Unverified Commit d5712f7c authored by Thomas Wolf's avatar Thomas Wolf Committed by GitHub
Browse files

Merge branch 'master' into check-link-validity

parents f230d91b 9c58b236
......@@ -72,8 +72,7 @@ class ExamplesTests(unittest.TestCase):
logger.addHandler(stream_handler)
testargs = ["run_squad.py",
"--train_file=./examples/tests_samples/SQUAD/dev-v2.0-small.json",
"--predict_file=./examples/tests_samples/SQUAD/dev-v2.0-small.json",
"--data_dir=./examples/tests_samples/SQUAD",
"--model_name=bert-base-uncased",
"--output_dir=./examples/tests_samples/temp_dir",
"--max_steps=10",
......
{
"version": "v2.0",
"data": [{
"title": "Normans",
"paragraphs": [{
"qas": [{
"question": "In what country is Normandy located?",
"id": "56ddde6b9a695914005b9628",
"answers": [{
"text": "France",
"answer_start": 159
}],
"is_impossible": false
}, {
"question": "When were the Normans in Normandy?",
"id": "56ddde6b9a695914005b9629",
"answers": [{
"text": "10th and 11th centuries",
"answer_start": 94
}],
"is_impossible": false
}, {
"question": "From which countries did the Norse originate?",
"id": "56ddde6b9a695914005b962a",
"answers": [{
"text": "Denmark, Iceland and Norway",
"answer_start": 256
}],
"is_impossible": false
}, {
"plausible_answers": [{
"text": "Rollo",
"answer_start": 308
}],
"question": "Who did King Charles III swear fealty to?",
"id": "5ad39d53604f3c001a3fe8d3",
"answers": [],
"is_impossible": true
}, {
"plausible_answers": [{
"text": "10th century",
"answer_start": 671
}],
"question": "When did the Frankish identity emerge?",
"id": "5ad39d53604f3c001a3fe8d4",
"answers": [],
"is_impossible": true
}],
"context": "The Normans (Norman: Nourmands; French: Normands; Latin: Normanni) were the people who in the 10th and 11th centuries gave their name to Normandy, a region in France. They were descended from Norse (\"Norman\" comes from \"Norseman\") raiders and pirates from Denmark, Iceland and Norway who, under their leader Rollo, agreed to swear fealty to King Charles III of West Francia. Through generations of assimilation and mixing with the native Frankish and Roman-Gaulish populations, their descendants would gradually merge with the Carolingian-based cultures of West Francia. The distinct cultural and ethnic identity of the Normans emerged initially in the first half of the 10th century, and it continued to evolve over the succeeding centuries."
}, {
"qas": [{
"question": "Who was the duke in the battle of Hastings?",
"id": "56dddf4066d3e219004dad5f",
"answers": [{
"text": "William the Conqueror",
"answer_start": 1022
}],
"is_impossible": false
}, {
"plausible_answers": [{
"text": "Antioch",
"answer_start": 1295
}],
"question": "What principality did William the conquerer found?",
"id": "5ad3a266604f3c001a3fea2b",
"answers": [],
"is_impossible": true
}],
"context": "The Norman dynasty had a major political, cultural and military impact on medieval Europe and even the Near East. The Normans were famed for their martial spirit and eventually for their Christian piety, becoming exponents of the Catholic orthodoxy into which they assimilated. They adopted the Gallo-Romance language of the Frankish land they settled, their dialect becoming known as Norman, Normaund or Norman French, an important literary language. The Duchy of Normandy, which they formed by treaty with the French crown, was a great fief of medieval France, and under Richard I of Normandy was forged into a cohesive and formidable principality in feudal tenure. The Normans are noted both for their culture, such as their unique Romanesque architecture and musical traditions, and for their significant military accomplishments and innovations. Norman adventurers founded the Kingdom of Sicily under Roger II after conquering southern Italy on the Saracens and Byzantines, and an expedition on behalf of their duke, William the Conqueror, led to the Norman conquest of England at the Battle of Hastings in 1066. Norman cultural and military influence spread from these new European centres to the Crusader states of the Near East, where their prince Bohemond I founded the Principality of Antioch in the Levant, to Scotland and Wales in Great Britain, to Ireland, and to the coasts of north Africa and the Canary Islands."
}]
}, {
"title": "Computational_complexity_theory",
"paragraphs": [{
"qas": [{
"question": "What branch of theoretical computer science deals with broadly classifying computational problems by difficulty and class of relationship?",
"id": "56e16182e3433e1400422e28",
"answers": [{
"text": "Computational complexity theory",
"answer_start": 0
}],
"is_impossible": false
}, {
"plausible_answers": [{
"text": "algorithm",
"answer_start": 472
}],
"question": "What is a manual application of mathematical steps?",
"id": "5ad5316b5b96ef001a10ab76",
"answers": [],
"is_impossible": true
}],
"context": "Computational complexity theory is a branch of the theory of computation in theoretical computer science that focuses on classifying computational problems according to their inherent difficulty, and relating those classes to each other. A computational problem is understood to be a task that is in principle amenable to being solved by a computer, which is equivalent to stating that the problem may be solved by mechanical application of mathematical steps, such as an algorithm."
