclass(module.source_cls, Conv1D):
            if fp16_statistics is None:
                new_value = new_value.T

        kwargs = old_value.__dict__
        new_value = bnb.nn.Int8Params(new_value, requires_grad=False, **kwargs).to(target_device)

        module._parameters[tensor_name] = new_value
        if fp16_statistics is not None:
            setattr(module.weight, "SCB", fp16_statistics.to(target_device))
            if unexpected_keys is not None:
                unexpected_keys.remove(fp16_statistics_key)

        # We just need to pop the `weight_format` keys from the state dict to remove unneeded
        # messages. The correct format is correctly retrieved during the first forward pass.
        if fp16_weights_format is not None and unexpected_keys is not None:
            unexpected_keys.remove(fp16_weights_format_key)

    def _process_model_after_weight_loading(self, model: "PreTrainedModel", **kwargs):
        model.is_loaded_in_8bit = True
        model.is_8bit_serializable = self.is_serializable
        return model

    def _process_model_before_weight_loading(
        self,
        model: "PreTrainedModel",
        device_map,
        keep_in_fp32_modules: List[str] = [],
        **kwargs,
    ):
        from ..integrations import get_keys_to_not_convert, replace_with_bnb_linear

        load_in_8bit_fp32_cpu_offload = self.quantization_config.llm_int8_enable_fp32_cpu_offload

        # We keep some modules such as the lm_head in their original dtype for numerical stability reasons
        if self.quantization_config.llm_int8_skip_modules is None:
            self.modules_to_not_convert = get_keys_to_not_convert(model)
        else:
            self.modules_to_not_convert = self.quantization_config.llm_int8_skip_modules

        if not isinstance(self.modules_to_not_convert, list):
            self.modules_to_not_convert = [self.modules_to_not_convert]

        self.modules_to_not_convert.extend(keep_in_fp32_modules)

        # Extend `self.modules_to_not_convert` to keys that are supposed to be offloaded to `cpu` or `disk`
        if isinstance(device_map, dict) and len(device_map.keys()) > 1:
            keys_on_cpu = [key for key, value in device_map.items() if value in ["disk", "cpu"]]

            if len(keys_on_cpu) > 0 and not load_in_8bit_fp32_cpu_offload:
                raise ValueError(
                    "If you want to offload some keys to `cpu` or `disk`, you need to set "
                    "`llm_int8_enable_fp32_cpu_offload=True`. Note that these modules will not be "
                    " converted to 8-bit but kept in 32-bit."
                )
            self.modules_to_not_convert.extend(keys_on_cpu)

        model = replace_with_bnb_linear(
            model, modules_to_not_convert=self.modules_to_not_convert, quantization_config=self.quantization_config
        )
        # TODO: consider bringing replace_with_bnb_linear() code from ..integrations/bitsandbyter.py to here

        model.config.quantization_config = self.quantization_config

    @property
    def is_serializable(self):
        _bnb_supports_8bit_serialization = version.parse(importlib.metadata.version("bitsandbytes")) > version.parse(
            "0.37.2"
        )

        if not _bnb_supports_8bit_serialization:
            logger.warning(
                "You are calling `save_pretrained` to a 8-bit converted model, but your `bitsandbytes` version doesn't support it. "
                "If you want to save 8-bit models, make sure to have `bitsandbytes>0.37.2` installed. You will most likely face errors or"
                " unexpected behaviours."
            )
            return False

        return True

    @property
    def is_trainable(self) -> bool:
        return version.parse(importlib.metadata.version("bitsandbytes")) >= version.parse("0.37.0")

    def _dequantize(self, model):
        from ..integrations import dequantize_and_replace

        model = dequantize_and_replace(
            model, self.modules_to_not_convert, quantization_config=self.quantization_config
        )
        return model