File size: 5,357 Bytes
74e8f2f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
# Copyright 2024 Big Vision Authors.
#
# 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.

"""Evaluator for ChartQA variants."""

import functools

import big_vision.evaluators.common as c
import big_vision.pp.tokenizer
import big_vision.utils as u


# Temporary global flag to facilitate backwards compatability. Will be removed
# by the end of year 2023.
API = "jit"


class Evaluator:
  """Evaluator for simple VQA tasks."""

  def __init__(
      self, predict_fn, tokenizer, to_lower=False,
      outfile="{workdir}/{split}.json",
      out_question_key="question_id", out_answer_key="answer",
      *, data, devices, **kw):
    self.get_data_iter, self.steps = c.eval_input_pipeline(
        keep_on_cpu={"answer", "question_id"}, data=data, devices=devices, **kw)

    self.outfile = c.resolve_outfile(outfile, split=data.get("split"))
    self.out_question_key = out_question_key
    self.out_answer_key = out_answer_key

    # We'll need the tokenizer to detokenize the model outputs later.
    self.tok = big_vision.pp.tokenizer.get_tokenizer(tokenizer)
    self.postproc = (lambda s: s.lower()) if to_lower else lambda s: s
    self.decode = functools.partial(
        predict_fn, devices=devices, eos_token=self.tok.eos_token)

  def run(self, train_state):
    """Does one evaluation run, yields metrics."""

    accuracies = []
    relaxed_accuracies = []
    json_out = []
    for _, batch in zip(range(self.steps), self.get_data_iter()):
      # (batch, seqlen) array of decoded generated tokens.
      tokens = self.decode(train_state, batch)

      # (local_batch,) that indicates padding examples (0) vs real examples (1).
      tokens = u.get_local_slice_from_fsarray(tokens)
      ex_masks = u.get_local_slice_from_fsarray(batch["_mask"])

      # Turn predictions into texts and then scores, one by one.
      for i in range(len(tokens)):
        if ex_masks[i] == 0:  # Skip last-batch padding examples
          continue

        answer = self.postproc(self.tok.to_str(tokens[i], stop_at_eos=True))

        gt = self.postproc(batch["answer"][i])
        accuracies.append(float(answer == gt))
        relaxed_accuracies.append(_relaxed_match(gt, answer))
        json_out.append({
            self.out_question_key: batch["question_id"][i].item(),
            self.out_answer_key: answer,
            "gt": gt,
            "relaxed_match": relaxed_accuracies[-1],
        })

    # At this point `accuracies` is a list of per-example scores. However,
    # remember that each host holds a different subset of the examples! So if
    # we were to just return the mean accuracy here, we would effectively only
    # have evaluated on the main host's (who writes metrics) subset!
    # So now, we need to compute global means.
    # There is one more caveat: `process_sum` needs the summands on each host
    # to have the same size. So we either need to include dummy values for
    # the padding examples (last batch, annoying), or we only sum scalars as in
    # sufficient statistics, which we do here.
    sum_accs, sum_relaxed_accs, num = c.process_sum(
        [sum(accuracies), sum(relaxed_accuracies), len(accuracies)])

    # Yielding metric_name, value means logging the metric.
    yield "acc", sum_accs / num
    yield "relaxed_acc", sum_relaxed_accs / num
    yield "num", num  # Just for sanity checks.
    c.multiprocess_write_json(self.outfile, json_out)


def _to_float(text: str) -> float | None:
  try:
    if text.endswith("%"):
      # Convert percentages to floats.
      return float(text.rstrip("%")) / 100.0
    else:
      return float(text)
  except ValueError:
    return None


def _relaxed_match(target: str,
                   prediction: str,
                   max_relative_error: float = 0.05) -> bool:
  """Calculates relaxed correctness.

  The correctness tolerates certain error ratio defined by max_relative_error.
  See https://arxiv.org/pdf/2203.10244.pdf, end of section 5.1:
  “Following Methani et al. (2020), we use a relaxed accuracy measure for the
  numeric answers to allow a minor inaccuracy that may result from the automatic
  data extraction process. We consider an answer to be correct if it is within
  5% of the gold answer. For non-numeric answers, we still need an exact match
  to consider an answer to be correct.”

  Args:
    target: Target string.
    prediction: Predicted string.
    max_relative_error: Maximum relative error.

  Returns:
    Whether the prediction was correct given the specified tolerance.
  """
  prediction_float = _to_float(prediction)
  target_float = _to_float(target)
  # When the target is 0 is always required an exact match.
  if prediction_float is not None and target_float:
    relative_error = abs(prediction_float - target_float) / abs(target_float)
    return relative_error <= max_relative_error
  else:
    return prediction == target