Skip to content

Commit

Permalink
update for xlnet codes
Browse files Browse the repository at this point in the history
  • Loading branch information
@ewrfcas
ewrfcas committed Oct 24, 2019
1 parent 2e1e33f commit 5fc95b4
Show file tree
Hide file tree
Showing 8 changed files with 3,030 additions and 120 deletions.
359 changes: 359 additions & 0 deletions DRCD_finetune_xlnet.py
View file

Large diffs are not rendered by default.

359 changes: 359 additions & 0 deletions cmrc2018_finetune_xlnet.py
View file

Large diffs are not rendered by default.

122 changes: 62 additions & 60 deletions evaluate/DRCD_output.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -7,12 +7,18 @@
from tqdm import tqdm


def write_predictions_topk(config, all_examples, all_features, all_results, n_best_size,
max_answer_length, do_lower_case, output_prediction_file,
output_nbest_file):
def write_predictions_topk(FLAGS, all_examples, all_features, all_results, n_best_size,
max_answer_length, output_prediction_file, output_nbest_file):
_PrelimPrediction = collections.namedtuple( # pylint: disable=invalid-name
"PrelimPrediction",
["feature_index", "start_index", "end_index",
"start_log_prob", "end_log_prob"])

_NbestPrediction = collections.namedtuple( # pylint: disable=invalid-name
"NbestPrediction", ["text", "start_log_prob", "end_log_prob"])

"""Write final predictions to the json file and log-odds of null if needed."""
print("Writing predictions to: %s" % (output_prediction_file))
print("Writing nbest to: %s" % (output_nbest_file))

example_index_to_features = collections.defaultdict(list)
for feature in all_features:
Expand All @@ -22,37 +28,45 @@ def write_predictions_topk(config, all_examples, all_features, all_results, n_be
for result in all_results:
unique_id_to_result[result.unique_id] = result

_PrelimPrediction = collections.namedtuple( # pylint: disable=invalid-name
"PrelimPrediction",
["feature_index", "start_index", "end_index", "start_logit", "end_logit"])

all_predictions = collections.OrderedDict()
all_nbest_json = collections.OrderedDict()
scores_diff_json = collections.OrderedDict()

for (example_index, example) in enumerate(tqdm(all_examples)):
for (example_index, example) in enumerate(all_examples):
features = example_index_to_features[example_index]

prelim_predictions = []
# keep track of the minimum score of null start+end of position 0
score_null = 1000000 # large and positive

for (feature_index, feature) in enumerate(features):
result = unique_id_to_result[feature['unique_id']]
for i in range(config.start_n_top):
for j in range(config.end_n_top):
start_logit = result.start_top_logits[i]

cur_null_score = result.cls_logits

# if we could have irrelevant answers, get the min score of irrelevant
score_null = min(score_null, cur_null_score)

for i in range(FLAGS.start_n_top):
for j in range(FLAGS.end_n_top):
start_log_prob = result.start_top_log_probs[i]
start_index = result.start_top_index[i]

j_index = i * config.end_n_top + j
j_index = i * FLAGS.end_n_top + j

end_logit = result.end_top_logits[j_index]
end_log_prob = result.end_top_log_probs[j_index]
end_index = result.end_top_index[j_index]

# We could hypothetically create invalid predictions, e.g., predict
# that the start of the span is in the question. We throw out all
# invalid predictions.
if start_index >= len(feature['tokens']):
if start_index >= feature['paragraph_len'] - 1:
continue
if end_index >= len(feature['tokens']):
if end_index >= feature['paragraph_len'] - 1:
continue
if not feature['token_is_max_context'].get(str(start_index), False):

if not feature['token_is_max_context'].get(start_index, False) and \
not feature['token_is_max_context'].get(str(start_index), False):
continue
if end_index < start_index:
continue
Expand All @@ -65,87 +79,75 @@ def write_predictions_topk(config, all_examples, all_features, all_results, n_be
feature_index=feature_index,
start_index=start_index,
end_index=end_index,
start_logit=start_logit,
end_logit=end_logit))
start_log_prob=start_log_prob,
end_log_prob=end_log_prob))

prelim_predictions = sorted(
prelim_predictions,
key=lambda x: (x.start_logit + x.end_logit),
key=lambda x: (x.start_log_prob + x.end_log_prob),
reverse=True)