}, {
"qas": [{
"question": "What measure of a computational problem broadly defines the inherent difficulty of the solution?",
"id": "56e16839cd28a01900c67887",
"answers": [{
"text": "if its solution requires significant resources",
"answer_start": 46
}],
"is_impossible": false
}, {
"question": "What method is used to intuitively assess or quantify the amount of resources required to solve a computational problem?",
"id": "56e16839cd28a01900c67888",
"answers": [{
"text": "mathematical models of computation",
"answer_start": 176
}],
"is_impossible": false
}, {
"question": "What are two basic primary resources used to guage complexity?",
"id": "56e16839cd28a01900c67889",
"answers": [{
"text": "time and storage",
"answer_start": 305
}],
"is_impossible": false
}, {
"plausible_answers": [{
"text": "the number of gates in a circuit",
"answer_start": 436
}],
"question": "What unit is measured to determine circuit simplicity?",
"id": "5ad532575b96ef001a10ab7f",
"answers": [],
"is_impossible": true
}, {
"plausible_answers": [{
"text": "the number of processors",
"answer_start": 502
}],
"question": "What number is used in perpendicular computing?",
"id": "5ad532575b96ef001a10ab80",
"answers": [],
"is_impossible": true
}],
"context": "A problem is regarded as inherently difficult if its solution requires significant resources, whatever the algorithm used. The theory formalizes this intuition, by introducing mathematical models of computation to study these problems and quantifying the amount of resources needed to solve them, such as time and storage. Other complexity measures are also used, such as the amount of communication (used in communication complexity), the number of gates in a circuit (used in circuit complexity) and the number of processors (used in parallel computing). One of the roles of computational complexity theory is to determine the practical limits on what computers can and cannot do."
}]
}]
}
\ No newline at end of file
""" Official evaluation script for SQuAD version 2.0.
Modified by XLNet authors to update `find_best_threshold` scripts for SQuAD V2.0
In addition to basic functionality, we also compute additional statistics and
plot precision-recall curves if an additional na_prob.json file is provided.
This file is expected to map question ID's to the model's predicted probability
that a question is unanswerable.
"""
import argparse
import collections
import json
import numpy as np
import os
import re
import string
import sys
class EVAL_OPTS():
def __init__(self, data_file, pred_file, out_file="",
na_prob_file="na_prob.json", na_prob_thresh=1.0,
out_image_dir=None, verbose=False):
self.data_file = data_file
self.pred_file = pred_file
self.out_file = out_file
self.na_prob_file = na_prob_file
self.na_prob_thresh = na_prob_thresh
self.out_image_dir = out_image_dir
self.verbose = verbose
OPTS = None
def parse_args():
parser = argparse.ArgumentParser('Official evaluation script for SQuAD version 2.0.')
parser.add_argument('data_file', metavar='data.json', help='Input data JSON file.')
parser.add_argument('pred_file', metavar='pred.json', help='Model predictions.')
parser.add_argument('--out-file', '-o', metavar='eval.json',
help='Write accuracy metrics to file (default is stdout).')
parser.add_argument('--na-prob-file', '-n', metavar='na_prob.json',
help='Model estimates of probability of no answer.')
parser.add_argument('--na-prob-thresh', '-t', type=float, default=1.0,
help='Predict "" if no-answer probability exceeds this (default = 1.0).')
parser.add_argument('--out-image-dir', '-p', metavar='out_images', default=None,
help='Save precision-recall curves to directory.')
parser.add_argument('--verbose', '-v', action='store_true')
if len(sys.argv) == 1:
parser.print_help()
sys.exit(1)
return parser.parse_args()
def make_qid_to_has_ans(dataset):
qid_to_has_ans = {}
for article in dataset:
for p in article['paragraphs']:
for qa in p['qas']:
qid_to_has_ans[qa['id']] = bool(qa['answers'])
return qid_to_has_ans
def normalize_answer(s):
"""Lower text and remove punctuation, articles and extra whitespace."""
def remove_articles(text):
regex = re.compile(r'\b(a|an|the)\b', re.UNICODE)
return re.sub(regex, ' ', text)
def white_space_fix(text):
return ' '.join(text.split())
def remove_punc(text):
exclude = set(string.punctuation)
return ''.join(ch for ch in text if ch not in exclude)
def lower(text):
return text.lower()
return white_space_fix(remove_articles(remove_punc(lower(s))))