_NbestPrediction = collections.namedtuple( # pylint: disable=invalid-name
"NbestPrediction", ["text", "start_logit", "end_logit"])

seen_predictions = {}
nbest = []
# ipdb.set_trace()
for pred in prelim_predictions:
if len(nbest) >= n_best_size:
break
feature = features[pred.feature_index]
if pred.start_index > 0: # this is a non-null prediction
tok_tokens = feature['tokens'][pred.start_index:(pred.end_index + 1)]
orig_doc_start = feature['token_to_orig_map'][str(pred.start_index)]
orig_doc_end = feature['token_to_orig_map'][str(pred.end_index)]
orig_tokens = example['ori_doc_tokens'][orig_doc_start:(orig_doc_end + 1)]
tok_text = "".join(tok_tokens)

# De-tokenize WordPieces that have been split off.
tok_text = tok_text.replace(" ##", "")
tok_text = tok_text.replace("##", "")
tok_start_to_orig_index = feature['tok_start_to_orig_index']
tok_end_to_orig_index = feature['tok_end_to_orig_index']
start_orig_pos = tok_start_to_orig_index[pred.start_index]
end_orig_pos = tok_end_to_orig_index[pred.end_index]

# Clean whitespace
tok_text = tok_text.strip()
tok_text = " ".join(tok_text.split())
orig_text = "".join(orig_tokens)
paragraph_text = example['ori_paragraph_text']
final_text = paragraph_text[start_orig_pos: end_orig_pos + 1].strip()

final_text = get_final_text(tok_text, orig_text, do_lower_case)
if final_text in seen_predictions:
continue
if final_text in seen_predictions:
continue

seen_predictions[final_text] = True
else:
final_text = ""
seen_predictions[final_text] = True
seen_predictions[final_text] = True

nbest.append(
_NbestPrediction(
text=final_text,
start_logit=pred.start_logit,
end_logit=pred.end_logit))
start_log_prob=pred.start_log_prob,
end_log_prob=pred.end_log_prob))

# In very rare edge cases we could have no valid predictions. So we
# just create a nonce prediction in this case to avoid failure.
if not nbest:
nbest.append(_NbestPrediction(text="empty", start_logit=0.0, end_logit=0.0))

assert len(nbest) >= 1
# ipdb.set_trace()
nbest.append(
_NbestPrediction(text="", start_log_prob=-1e6,
end_log_prob=-1e6))

total_scores = []
best_non_null_entry = None
for entry in nbest:
total_scores.append(entry.start_logit + entry.end_logit)
total_scores.append(entry.start_log_prob + entry.end_log_prob)
if not best_non_null_entry:
best_non_null_entry = entry

probs = _compute_softmax(total_scores)

# ipdb.set_trace()

nbest_json = []
# ipdb.set_trace()
for (i, entry) in enumerate(nbest):
output = collections.OrderedDict()
output["text"] = entry.text
output["probability"] = float(probs[i])
output["start_logit"] = float(entry.start_logit)
output["end_logit"] = float(entry.end_logit)
output["probability"] = probs[i]
output["start_log_prob"] = entry.start_log_prob
output["end_log_prob"] = entry.end_log_prob
nbest_json.append(output)

assert len(nbest_json) >= 1
assert best_non_null_entry is not None

# ipdb.set_trace()
score_diff = score_null
scores_diff_json[example['qas_id']] = score_diff
# note(zhiliny): always predict best_non_null_entry
# and the evaluation script will search for the best threshold
all_predictions[example['qas_id']] = best_non_null_entry.text

all_predictions[example['qid']] = nbest_json[0]["text"]
all_nbest_json[example['qid']] = nbest_json
all_nbest_json[example['qas_id']] = nbest_json

with open(output_prediction_file, "w") as writer:
writer.write(json.dumps(all_predictions, indent=4, ensure_ascii=False) + "\n")
Expand Down
Loading

0 comments on commit 5fc95b4

Please sign in to comment.