def get_tokens(s):
if not s: return []
return normalize_answer(s).split()
def compute_exact(a_gold, a_pred):
return int(normalize_answer(a_gold) == normalize_answer(a_pred))
def compute_f1(a_gold, a_pred):
gold_toks = get_tokens(a_gold)
pred_toks = get_tokens(a_pred)
common = collections.Counter(gold_toks) & collections.Counter(pred_toks)
num_same = sum(common.values())
if len(gold_toks) == 0 or len(pred_toks) == 0:
# If either is no-answer, then F1 is 1 if they agree, 0 otherwise
return int(gold_toks == pred_toks)
if num_same == 0:
return 0
precision = 1.0 * num_same / len(pred_toks)
recall = 1.0 * num_same / len(gold_toks)
f1 = (2 * precision * recall) / (precision + recall)
return f1
def get_raw_scores(dataset, preds):
exact_scores = {}
f1_scores = {}
for article in dataset:
for p in article['paragraphs']:
for qa in p['qas']:
qid = qa['id']
gold_answers = [a['text'] for a in qa['answers']
if normalize_answer(a['text'])]
if not gold_answers:
# For unanswerable questions, only correct answer is empty string
gold_answers = ['']
if qid not in preds:
print('Missing prediction for %s' % qid)
continue
a_pred = preds[qid]
# Take max over all gold answers
exact_scores[qid] = max(compute_exact(a, a_pred) for a in gold_answers)
f1_scores[qid] = max(compute_f1(a, a_pred) for a in gold_answers)
return exact_scores, f1_scores
def apply_no_ans_threshold(scores, na_probs, qid_to_has_ans, na_prob_thresh):
new_scores = {}
for qid, s in scores.items():
pred_na = na_probs[qid] > na_prob_thresh
if pred_na:
new_scores[qid] = float(not qid_to_has_ans[qid])
else:
new_scores[qid] = s
return new_scores
def make_eval_dict(exact_scores, f1_scores, qid_list=None):
if not qid_list:
total = len(exact_scores)
return collections.OrderedDict([
('exact', 100.0 * sum(exact_scores.values()) / total),
('f1', 100.0 * sum(f1_scores.values()) / total),
('total', total),
])
else:
total = len(qid_list)
return collections.OrderedDict([
('exact', 100.0 * sum(exact_scores[k] for k in qid_list) / total),
('f1', 100.0 * sum(f1_scores[k] for k in qid_list) / total),
('total', total),
])
def merge_eval(main_eval, new_eval, prefix):
for k in new_eval:
main_eval['%s_%s' % (prefix, k)] = new_eval[k]
def plot_pr_curve(precisions, recalls, out_image, title):
plt.step(recalls, precisions, color='b', alpha=0.2, where='post')
plt.fill_between(recalls, precisions, step='post', alpha=0.2, color='b')
plt.xlabel('Recall')
plt.ylabel('Precision')
plt.xlim([0.0, 1.05])
plt.ylim([0.0, 1.05])
plt.title(title)
plt.savefig(out_image)
plt.clf()
def make_precision_recall_eval(scores, na_probs, num_true_pos, qid_to_has_ans,
out_image=None, title=None):
qid_list = sorted(na_probs, key=lambda k: na_probs[k])
true_pos = 0.0
cur_p = 1.0
cur_r = 0.0
precisions = [1.0]
recalls = [0.0]
avg_prec = 0.0
for i, qid in enumerate(qid_list):
if qid_to_has_ans[qid]:
true_pos += scores[qid]
cur_p = true_pos / float(i+1)
cur_r = true_pos / float(num_true_pos)
if i == len(qid_list) - 1 or na_probs[qid] != na_probs[qid_list[i+1]]:
# i.e., if we can put a threshold after this point
avg_prec += cur_p * (cur_r - recalls[-1])
precisions.append(cur_p)
recalls.append(cur_r)
if out_image:
plot_pr_curve(precisions, recalls, out_image, title)
return {'ap': 100.0 * avg_prec}
def run_precision_recall_analysis(main_eval, exact_raw, f1_raw, na_probs,
qid_to_has_ans, out_image_dir):
if out_image_dir and not os.path.exists(out_image_dir):
os.makedirs(out_image_dir)
num_true_pos = sum(1 for v in qid_to_has_ans.values() if v)
if num_true_pos == 0:
return
pr_exact = make_precision_recall_eval(
exact_raw, na_probs, num_true_pos, qid_to_has_ans,
out_image=os.path.join(out_image_dir, 'pr_exact.png'),
title='Precision-Recall curve for Exact Match score')
pr_f1 = make_precision_recall_eval(
f1_raw, na_probs, num_true_pos, qid_to_has_ans,
out_image=os.path.join(out_image_dir, 'pr_f1.png'),
title='Precision-Recall curve for F1 score')
oracle_scores = {k: float(v) for k, v in qid_to_has_ans.items()}
pr_oracle = make_precision_recall_eval(
oracle_scores, na_probs, num_true_pos, qid_to_has_ans,
out_image=os.path.join(out_image_dir, 'pr_oracle.png'),
title='Oracle Precision-Recall curve (binary task of HasAns vs. NoAns)')
merge_eval(main_eval, pr_exact, 'pr_exact')
merge_eval(main_eval, pr_f1, 'pr_f1')
merge_eval(main_eval, pr_oracle, 'pr_oracle')
def histogram_na_prob(na_probs, qid_list, image_dir, name):
if not qid_list:
return
x = [na_probs[k] for k in qid_list]
weights = np.ones_like(x) / float(len(x))
plt.hist(x, weights=weights, bins=20, range=(0.0, 1.0))
plt.xlabel('Model probability of no-answer')
plt.ylabel('Proportion of dataset')
plt.title('Histogram of no-answer probability: %s' % name)
plt.savefig(os.path.join(image_dir, 'na_prob_hist_%s.png' % name))
plt.clf()
def find_best_thresh(preds, scores, na_probs, qid_to_has_ans):
num_no_ans = sum(1 for k in qid_to_has_ans if not qid_to_has_ans[k])
cur_score = num_no_ans
best_score = cur_score
best_thresh = 0.0
qid_list = sorted(na_probs, key=lambda k: na_probs[k])
for i, qid in enumerate(qid_list):
if qid not in scores: continue
if qid_to_has_ans[qid]:
diff = scores[qid]
else:
if preds[qid]:
diff = -1
else:
diff = 0
cur_score += diff
if cur_score > best_score:
best_score = cur_score
best_thresh = na_probs[qid]
return 100.0 * best_score / len(scores), best_thresh
def find_best_thresh_v2(preds, scores, na_probs, qid_to_has_ans):
num_no_ans = sum(1 for k in qid_to_has_ans if not qid_to_has_ans[k])
cur_score = num_no_ans
best_score = cur_score
best_thresh = 0.0
qid_list = sorted(na_probs, key=lambda k: na_probs[k])
for i, qid in enumerate(qid_list):
if qid not in scores: continue
if qid_to_has_ans[qid]:
diff = scores[qid]
else:
if preds[qid]:
diff = -1
else:
diff = 0
cur_score += diff
if cur_score > best_score:
best_score = cur_score
best_thresh = na_probs[qid]
has_ans_score, has_ans_cnt = 0, 0
for qid in qid_list:
if not qid_to_has_ans[qid]: continue
has_ans_cnt += 1
if qid not in scores: continue
has_ans_score += scores[qid]
return 100.0 * best_score / len(scores), best_thresh, 1.0 * has_ans_score / has_ans_cnt
def find_all_best_thresh(main_eval, preds, exact_raw, f1_raw, na_probs, qid_to_has_ans):
best_exact, exact_thresh = find_best_thresh(preds, exact_raw, na_probs, qid_to_has_ans)
best_f1, f1_thresh = find_best_thresh(preds, f1_raw, na_probs, qid_to_has_ans)
main_eval['best_exact'] = best_exact
main_eval['best_exact_thresh'] = exact_thresh
main_eval['best_f1'] = best_f1
main_eval['best_f1_thresh'] = f1_thresh
def find_all_best_thresh_v2(main_eval, preds, exact_raw, f1_raw, na_probs, qid_to_has_ans):
best_exact, exact_thresh, has_ans_exact = find_best_thresh_v2(preds, exact_raw, na_probs, qid_to_has_ans)
best_f1, f1_thresh, has_ans_f1 = find_best_thresh_v2(preds, f1_raw, na_probs, qid_to_has_ans)
main_eval['best_exact'] = best_exact
main_eval['best_exact_thresh'] = exact_thresh
main_eval['best_f1'] = best_f1
main_eval['best_f1_thresh'] = f1_thresh
main_eval['has_ans_exact'] = has_ans_exact
main_eval['has_ans_f1'] = has_ans_f1
def main(OPTS):
with open(OPTS.data_file) as f:
dataset_json = json.load(f)
dataset = dataset_json['data']
with open(OPTS.pred_file) as f:
preds = json.load(f)
if OPTS.na_prob_file:
with open(OPTS.na_prob_file) as f:
na_probs = json.load(f)
else:
na_probs = {k: 0.0 for k in preds}
qid_to_has_ans = make_qid_to_has_ans(dataset) # maps qid to True/False
has_ans_qids = [k for k, v in qid_to_has_ans.items() if v]
no_ans_qids = [k for k, v in qid_to_has_ans.items() if not v]
exact_raw, f1_raw = get_raw_scores(dataset, preds)
exact_thresh = apply_no_ans_threshold(exact_raw, na_probs, qid_to_has_ans,
OPTS.na_prob_thresh)
f1_thresh = apply_no_ans_threshold(f1_raw, na_probs, qid_to_has_ans,
OPTS.na_prob_thresh)
out_eval = make_eval_dict(exact_thresh, f1_thresh)
if has_ans_qids:
has_ans_eval = make_eval_dict(exact_thresh, f1_thresh, qid_list=has_ans_qids)
merge_eval(out_eval, has_ans_eval, 'HasAns')
if no_ans_qids:
no_ans_eval = make_eval_dict(exact_thresh, f1_thresh, qid_list=no_ans_qids)
merge_eval(out_eval, no_ans_eval, 'NoAns')
if OPTS.na_prob_file:
find_all_best_thresh(out_eval, preds, exact_raw, f1_raw, na_probs, qid_to_has_ans)
if OPTS.na_prob_file and OPTS.out_image_dir:
run_precision_recall_analysis(out_eval, exact_raw, f1_raw, na_probs,
qid_to_has_ans, OPTS.out_image_dir)
histogram_na_prob(na_probs, has_ans_qids, OPTS.out_image_dir, 'hasAns')
histogram_na_prob(na_probs, no_ans_qids, OPTS.out_image_dir, 'noAns')
if OPTS.out_file:
with open(OPTS.out_file, 'w') as f:
json.dump(out_eval, f)
else:
print(json.dumps(out_eval, indent=2))
return out_eval
if __name__ == '__main__':
OPTS = parse_args()
if OPTS.out_image_dir:
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
main(OPTS)
......@@ -72,7 +72,6 @@ setup(
'transformers-cli'
],
# python_requires='>=3.5.0',
tests_require=['pytest'],
classifiers=[
'Intended Audience :: Science/Research',
'License :: OSI Approved :: Apache Software License',
......
......@@ -18,11 +18,11 @@ from __future__ import print_function
import unittest
import shutil
import pytest
import sys
from .modeling_tf_common_test import (TFCommonTestCases, ids_tensor)
from .configuration_common_test import ConfigTester
from .utils import require_tf, slow
from transformers import XxxConfig, is_tf_available
......@@ -33,10 +33,9 @@ if is_tf_available():
TFXxxForTokenClassification,
TFXxxForQuestionAnswering,
TF_XXX_PRETRAINED_MODEL_ARCHIVE_MAP)
else:
pytestmark = pytest.mark.skip("Require TensorFlow")
@require_tf
class TFXxxModelTest(TFCommonTestCases.TFCommonModelTester):
all_model_classes = (TFXxxModel, TFXxxForMaskedLM, TFXxxForQuestionAnswering,
......@@ -244,7 +243,7 @@ class TFXxxModelTest(TFCommonTestCases.TFCommonModelTester):
config_and_inputs = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_xxx_for_token_classification(*config_and_inputs)
@pytest.mark.slow
@slow
def test_model_from_pretrained(self):
cache_dir = "/tmp/transformers_test/"
for model_name in ['xxx-base-uncased']:
......
......@@ -18,12 +18,12 @@ from __future__ import print_function
import unittest
import shutil
import pytest
from transformers import is_torch_available
from .modeling_common_test import (CommonTestCases, ids_tensor)
from .configuration_common_test import ConfigTester
from .utils import require_torch, slow, torch_device
if is_torch_available():
from transformers import (XxxConfig, XxxModel, XxxForMaskedLM,
......@@ -31,10 +31,9 @@ if is_torch_available():
XxxForQuestionAnswering, XxxForSequenceClassification,
XxxForTokenClassification, XxxForMultipleChoice)
from transformers.modeling_xxx import XXX_PRETRAINED_MODEL_ARCHIVE_MAP
else:
pytestmark = pytest.mark.skip("Require Torch")
@require_torch
class XxxModelTest(CommonTestCases.CommonModelTester):
all_model_classes = (XxxModel, XxxForMaskedLM, XxxForQuestionAnswering,
......@@ -131,6 +130,7 @@ class XxxModelTest(CommonTestCases.CommonModelTester):
def create_and_check_xxx_model(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
model = XxxModel(config=config)
model.to(torch_device)
model.eval()
sequence_output, pooled_output = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids)
sequence_output, pooled_output = model(input_ids, token_type_ids=token_type_ids)
......@@ -148,6 +148,7 @@ class XxxModelTest(CommonTestCases.CommonModelTester):
def create_and_check_xxx_for_masked_lm(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
model = XxxForMaskedLM(config=config)
model.to(torch_device)
model.eval()
loss, prediction_scores = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, masked_lm_labels=token_labels)
result = {
......@@ -162,6 +163,7 @@ class XxxModelTest(CommonTestCases.CommonModelTester):
def create_and_check_xxx_for_question_answering(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
model = XxxForQuestionAnswering(config=config)
model.to(torch_device)
model.eval()
loss, start_logits, end_logits = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids,
start_positions=sequence_labels, end_positions=sequence_labels)
......@@ -182,6 +184,7 @@ class XxxModelTest(CommonTestCases.CommonModelTester):
def create_and_check_xxx_for_sequence_classification(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
config.num_labels = self.num_labels
model = XxxForSequenceClassification(config)
model.to(torch_device)
model.eval()
loss, logits = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, labels=sequence_labels)
result = {
......@@ -197,6 +200,7 @@ class XxxModelTest(CommonTestCases.CommonModelTester):
def create_and_check_xxx_for_token_classification(self, config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels):
config.num_labels = self.num_labels
model = XxxForTokenClassification(config=config)
model.to(torch_device)
model.eval()
loss, logits = model(input_ids, attention_mask=input_mask, token_type_ids=token_type_ids, labels=token_labels)
result = {
......@@ -243,7 +247,7 @@ class XxxModelTest(CommonTestCases.CommonModelTester):
config_and_inputs = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_xxx_for_token_classification(*config_and_inputs)
@pytest.mark.slow
@slow
def test_model_from_pretrained(self):
cache_dir = "/tmp/transformers_test/"
for model_name in list(XXX_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
......
......@@ -26,7 +26,9 @@ from .data import (is_sklearn_available,
InputExample, InputFeatures, DataProcessor,
glue_output_modes, glue_convert_examples_to_features,
glue_processors, glue_tasks_num_labels,
xnli_output_modes, xnli_processors, xnli_tasks_num_labels)
xnli_output_modes, xnli_processors, xnli_tasks_num_labels,
squad_convert_examples_to_features, SquadFeatures,
SquadExample, SquadV1Processor, SquadV2Processor)
if is_sklearn_available():
from .data import glue_compute_metrics, xnli_compute_metrics
......@@ -35,6 +37,7 @@ if is_sklearn_available():
from .tokenization_utils import (PreTrainedTokenizer)
from .tokenization_auto import AutoTokenizer
from .tokenization_bert import BertTokenizer, BasicTokenizer, WordpieceTokenizer
from .tokenization_bert_japanese import BertJapaneseTokenizer, MecabTokenizer, CharacterTokenizer
from .tokenization_openai import OpenAIGPTTokenizer
from .tokenization_transfo_xl import (TransfoXLTokenizer, TransfoXLCorpus)
from .tokenization_gpt2 import GPT2Tokenizer
......@@ -164,6 +167,7 @@ if is_tf_available():
from .modeling_tf_distilbert import (TFDistilBertPreTrainedModel, TFDistilBertMainLayer,
TFDistilBertModel, TFDistilBertForMaskedLM,
TFDistilBertForSequenceClassification,
TFDistilBertForTokenClassification,
TFDistilBertForQuestionAnswering,
TF_DISTILBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
......@@ -174,6 +178,8 @@ if is_tf_available():
from .modeling_tf_albert import (TFAlbertPreTrainedModel, TFAlbertModel, TFAlbertForMaskedLM,
TFAlbertForSequenceClassification,
TF_ALBERT_PRETRAINED_MODEL_ARCHIVE_MAP)
# Optimization
from .optimization_tf import (WarmUp, create_optimizer, AdamWeightDecay, GradientAccumulator)
# TF 2.0 <=> PyTorch conversion utilities
from .modeling_tf_pytorch_utils import (convert_tf_weight_name_to_pt_weight_name,
......
......@@ -42,6 +42,10 @@ BERT_PRETRAINED_CONFIG_ARCHIVE_MAP = {
'bert-base-cased-finetuned-mrpc': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-cased-finetuned-mrpc-config.json",
'bert-base-german-dbmdz-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-cased-config.json",
'bert-base-german-dbmdz-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-uncased-config.json",
'bert-base-japanese': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-config.json",
'bert-base-japanese-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-whole-word-masking-config.json",
'bert-base-japanese-char': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-config.json",
'bert-base-japanese-char-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-config.json"
}
......
......@@ -24,7 +24,7 @@ import logging
import os
from io import open
from .file_utils import cached_path, CONFIG_NAME
from .file_utils import CONFIG_NAME, cached_path, is_remote_url, hf_bucket_url
logger = logging.getLogger(__name__)
......@@ -131,8 +131,10 @@ class PretrainedConfig(object):
config_file = cls.pretrained_config_archive_map[pretrained_model_name_or_path]
elif os.path.isdir(pretrained_model_name_or_path):
config_file = os.path.join(pretrained_model_name_or_path, CONFIG_NAME)
else:
elif os.path.isfile(pretrained_model_name_or_path) or is_remote_url(pretrained_model_name_or_path):
config_file = pretrained_model_name_or_path
else:
config_file = hf_bucket_url(pretrained_model_name_or_path, postfix=CONFIG_NAME)
# redirect to the cache, if necessary
try:
resolved_config_file = cached_path(config_file, cache_dir=cache_dir, force_download=force_download,
......@@ -187,7 +189,7 @@ class PretrainedConfig(object):
@classmethod
def from_json_file(cls, json_file):
"""Constructs a `BertConfig` from a json file of parameters."""
"""Constructs a `Config` from a json file of parameters."""
with open(json_file, "r", encoding='utf-8') as reader:
text = reader.read()
return cls.from_dict(json.loads(text))
......
......@@ -119,10 +119,11 @@ def convert_pt_checkpoint_to_tf(model_type, pytorch_checkpoint_path, config_file
tf_inputs = tf.constant(inputs_list)
tfo = tf_model(tf_inputs, training=False) # build the network
pt_model = pt_model_class.from_pretrained(None,
state_dict = torch.load(pytorch_checkpoint_path, map_location='cpu')
pt_model = pt_model_class.from_pretrained(pretrained_model_name_or_path=None,
config=config,
state_dict=torch.load(pytorch_checkpoint_path,
map_location='cpu'))
state_dict=state_dict)
pt_inputs = torch.tensor(inputs_list)
with torch.no_grad():
pto = pt_model(pt_inputs)
......@@ -139,7 +140,7 @@ def convert_pt_checkpoint_to_tf(model_type, pytorch_checkpoint_path, config_file
def convert_all_pt_checkpoints_to_tf(args_model_type, tf_dump_path, model_shortcut_names_or_path=None, config_shortcut_names_or_path=None,
compare_with_pt_model=False, use_cached_models=False, only_convert_finetuned_models=False):
compare_with_pt_model=False, use_cached_models=False, remove_cached_files=False, only_convert_finetuned_models=False):
assert os.path.isdir(args.tf_dump_path), "--tf_dump_path should be a directory"
if args_model_type is None:
......@@ -187,13 +188,15 @@ def convert_all_pt_checkpoints_to_tf(args_model_type, tf_dump_path, model_shortc
if os.path.isfile(model_shortcut_name):
model_shortcut_name = 'converted_model'
convert_pt_checkpoint_to_tf(model_type=model_type,
pytorch_checkpoint_path=model_file,
config_file=config_file,
tf_dump_path=os.path.join(tf_dump_path, model_shortcut_name + '-tf_model.h5'),
compare_with_pt_model=compare_with_pt_model)
os.remove(config_file)
os.remove(model_file)
if remove_cached_files:
os.remove(config_file)
os.remove(model_file)
if __name__ == "__main__":
......@@ -226,6 +229,9 @@ if __name__ == "__main__":
parser.add_argument("--use_cached_models",
action='store_true',
help = "Use cached models if possible instead of updating to latest checkpoint versions.")
parser.add_argument("--remove_cached_files",
action='store_true',
help = "Remove pytorch models after conversion (save memory when converting in batches).")
parser.add_argument("--only_convert_finetuned_models",
action='store_true',
help = "Only convert finetuned models.")
......@@ -245,4 +251,5 @@ if __name__ == "__main__":
config_shortcut_names_or_path=[args.config_file] if args.config_file is not None else None,
compare_with_pt_model=args.compare_with_pt_model,
use_cached_models=args.use_cached_models,
remove_cached_files=args.remove_cached_files,
only_convert_finetuned_models=args.only_convert_finetuned_models)
from .processors import InputExample, InputFeatures, DataProcessor
from .processors import InputExample, InputFeatures, DataProcessor, SquadFeatures
from .processors import glue_output_modes, glue_processors, glue_tasks_num_labels, glue_convert_examples_to_features
from .processors import squad_convert_examples_to_features, SquadExample, SquadV1Processor, SquadV2Processor
from .processors import xnli_output_modes, xnli_processors, xnli_tasks_num_labels
from .metrics import is_sklearn_available
......
from .utils import InputExample, InputFeatures, DataProcessor
from .glue import glue_output_modes, glue_processors, glue_tasks_num_labels, glue_convert_examples_to_features
from .xnli import xnli_output_modes, xnli_processors, xnli_tasks_num_labels
from .squad import squad_convert_examples_to_features, SquadFeatures, SquadExample, SquadV1Processor, SquadV2Processor
from .xnli import xnli_output_modes, xnli_processors, xnli_tasks_num_labels
\ No newline at end of file
This diff is collapsed.
......@@ -21,7 +21,7 @@ import boto3
from botocore.config import Config
from botocore.exceptions import ClientError
import requests
from tqdm import tqdm
from tqdm.auto import tqdm
from contextlib import contextmanager
logger = logging.getLogger(__name__) # pylint: disable=invalid-name
......@@ -73,6 +73,8 @@ TF2_WEIGHTS_NAME = 'tf_model.h5'
TF_WEIGHTS_NAME = 'model.ckpt'
CONFIG_NAME = "config.json"
S3_BUCKET_PREFIX = "https://s3.amazonaws.com/models.huggingface.co/bert"
def is_torch_available():
return _torch_available
......@@ -103,6 +105,18 @@ else:
return fn
return docstring_decorator
def is_remote_url(url_or_filename):
parsed = urlparse(url_or_filename)
return parsed.scheme in ('http', 'https', 's3')
def hf_bucket_url(identifier, postfix=None):
if postfix is None:
return "/".join((S3_BUCKET_PREFIX, identifier))
else:
return "/".join((S3_BUCKET_PREFIX, identifier, postfix))
def url_to_filename(url, etag=None):
"""
Convert `url` into a hashed filename in a repeatable way.
......@@ -171,9 +185,7 @@ def cached_path(url_or_filename, cache_dir=None, force_download=False, proxies=N
if sys.version_info[0] == 3 and isinstance(cache_dir, Path):
cache_dir = str(cache_dir)
parsed = urlparse(url_or_filename)
if parsed.scheme in ('http', 'https', 's3'):
if is_remote_url(url_or_filename):
# URL, so get it from the cache (downloading if necessary)
return get_from_cache(url_or_filename, cache_dir=cache_dir,
force_download=force_download, proxies=proxies,
......@@ -181,7 +193,7 @@ def cached_path(url_or_filename, cache_dir=None, force_download=False, proxies=N
elif os.path.exists(url_or_filename):
# File, and it exists.
return url_or_filename
elif parsed.scheme == '':
elif urlparse(url_or_filename).scheme == '':
# File, but it doesn't exist.
raise EnvironmentError("file {} not found".format(url_or_filename))
else:
......@@ -245,7 +257,7 @@ def http_get(url, temp_file, proxies=None, resume_size=0):
return
content_length = response.headers.get('Content-Length')
total = resume_size + int(content_length) if content_length is not None else None
progress = tqdm(unit="B", total=total, initial=resume_size)
progress = tqdm(unit="B", unit_scale=True, total=total, initial=resume_size, desc="Downloading")
for chunk in response.iter_content(chunk_size=1024):
if chunk: # filter out keep-alive new chunks
progress.update(len(chunk))
......
......@@ -16,10 +16,11 @@ from __future__ import absolute_import, division, print_function
import os
from os.path import expanduser
import six
import requests
import six
from requests.exceptions import HTTPError
from tqdm import tqdm
ENDPOINT = "https://huggingface.co"
......@@ -129,10 +130,13 @@ class HfApi:
# Even though we presign with the correct content-type,
# the client still has to specify it when uploading the file.
with open(filepath, "rb") as f:
pf = TqdmProgressFileReader(f)
r = requests.put(urls.write, data=f, headers={
"content-type": urls.type,
})
r.raise_for_status()
pf.close()
return urls.access
def list_objs(self, token):
......@@ -148,6 +152,34 @@ class HfApi:
class TqdmProgressFileReader:
"""
Wrap an io.BufferedReader `f` (such as the output of `open(…, "rb")`)
and override `f.read()` so as to display a tqdm progress bar.
see github.com/huggingface/transformers/pull/2078#discussion_r354739608
for implementation details.
"""
def __init__(
self,
f # type: io.BufferedReader
):
self.f = f
self.total_size = os.fstat(f.fileno()).st_size # type: int
self.pbar = tqdm(total=self.total_size, leave=False)
if six.PY3:
# does not work unless PY3
# no big deal as the CLI does not currently support PY2 anyways.
self.read = f.read
f.read = self._read
def _read(self, n=-1):
self.pbar.update(n)
return self.read(n)
def close(self):
self.pbar.close()
class HfFolder:
......
......@@ -28,7 +28,6 @@ from .modeling_xlm import XLMModel, XLMWithLMHeadModel, XLMForSequenceClassifica
from .modeling_roberta import RobertaModel, RobertaForMaskedLM, RobertaForSequenceClassification
from .modeling_distilbert import DistilBertModel, DistilBertForQuestionAnswering, DistilBertForMaskedLM, DistilBertForSequenceClassification
from .modeling_camembert import CamembertModel, CamembertForMaskedLM, CamembertForSequenceClassification, CamembertForMultipleChoice
from .modeling_camembert import CamembertModel, CamembertForMaskedLM, CamembertForSequenceClassification, CamembertForMultipleChoice
from .modeling_albert import AlbertModel, AlbertForMaskedLM, AlbertForSequenceClassification, AlbertForQuestionAnswering
from .modeling_utils import PreTrainedModel, SequenceSummary
......
This diff is collapsed.
......@@ -48,6 +48,10 @@ BERT_PRETRAINED_MODEL_ARCHIVE_MAP = {
'bert-base-cased-finetuned-mrpc': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-cased-finetuned-mrpc-pytorch_model.bin",
'bert-base-german-dbmdz-cased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-cased-pytorch_model.bin",
'bert-base-german-dbmdz-uncased': "https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-german-dbmdz-uncased-pytorch_model.bin",
'bert-base-japanese': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-pytorch_model.bin",
'bert-base-japanese-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-whole-word-masking-pytorch_model.bin",
'bert-base-japanese-char': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-pytorch_model.bin",
'bert-base-japanese-char-whole-word-masking': "https://s3.amazonaws.com/models.huggingface.co/bert/cl-tohoku/bert-base-japanese-char-whole-word-masking-pytorch_model.bin"
}
......@@ -667,18 +671,20 @@ class BertModel(BertPreTrainedModel):
# ourselves in which case we just need to make it broadcastable to all heads.
if attention_mask.dim() == 3:
extended_attention_mask = attention_mask[:, None, :, :]
# Provided a padding mask of dimensions [batch_size, seq_length]
# - if the model is a decoder, apply a causal mask in addition to the padding mask
# - if the model is an encoder, make the mask broadcastable to [batch_size, num_heads, seq_length, seq_length]
if attention_mask.dim() == 2:
elif attention_mask.dim() == 2:
# Provided a padding mask of dimensions [batch_size, seq_length]
# - if the model is a decoder, apply a causal mask in addition to the padding mask
# - if the model is an encoder, make the mask broadcastable to [batch_size, num_heads, seq_length, seq_length]
if self.config.is_decoder:
batch_size, seq_length = input_shape
seq_ids = torch.arange(seq_length, device=device)
causal_mask = seq_ids[None, None, :].repeat(batch_size, seq_length, 1) <= seq_ids[None, :, None]
causal_mask = causal_mask.to(torch.long) # not converting to long will cause errors with pytorch version < 1.3
extended_attention_mask = causal_mask[:, None, :, :] * attention_mask[:, None, None, :]
else:
extended_attention_mask = attention_mask[:, None, None, :]
else:
raise ValueError("Wrong shape for input_ids (shape {}) or attention_mask (shape {})".format(input_shape, attention_mask.shape))
# Since attention_mask is 1.0 for positions we want to attend and 0.0 for
# masked positions, this operation will create a tensor which is 0.0 for
......@@ -690,14 +696,19 @@ class BertModel(BertPreTrainedModel):
# If a 2D ou 3D attention mask is provided for the cross-attention
# we need to make broadcastabe to [batch_size, num_heads, seq_length, seq_length]
if self.config.is_decoder:
if self.config.is_decoder and encoder_hidden_states is not None:
encoder_batch_size, encoder_sequence_length, _ = encoder_hidden_states.size()
encoder_hidden_shape = (encoder_batch_size, encoder_sequence_length)
if encoder_attention_mask is None:
encoder_attention_mask = torch.ones(input_shape, device=device)
encoder_attention_mask = torch.ones(encoder_hidden_shape, device=device)
if encoder_attention_mask.dim() == 3:
encoder_extended_attention_mask = encoder_attention_mask[:, None, :, :]
if encoder_attention_mask.dim() == 2:
elif encoder_attention_mask.dim() == 2:
encoder_extended_attention_mask = encoder_attention_mask[:, None, None, :]
else:
raise ValueError("Wrong shape for encoder_hidden_shape (shape {}) or encoder_attention_mask (shape {})".format(encoder_hidden_shape,
encoder_attention_mask.shape))
encoder_extended_attention_mask = encoder_extended_attention_mask.to(dtype=next(self.parameters()).dtype) # fp16 compatibility
encoder_extended_attention_mask = (1.0 - encoder_extended_attention_mask) * -10000.0
......@@ -1226,9 +1237,9 @@ class BertForQuestionAnswering(BertPreTrainedModel):
question, text = "Who was Jim Henson?", "Jim Henson was a nice puppet"
input_text = "[CLS] " + question + " [SEP] " + text + " [SEP]"
input_ids = tokenizer.encode(input_text)
token_type_ids = [0 if i <= input_ids.index(102) else 1 for i in range(len(input_ids))]
token_type_ids = [0 if i <= input_ids.index(102) else 1 for i in range(len(input_ids))]
start_scores, end_scores = model(torch.tensor([input_ids]), token_type_ids=torch.tensor([token_type_ids]))
all_tokens = tokenizer.convert_ids_to_tokens(input_ids)
all_tokens = tokenizer.convert_ids_to_tokens(input_ids)
print(' '.join(all_tokens[torch.argmax(start_scores) : torch.argmax(end_scores)+1]))
# a nice puppet
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment