SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Support for torch-lightning in NER examples (#2890)

Browse Source 
* initial pytorch lightning commit

* tested multigpu

* Fix learning rate schedule

* black formatting

* fix flake8

* isort

* isort

* .

Co-authored-by: Check your git settings! <chris@chris-laptop>
This commit is contained in:
srush
2020-02-20 11:50:05 -05:00
committed by GitHub
parent ab1238393c
commit b662f0e625
10 changed files with 753 additions and 196 deletions
Download Patch File Download Diff File Expand all files Collapse all files

179
examples/ner/README.md Normal file
View File

@@ -0,0 +1,179 @@
## Named Entity Recognition
Based on the scripts [`run_ner.py`](https://github.com/huggingface/transformers/blob/master/examples/ner/run_ner.py) for Pytorch and
[`run_tf_ner.py`](https://github.com/huggingface/transformers/blob/master/examples/ner/run_tf_ner.py) for Tensorflow 2.
This example fine-tune Bert Multilingual on GermEval 2014 (German NER).
Details and results for the fine-tuning provided by @stefan-it.
### Data (Download and pre-processing steps)
Data can be obtained from the [GermEval 2014](https://sites.google.com/site/germeval2014ner/data) shared task page.
Here are the commands for downloading and pre-processing train, dev and test datasets. The original data format has four (tab-separated) columns, in a pre-processing step only the two relevant columns (token and outer span NER annotation) are extracted:
```bash
curl -L 'https://sites.google.com/site/germeval2014ner/data/NER-de-train.tsv?attredirects=0&d=1' \
| grep -v "^#" | cut -f 2,3 | tr '\t' ' ' > train.txt.tmp
curl -L 'https://sites.google.com/site/germeval2014ner/data/NER-de-dev.tsv?attredirects=0&d=1' \
| grep -v "^#" | cut -f 2,3 | tr '\t' ' ' > dev.txt.tmp
curl -L 'https://sites.google.com/site/germeval2014ner/data/NER-de-test.tsv?attredirects=0&d=1' \
| grep -v "^#" | cut -f 2,3 | tr '\t' ' ' > test.txt.tmp
```
The GermEval 2014 dataset contains some strange "control character" tokens like `'\x96', '\u200e', '\x95', '\xad' or '\x80'`. One problem with these tokens is, that `BertTokenizer` returns an empty token for them, resulting in misaligned `InputExample`s. I wrote a script that a) filters these tokens and b) splits longer sentences into smaller ones (once the max. subtoken length is reached).
```bash
wget "https://raw.githubusercontent.com/stefan-it/fine-tuned-berts-seq/master/scripts/preprocess.py"
```
Let's define some variables that we need for further pre-processing steps and training the model:
```bash
export MAX_LENGTH=128
export BERT_MODEL=bert-base-multilingual-cased
```
Run the pre-processing script on training, dev and test datasets:
```bash
python3 preprocess.py train.txt.tmp $BERT_MODEL $MAX_LENGTH > train.txt
python3 preprocess.py dev.txt.tmp $BERT_MODEL $MAX_LENGTH > dev.txt
python3 preprocess.py test.txt.tmp $BERT_MODEL $MAX_LENGTH > test.txt
```
The GermEval 2014 dataset has much more labels than CoNLL-2002/2003 datasets, so an own set of labels must be used:
```bash
cat train.txt dev.txt test.txt | cut -d " " -f 2 | grep -v "^$"| sort | uniq > labels.txt
```
### Prepare the run
Additional environment variables must be set:
```bash
export OUTPUT_DIR=germeval-model
export BATCH_SIZE=32
export NUM_EPOCHS=3
export SAVE_STEPS=750
export SEED=1
```
### Run the Pytorch version
To start training, just run:
```bash
python3 run_ner.py --data_dir ./ \
--model_type bert \
--labels ./labels.txt \
--model_name_or_path $BERT_MODEL \
--output_dir $OUTPUT_DIR \
--max_seq_length $MAX_LENGTH \
--num_train_epochs $NUM_EPOCHS \
--per_gpu_train_batch_size $BATCH_SIZE \
--save_steps $SAVE_STEPS \
--seed $SEED \
--do_train \
--do_eval \
--do_predict
```
If your GPU supports half-precision training, just add the `--fp16` flag. After training, the model will be both evaluated on development and test datasets.
#### Evaluation
Evaluation on development dataset outputs the following for our example:
```bash
10/04/2019 00:42:06 - INFO - __main__ - ***** Eval results *****
10/04/2019 00:42:06 - INFO - __main__ - f1 = 0.8623348017621146
10/04/2019 00:42:06 - INFO - __main__ - loss = 0.07183869666975543
10/04/2019 00:42:06 - INFO - __main__ - precision = 0.8467916366258111
10/04/2019 00:42:06 - INFO - __main__ - recall = 0.8784592370979806
```
On the test dataset the following results could be achieved:
```bash
10/04/2019 00:42:42 - INFO - __main__ - ***** Eval results *****
10/04/2019 00:42:42 - INFO - __main__ - f1 = 0.8614389652384803
10/04/2019 00:42:42 - INFO - __main__ - loss = 0.07064602487454782
10/04/2019 00:42:42 - INFO - __main__ - precision = 0.8604651162790697
10/04/2019 00:42:42 - INFO - __main__ - recall = 0.8624150210424085
```
#### Comparing BERT (large, cased), RoBERTa (large, cased) and DistilBERT (base, uncased)
Here is a small comparison between BERT (large, cased), RoBERTa (large, cased) and DistilBERT (base, uncased) with the same hyperparameters as specified in the [example documentation](https://huggingface.co/transformers/examples.html#named-entity-recognition) (one run):
| Model | F-Score Dev | F-Score Test
| --------------------------------- | ------- | --------
| `bert-large-cased` | 95.59 | 91.70
| `roberta-large` | 95.96 | 91.87
| `distilbert-base-uncased` | 94.34 | 90.32
### Run the Tensorflow 2 version
To start training, just run:
```bash
python3 run_tf_ner.py --data_dir ./ \
--model_type bert \
--labels ./labels.txt \
--model_name_or_path $BERT_MODEL \
--output_dir $OUTPUT_DIR \
--max_seq_length $MAX_LENGTH \
--num_train_epochs $NUM_EPOCHS \
--per_device_train_batch_size $BATCH_SIZE \
--save_steps $SAVE_STEPS \
--seed $SEED \
--do_train \
--do_eval \
--do_predict
```
Such as the Pytorch version, if your GPU supports half-precision training, just add the `--fp16` flag. After training, the model will be both evaluated on development and test datasets.
#### Evaluation
Evaluation on development dataset outputs the following for our example:
```bash
precision recall f1-score support
LOCderiv 0.7619 0.6154 0.6809 52
PERpart 0.8724 0.8997 0.8858 4057
OTHpart 0.9360 0.9466 0.9413 711
ORGpart 0.7015 0.6989 0.7002 269
LOCpart 0.7668 0.8488 0.8057 496
LOC 0.8745 0.9191 0.8963 235
ORGderiv 0.7723 0.8571 0.8125 91
OTHderiv 0.4800 0.6667 0.5581 18
OTH 0.5789 0.6875 0.6286 16
PERderiv 0.5385 0.3889 0.4516 18
PER 0.5000 0.5000 0.5000 2
ORG 0.0000 0.0000 0.0000 3
micro avg 0.8574 0.8862 0.8715 5968
macro avg 0.8575 0.8862 0.8713 5968
```
On the test dataset the following results could be achieved:
```bash
precision recall f1-score support
PERpart 0.8847 0.8944 0.8896 9397
OTHpart 0.9376 0.9353 0.9365 1639
ORGpart 0.7307 0.7044 0.7173 697
LOC 0.9133 0.9394 0.9262 561
LOCpart 0.8058 0.8157 0.8107 1150
ORG 0.0000 0.0000 0.0000 8
OTHderiv 0.5882 0.4762 0.5263 42
PERderiv 0.6571 0.5227 0.5823 44
OTH 0.4906 0.6667 0.5652 39
ORGderiv 0.7016 0.7791 0.7383 172
LOCderiv 0.8256 0.6514 0.7282 109
PER 0.0000 0.0000 0.0000 11
micro avg 0.8722 0.8774 0.8748 13869
macro avg 0.8712 0.8774 0.8740 13869
```

32
examples/ner/run.sh Normal file
View File

@@ -0,0 +1,32 @@
curl -L 'https://sites.google.com/site/germeval2014ner/data/NER-de-train.tsv?attredirects=0&d=1' \
| grep -v "^#" | cut -f 2,3 | tr '\t' ' ' > train.txt.tmp
curl -L 'https://sites.google.com/site/germeval2014ner/data/NER-de-dev.tsv?attredirects=0&d=1' \
| grep -v "^#" | cut -f 2,3 | tr '\t' ' ' > dev.txt.tmp
curl -L 'https://sites.google.com/site/germeval2014ner/data/NER-de-test.tsv?attredirects=0&d=1' \
| grep -v "^#" | cut -f 2,3 | tr '\t' ' ' > test.txt.tmp
wget "https://raw.githubusercontent.com/stefan-it/fine-tuned-berts-seq/master/scripts/preprocess.py"
export MAX_LENGTH=128
export BERT_MODEL=bert-base-multilingual-cased
python3 preprocess.py train.txt.tmp $BERT_MODEL $MAX_LENGTH > train.txt
python3 preprocess.py dev.txt.tmp $BERT_MODEL $MAX_LENGTH > dev.txt
python3 preprocess.py test.txt.tmp $BERT_MODEL $MAX_LENGTH > test.txt
cat train.txt dev.txt test.txt | cut -d " " -f 2 | grep -v "^$"| sort | uniq > labels.txt
export OUTPUT_DIR=germeval-model
export BATCH_SIZE=32
export NUM_EPOCHS=3
export SAVE_STEPS=750
export SEED=1
python3 run_ner.py --data_dir ./ \
--model_type bert \
--labels ./labels.txt \
--model_name_or_path $BERT_MODEL \
--output_dir $OUTPUT_DIR \
--max_seq_length $MAX_LENGTH \
--num_train_epochs $NUM_EPOCHS \
--per_gpu_train_batch_size $BATCH_SIZE \
--save_steps $SAVE_STEPS \
--seed $SEED \
--do_train \
--do_eval \
--do_predict

688
examples/ner/run_ner.py Normal file
View File

@@ -0,0 +1,688 @@
# coding=utf-8
# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
# Copyright (c) 2018, NVIDIA CORPORATION. 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.
""" Fine-tuning the library models for named entity recognition on CoNLL-2003 (Bert or Roberta). """
import argparse
import glob
import logging
import os
import random
import numpy as np
import torch
from seqeval.metrics import f1_score, precision_score, recall_score
from torch.nn import CrossEntropyLoss
from torch.utils.data import DataLoader, RandomSampler, SequentialSampler, TensorDataset
from torch.utils.data.distributed import DistributedSampler
from tqdm import tqdm, trange
from transformers import (
WEIGHTS_NAME,
AdamW,
BertConfig,
BertForTokenClassification,
BertTokenizer,
CamembertConfig,
CamembertForTokenClassification,
CamembertTokenizer,
DistilBertConfig,
DistilBertForTokenClassification,
DistilBertTokenizer,
RobertaConfig,
RobertaForTokenClassification,
RobertaTokenizer,
XLMRobertaConfig,
XLMRobertaForTokenClassification,
XLMRobertaTokenizer,
get_linear_schedule_with_warmup,
)
from utils_ner import convert_examples_to_features, get_labels, read_examples_from_file
try:
from torch.utils.tensorboard import SummaryWriter
except ImportError:
from tensorboardX import SummaryWriter
logger = logging.getLogger(__name__)
ALL_MODELS = sum(
(
tuple(conf.pretrained_config_archive_map.keys())
for conf in (BertConfig, RobertaConfig, DistilBertConfig, CamembertConfig, XLMRobertaConfig)
),
(),
)
MODEL_CLASSES = {
"bert": (BertConfig, BertForTokenClassification, BertTokenizer),
"roberta": (RobertaConfig, RobertaForTokenClassification, RobertaTokenizer),
"distilbert": (DistilBertConfig, DistilBertForTokenClassification, DistilBertTokenizer),
"camembert": (CamembertConfig, CamembertForTokenClassification, CamembertTokenizer),
"xlmroberta": (XLMRobertaConfig, XLMRobertaForTokenClassification, XLMRobertaTokenizer),
}
def set_seed(args):
random.seed(args.seed)
np.random.seed(args.seed)
torch.manual_seed(args.seed)
if args.n_gpu > 0:
torch.cuda.manual_seed_all(args.seed)
def train(args, train_dataset, model, tokenizer, labels, pad_token_label_id):
""" Train the model """
if args.local_rank in [-1, 0]:
tb_writer = SummaryWriter()
args.train_batch_size = args.per_gpu_train_batch_size * max(1, args.n_gpu)
train_sampler = RandomSampler(train_dataset) if args.local_rank == -1 else DistributedSampler(train_dataset)
train_dataloader = DataLoader(train_dataset, sampler=train_sampler, batch_size=args.train_batch_size)
if args.max_steps > 0:
t_total = args.max_steps
args.num_train_epochs = args.max_steps // (len(train_dataloader) // args.gradient_accumulation_steps) + 1
else:
t_total = len(train_dataloader) // args.gradient_accumulation_steps * args.num_train_epochs
# Prepare optimizer and schedule (linear warmup and decay)
no_decay = ["bias", "LayerNorm.weight"]
optimizer_grouped_parameters = [
{
"params": [p for n, p in model.named_parameters() if not any(nd in n for nd in no_decay)],
"weight_decay": args.weight_decay,
},
{"params": [p for n, p in model.named_parameters() if any(nd in n for nd in no_decay)], "weight_decay": 0.0},
]
optimizer = AdamW(optimizer_grouped_parameters, lr=args.learning_rate, eps=args.adam_epsilon)
scheduler = get_linear_schedule_with_warmup(
optimizer, num_warmup_steps=args.warmup_steps, num_training_steps=t_total
)
# Check if saved optimizer or scheduler states exist
if os.path.isfile(os.path.join(args.model_name_or_path, "optimizer.pt")) and os.path.isfile(
os.path.join(args.model_name_or_path, "scheduler.pt")
):
# Load in optimizer and scheduler states
optimizer.load_state_dict(torch.load(os.path.join(args.model_name_or_path, "optimizer.pt")))
scheduler.load_state_dict(torch.load(os.path.join(args.model_name_or_path, "scheduler.pt")))
if args.fp16:
try:
from apex import amp
except ImportError:
raise ImportError("Please install apex from https://www.github.com/nvidia/apex to use fp16 training.")
model, optimizer = amp.initialize(model, optimizer, opt_level=args.fp16_opt_level)
# multi-gpu training (should be after apex fp16 initialization)
if args.n_gpu > 1:
model = torch.nn.DataParallel(model)
# Distributed training (should be after apex fp16 initialization)
if args.local_rank != -1:
model = torch.nn.parallel.DistributedDataParallel(
model, device_ids=[args.local_rank], output_device=args.local_rank, find_unused_parameters=True
)
# Train!
logger.info("***** Running training *****")
logger.info(" Num examples = %d", len(train_dataset))
logger.info(" Num Epochs = %d", args.num_train_epochs)
logger.info(" Instantaneous batch size per GPU = %d", args.per_gpu_train_batch_size)
logger.info(
" Total train batch size (w. parallel, distributed & accumulation) = %d",
args.train_batch_size
* args.gradient_accumulation_steps
* (torch.distributed.get_world_size() if args.local_rank != -1 else 1),
)
logger.info(" Gradient Accumulation steps = %d", args.gradient_accumulation_steps)
logger.info(" Total optimization steps = %d", t_total)
global_step = 0
epochs_trained = 0
steps_trained_in_current_epoch = 0
# Check if continuing training from a checkpoint
if os.path.exists(args.model_name_or_path):
# set global_step to gobal_step of last saved checkpoint from model path
try:
global_step = int(args.model_name_or_path.split("-")[-1].split("/")[0])
except ValueError:
global_step = 0
epochs_trained = global_step // (len(train_dataloader) // args.gradient_accumulation_steps)
steps_trained_in_current_epoch = global_step % (len(train_dataloader) // args.gradient_accumulation_steps)
logger.info(" Continuing training from checkpoint, will skip to saved global_step")
logger.info(" Continuing training from epoch %d", epochs_trained)
logger.info(" Continuing training from global step %d", global_step)
logger.info(" Will skip the first %d steps in the first epoch", steps_trained_in_current_epoch)
tr_loss, logging_loss = 0.0, 0.0
model.zero_grad()
train_iterator = trange(
epochs_trained, int(args.num_train_epochs), desc="Epoch", disable=args.local_rank not in [-1, 0]
)
set_seed(args) # Added here for reproductibility
for _ in train_iterator:
epoch_iterator = tqdm(train_dataloader, desc="Iteration", disable=args.local_rank not in [-1, 0])
for step, batch in enumerate(epoch_iterator):
# Skip past any already trained steps if resuming training
if steps_trained_in_current_epoch > 0:
steps_trained_in_current_epoch -= 1
continue
model.train()
batch = tuple(t.to(args.device) for t in batch)
inputs = {"input_ids": batch[0], "attention_mask": batch[1], "labels": batch[3]}
if args.model_type != "distilbert":
inputs["token_type_ids"] = (
batch[2] if args.model_type in ["bert", "xlnet"] else None
) # XLM and RoBERTa don"t use segment_ids
outputs = model(**inputs)
loss = outputs[0] # model outputs are always tuple in pytorch-transformers (see doc)
if args.n_gpu > 1:
loss = loss.mean() # mean() to average on multi-gpu parallel training
if args.gradient_accumulation_steps > 1:
loss = loss / args.gradient_accumulation_steps
if args.fp16:
with amp.scale_loss(loss, optimizer) as scaled_loss:
scaled_loss.backward()
else:
loss.backward()
tr_loss += loss.item()
if (step + 1) % args.gradient_accumulation_steps == 0:
if args.fp16:
torch.nn.utils.clip_grad_norm_(amp.master_params(optimizer), args.max_grad_norm)
else:
torch.nn.utils.clip_grad_norm_(model.parameters(), args.max_grad_norm)
scheduler.step() # Update learning rate schedule
optimizer.step()
model.zero_grad()
global_step += 1
if args.local_rank in [-1, 0] and args.logging_steps > 0 and global_step % args.logging_steps == 0:
# Log metrics
if (
args.local_rank == -1 and args.evaluate_during_training
): # Only evaluate when single GPU otherwise metrics may not average well
results, _ = evaluate(args, model, tokenizer, labels, pad_token_label_id, mode="dev")
for key, value in results.items():
tb_writer.add_scalar("eval_{}".format(key), value, global_step)
tb_writer.add_scalar("lr", scheduler.get_lr()[0], global_step)
tb_writer.add_scalar("loss", (tr_loss - logging_loss) / args.logging_steps, global_step)
logging_loss = tr_loss
if args.local_rank in [-1, 0] and args.save_steps > 0 and global_step % args.save_steps == 0:
# Save model checkpoint
output_dir = os.path.join(args.output_dir, "checkpoint-{}".format(global_step))
if not os.path.exists(output_dir):
os.makedirs(output_dir)
model_to_save = (
model.module if hasattr(model, "module") else model
) # Take care of distributed/parallel training
model_to_save.save_pretrained(output_dir)
tokenizer.save_pretrained(output_dir)
torch.save(args, os.path.join(output_dir, "training_args.bin"))
logger.info("Saving model checkpoint to %s", output_dir)
torch.save(optimizer.state_dict(), os.path.join(output_dir, "optimizer.pt"))
torch.save(scheduler.state_dict(), os.path.join(output_dir, "scheduler.pt"))
logger.info("Saving optimizer and scheduler states to %s", output_dir)
if args.max_steps > 0 and global_step > args.max_steps:
epoch_iterator.close()
break
if args.max_steps > 0 and global_step > args.max_steps:
train_iterator.close()
break
if args.local_rank in [-1, 0]:
tb_writer.close()
return global_step, tr_loss / global_step
def evaluate(args, model, tokenizer, labels, pad_token_label_id, mode, prefix=""):
eval_dataset = load_and_cache_examples(args, tokenizer, labels, pad_token_label_id, mode=mode)
args.eval_batch_size = args.per_gpu_eval_batch_size * max(1, args.n_gpu)
# Note that DistributedSampler samples randomly
eval_sampler = SequentialSampler(eval_dataset) if args.local_rank == -1 else DistributedSampler(eval_dataset)
eval_dataloader = DataLoader(eval_dataset, sampler=eval_sampler, batch_size=args.eval_batch_size)
# multi-gpu evaluate
if args.n_gpu > 1:
model = torch.nn.DataParallel(model)
# Eval!
logger.info("***** Running evaluation %s *****", prefix)
logger.info(" Num examples = %d", len(eval_dataset))
logger.info(" Batch size = %d", args.eval_batch_size)
eval_loss = 0.0
nb_eval_steps = 0
preds = None
out_label_ids = None
model.eval()
for batch in tqdm(eval_dataloader, desc="Evaluating"):
batch = tuple(t.to(args.device) for t in batch)
with torch.no_grad():
inputs = {"input_ids": batch[0], "attention_mask": batch[1], "labels": batch[3]}
if args.model_type != "distilbert":
inputs["token_type_ids"] = (
batch[2] if args.model_type in ["bert", "xlnet"] else None
) # XLM and RoBERTa don"t use segment_ids
outputs = model(**inputs)
tmp_eval_loss, logits = outputs[:2]
if args.n_gpu > 1:
tmp_eval_loss = tmp_eval_loss.mean() # mean() to average on multi-gpu parallel evaluating
eval_loss += tmp_eval_loss.item()
nb_eval_steps += 1
if preds is None:
preds = logits.detach().cpu().numpy()
out_label_ids = inputs["labels"].detach().cpu().numpy()
else:
preds = np.append(preds, logits.detach().cpu().numpy(), axis=0)
out_label_ids = np.append(out_label_ids, inputs["labels"].detach().cpu().numpy(), axis=0)
eval_loss = eval_loss / nb_eval_steps
preds = np.argmax(preds, axis=2)
label_map = {i: label for i, label in enumerate(labels)}
out_label_list = [[] for _ in range(out_label_ids.shape[0])]
preds_list = [[] for _ in range(out_label_ids.shape[0])]
for i in range(out_label_ids.shape[0]):
for j in range(out_label_ids.shape[1]):
if out_label_ids[i, j] != pad_token_label_id:
out_label_list[i].append(label_map[out_label_ids[i][j]])
preds_list[i].append(label_map[preds[i][j]])
results = {
"loss": eval_loss,
"precision": precision_score(out_label_list, preds_list),
"recall": recall_score(out_label_list, preds_list),
"f1": f1_score(out_label_list, preds_list),
}
logger.info("***** Eval results %s *****", prefix)
for key in sorted(results.keys()):
logger.info(" %s = %s", key, str(results[key]))
return results, preds_list
def load_and_cache_examples(args, tokenizer, labels, pad_token_label_id, mode):
if args.local_rank not in [-1, 0] and not evaluate:
torch.distributed.barrier() # Make sure only the first process in distributed training process the dataset, and the others will use the cache
# Load data features from cache or dataset file
cached_features_file = os.path.join(
args.data_dir,
"cached_{}_{}_{}".format(
mode, list(filter(None, args.model_name_or_path.split("/"))).pop(), str(args.max_seq_length)
),
)
if os.path.exists(cached_features_file) and not args.overwrite_cache:
logger.info("Loading features from cached file %s", cached_features_file)
features = torch.load(cached_features_file)
else:
logger.info("Creating features from dataset file at %s", args.data_dir)
examples = read_examples_from_file(args.data_dir, mode)
features = convert_examples_to_features(
examples,
labels,
args.max_seq_length,
tokenizer,
cls_token_at_end=bool(args.model_type in ["xlnet"]),
# xlnet has a cls token at the end
cls_token=tokenizer.cls_token,
cls_token_segment_id=2 if args.model_type in ["xlnet"] else 0,
sep_token=tokenizer.sep_token,
sep_token_extra=bool(args.model_type in ["roberta"]),
# roberta uses an extra separator b/w pairs of sentences, cf. github.com/pytorch/fairseq/commit/1684e166e3da03f5b600dbb7855cb98ddfcd0805
pad_on_left=bool(args.model_type in ["xlnet"]),
# pad on the left for xlnet
pad_token=tokenizer.convert_tokens_to_ids([tokenizer.pad_token])[0],
pad_token_segment_id=4 if args.model_type in ["xlnet"] else 0,
pad_token_label_id=pad_token_label_id,
)
if args.local_rank in [-1, 0]:
logger.info("Saving features into cached file %s", cached_features_file)
torch.save(features, cached_features_file)
if args.local_rank == 0 and not evaluate:
torch.distributed.barrier() # Make sure only the first process in distributed training process the dataset, and the others will use the cache
# Convert to Tensors and build dataset
all_input_ids = torch.tensor([f.input_ids for f in features], dtype=torch.long)
all_input_mask = torch.tensor([f.input_mask for f in features], dtype=torch.long)
all_segment_ids = torch.tensor([f.segment_ids for f in features], dtype=torch.long)
all_label_ids = torch.tensor([f.label_ids for f in features], dtype=torch.long)
dataset = TensorDataset(all_input_ids, all_input_mask, all_segment_ids, all_label_ids)
return dataset
def main():
parser = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"--data_dir",
default=None,
type=str,
required=True,
help="The input data dir. Should contain the training files for the CoNLL-2003 NER task.",
)
parser.add_argument(
"--model_type",
default=None,
type=str,
required=True,
help="Model type selected in the list: " + ", ".join(MODEL_CLASSES.keys()),
)
parser.add_argument(
"--model_name_or_path",
default=None,
type=str,
required=True,
help="Path to pre-trained model or shortcut name selected in the list: " + ", ".join(ALL_MODELS),
)
parser.add_argument(
"--output_dir",
default=None,
type=str,
required=True,
help="The output directory where the model predictions and checkpoints will be written.",
)
# Other parameters
parser.add_argument(
"--labels",
default="",
type=str,
help="Path to a file containing all labels. If not specified, CoNLL-2003 labels are used.",
)
parser.add_argument(
"--config_name", default="", type=str, help="Pretrained config name or path if not the same as model_name"
)
parser.add_argument(
"--tokenizer_name",
default="",
type=str,
help="Pretrained tokenizer name or path if not the same as model_name",
)
parser.add_argument(
"--cache_dir",
default="",
type=str,
help="Where do you want to store the pre-trained models downloaded from s3",
)
parser.add_argument(
"--max_seq_length",
default=128,
type=int,
help="The maximum total input sequence length after tokenization. Sequences longer "
"than this will be truncated, sequences shorter will be padded.",
)
parser.add_argument("--do_train", action="store_true", help="Whether to run training.")
parser.add_argument("--do_eval", action="store_true", help="Whether to run eval on the dev set.")
parser.add_argument("--do_predict", action="store_true", help="Whether to run predictions on the test set.")
parser.add_argument(
"--evaluate_during_training",
action="store_true",
help="Whether to run evaluation during training at each logging step.",
)
parser.add_argument(
"--do_lower_case", action="store_true", help="Set this flag if you are using an uncased model."
)
parser.add_argument("--per_gpu_train_batch_size", default=8, type=int, help="Batch size per GPU/CPU for training.")
parser.add_argument(
"--per_gpu_eval_batch_size", default=8, type=int, help="Batch size per GPU/CPU for evaluation."
)
parser.add_argument(
"--gradient_accumulation_steps",
type=int,
default=1,
help="Number of updates steps to accumulate before performing a backward/update pass.",
)
parser.add_argument("--learning_rate", default=5e-5, type=float, help="The initial learning rate for Adam.")
parser.add_argument("--weight_decay", default=0.0, type=float, help="Weight decay if we apply some.")
parser.add_argument("--adam_epsilon", default=1e-8, type=float, help="Epsilon for Adam optimizer.")
parser.add_argument("--max_grad_norm", default=1.0, type=float, help="Max gradient norm.")
parser.add_argument(
"--num_train_epochs", default=3.0, type=float, help="Total number of training epochs to perform."
)
parser.add_argument(
"--max_steps",
default=-1,
type=int,
help="If > 0: set total number of training steps to perform. Override num_train_epochs.",
)
parser.add_argument("--warmup_steps", default=0, type=int, help="Linear warmup over warmup_steps.")
parser.add_argument("--logging_steps", type=int, default=500, help="Log every X updates steps.")
parser.add_argument("--save_steps", type=int, default=500, help="Save checkpoint every X updates steps.")
parser.add_argument(
"--eval_all_checkpoints",
action="store_true",
help="Evaluate all checkpoints starting with the same prefix as model_name ending and ending with step number",
)
parser.add_argument("--no_cuda", action="store_true", help="Avoid using CUDA when available")
parser.add_argument(
"--overwrite_output_dir", action="store_true", help="Overwrite the content of the output directory"
)
parser.add_argument(
"--overwrite_cache", action="store_true", help="Overwrite the cached training and evaluation sets"
)
parser.add_argument("--seed", type=int, default=42, help="random seed for initialization")
parser.add_argument(
"--fp16",
action="store_true",
help="Whether to use 16-bit (mixed) precision (through NVIDIA apex) instead of 32-bit",
)
parser.add_argument(
"--fp16_opt_level",
type=str,
default="O1",
help="For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']."
"See details at https://nvidia.github.io/apex/amp.html",
)
parser.add_argument("--local_rank", type=int, default=-1, help="For distributed training: local_rank")
parser.add_argument("--server_ip", type=str, default="", help="For distant debugging.")
parser.add_argument("--server_port", type=str, default="", help="For distant debugging.")
args = parser.parse_args()
if (
os.path.exists(args.output_dir)
and os.listdir(args.output_dir)
and args.do_train
and not args.overwrite_output_dir
):
raise ValueError(
"Output directory ({}) already exists and is not empty. Use --overwrite_output_dir to overcome.".format(
args.output_dir
)
)
# Setup distant debugging if needed
if args.server_ip and args.server_port:
# Distant debugging - see https://code.visualstudio.com/docs/python/debugging#_attach-to-a-local-script
import ptvsd
print("Waiting for debugger attach")
ptvsd.enable_attach(address=(args.server_ip, args.server_port), redirect_output=True)
ptvsd.wait_for_attach()
# Setup CUDA, GPU & distributed training
if args.local_rank == -1 or args.no_cuda:
device = torch.device("cuda" if torch.cuda.is_available() and not args.no_cuda else "cpu")
args.n_gpu = torch.cuda.device_count()
else: # Initializes the distributed backend which will take care of sychronizing nodes/GPUs
torch.cuda.set_device(args.local_rank)
device = torch.device("cuda", args.local_rank)
torch.distributed.init_process_group(backend="nccl")
args.n_gpu = 1
args.device = device
# Setup logging
logging.basicConfig(
format="%(asctime)s - %(levelname)s - %(name)s - %(message)s",
datefmt="%m/%d/%Y %H:%M:%S",
level=logging.INFO if args.local_rank in [-1, 0] else logging.WARN,
)
logger.warning(
"Process rank: %s, device: %s, n_gpu: %s, distributed training: %s, 16-bits training: %s",
args.local_rank,
device,
args.n_gpu,
bool(args.local_rank != -1),
args.fp16,
)
# Set seed
set_seed(args)
# Prepare CONLL-2003 task
labels = get_labels(args.labels)
num_labels = len(labels)
# Use cross entropy ignore index as padding label id so that only real label ids contribute to the loss later
pad_token_label_id = CrossEntropyLoss().ignore_index
# Load pretrained model and tokenizer
if args.local_rank not in [-1, 0]:
torch.distributed.barrier() # Make sure only the first process in distributed training will download model & vocab
args.model_type = args.model_type.lower()
config_class, model_class, tokenizer_class = MODEL_CLASSES[args.model_type]
config = config_class.from_pretrained(
args.config_name if args.config_name else args.model_name_or_path,
num_labels=num_labels,
cache_dir=args.cache_dir if args.cache_dir else None,
)
tokenizer = tokenizer_class.from_pretrained(
args.tokenizer_name if args.tokenizer_name else args.model_name_or_path,
do_lower_case=args.do_lower_case,
cache_dir=args.cache_dir if args.cache_dir else None,
)
model = model_class.from_pretrained(
args.model_name_or_path,
from_tf=bool(".ckpt" in args.model_name_or_path),
config=config,
cache_dir=args.cache_dir if args.cache_dir else None,
)
if args.local_rank == 0:
torch.distributed.barrier() # Make sure only the first process in distributed training will download model & vocab
model.to(args.device)
logger.info("Training/evaluation parameters %s", args)
# Training
if args.do_train:
train_dataset = load_and_cache_examples(args, tokenizer, labels, pad_token_label_id, mode="train")
global_step, tr_loss = train(args, train_dataset, model, tokenizer, labels, pad_token_label_id)
logger.info(" global_step = %s, average loss = %s", global_step, tr_loss)
# Saving best-practices: if you use defaults names for the model, you can reload it using from_pretrained()
if args.do_train and (args.local_rank == -1 or torch.distributed.get_rank() == 0):
# Create output directory if needed
if not os.path.exists(args.output_dir) and args.local_rank in [-1, 0]:
os.makedirs(args.output_dir)
logger.info("Saving model checkpoint to %s", args.output_dir)
# Save a trained model, configuration and tokenizer using `save_pretrained()`.
# They can then be reloaded using `from_pretrained()`
model_to_save = (
model.module if hasattr(model, "module") else model
) # Take care of distributed/parallel training
model_to_save.save_pretrained(args.output_dir)
tokenizer.save_pretrained(args.output_dir)
# Good practice: save your training arguments together with the trained model
torch.save(args, os.path.join(args.output_dir, "training_args.bin"))
# Evaluation
results = {}
if args.do_eval and args.local_rank in [-1, 0]:
tokenizer = tokenizer_class.from_pretrained(args.output_dir, do_lower_case=args.do_lower_case)
checkpoints = [args.output_dir]
if args.eval_all_checkpoints:
checkpoints = list(
os.path.dirname(c) for c in sorted(glob.glob(args.output_dir + "/**/" + WEIGHTS_NAME, recursive=True))
)
logging.getLogger("pytorch_transformers.modeling_utils").setLevel(logging.WARN) # Reduce logging
logger.info("Evaluate the following checkpoints: %s", checkpoints)
for checkpoint in checkpoints:
global_step = checkpoint.split("-")[-1] if len(checkpoints) > 1 else ""
model = model_class.from_pretrained(checkpoint)
model.to(args.device)
result, _ = evaluate(args, model, tokenizer, labels, pad_token_label_id, mode="dev", prefix=global_step)
if global_step:
result = {"{}_{}".format(global_step, k): v for k, v in result.items()}
results.update(result)
output_eval_file = os.path.join(args.output_dir, "eval_results.txt")
with open(output_eval_file, "w") as writer:
for key in sorted(results.keys()):
writer.write("{} = {}\n".format(key, str(results[key])))
if args.do_predict and args.local_rank in [-1, 0]:
tokenizer = tokenizer_class.from_pretrained(args.output_dir, do_lower_case=args.do_lower_case)
model = model_class.from_pretrained(args.output_dir)
model.to(args.device)
result, predictions = evaluate(args, model, tokenizer, labels, pad_token_label_id, mode="test")
# Save results
output_test_results_file = os.path.join(args.output_dir, "test_results.txt")
with open(output_test_results_file, "w") as writer:
for key in sorted(result.keys()):
writer.write("{} = {}\n".format(key, str(result[key])))
# Save predictions
output_test_predictions_file = os.path.join(args.output_dir, "test_predictions.txt")
with open(output_test_predictions_file, "w") as writer:
with open(os.path.join(args.data_dir, "test.txt"), "r") as f:
example_id = 0
for line in f:
if line.startswith("-DOCSTART-") or line == "" or line == "\n":
writer.write(line)
if not predictions[example_id]:
example_id += 1
elif predictions[example_id]:
output_line = line.split()[0] + " " + predictions[example_id].pop(0) + "\n"
writer.write(output_line)
else:
logger.warning("Maximum sequence length exceeded: No prediction for '%s'.", line.split()[0])
return results
if __name__ == "__main__":
main()

21
examples/ner/run_pl.sh Normal file
View File

@@ -0,0 +1,21 @@
# Require pytorch-lightning=0.6
export MAX_LENGTH=128
export BERT_MODEL=bert-base-multilingual-cased
export OUTPUT_DIR=germeval-model
export BATCH_SIZE=32
export NUM_EPOCHS=3
export SAVE_STEPS=750
export SEED=1
python3 run_pl_ner.py --data_dir ./ \
--model_type bert \
--labels ./labels.txt \
--model_name_or_path $BERT_MODEL \
--output_dir $OUTPUT_DIR \
--max_seq_length $MAX_LENGTH \
--num_train_epochs $NUM_EPOCHS \
--train_batch_size 32 \
--save_steps $SAVE_STEPS \
--seed $SEED \
--do_train \
--do_predict

238
examples/ner/run_pl_ner.py Normal file
View File

@@ -0,0 +1,238 @@
import argparse
import glob
import logging
import os
import numpy as np
import torch
from seqeval.metrics import f1_score, precision_score, recall_score
from torch.nn import CrossEntropyLoss
from torch.utils.data import DataLoader, RandomSampler, SequentialSampler, TensorDataset
from torch.utils.data.distributed import DistributedSampler
from transformer_base import BaseTransformer, add_generic_args, generic_train
from utils_ner import convert_examples_to_features, get_labels, read_examples_from_file
logger = logging.getLogger(__name__)
class NERTransformer(BaseTransformer):
"""
A training module for NER. See BaseTransformer for the core options.
"""
def __init__(self, hparams):
self.labels = get_labels(hparams.labels)
num_labels = len(self.labels)
super(NERTransformer, self).__init__(hparams, num_labels)
def forward(self, **inputs):
return self.model(**inputs)
def training_step(self, batch, batch_num):
"Compute loss"
inputs = {"input_ids": batch[0], "attention_mask": batch[1], "labels": batch[3]}
if self.hparams.model_type != "distilbert":
inputs["token_type_ids"] = (
batch[2] if self.hparams.model_type in ["bert", "xlnet"] else None
) # XLM and RoBERTa don"t use segment_ids
outputs = self.forward(**inputs)
loss = outputs[0]
tensorboard_logs = {"loss": loss, "rate": self.lr_scheduler.get_last_lr()[-1]}
return {"loss": loss, "log": tensorboard_logs}
def load_dataset(self, mode, batch_size):
labels = get_labels(self.hparams.labels)
self.pad_token_label_id = CrossEntropyLoss().ignore_index
dataset = self.load_and_cache_examples(labels, self.pad_token_label_id, mode)
if mode == "train":
if self.hparams.n_gpu > 1:
sampler = DistributedSampler(dataset)
else:
sampler = RandomSampler(dataset)
else:
sampler = SequentialSampler(dataset)
dataloader = DataLoader(dataset, sampler=sampler, batch_size=batch_size)
return dataloader
def validation_step(self, batch, batch_nb):
inputs = {"input_ids": batch[0], "attention_mask": batch[1], "labels": batch[3]}
if self.hparams.model_type != "distilbert":
inputs["token_type_ids"] = (
batch[2] if self.hparams.model_type in ["bert", "xlnet"] else None
) # XLM and RoBERTa don"t use segment_ids
outputs = self.forward(**inputs)
tmp_eval_loss, logits = outputs[:2]
preds = logits.detach().cpu().numpy()
out_label_ids = inputs["labels"].detach().cpu().numpy()
return {"val_loss": tmp_eval_loss, "pred": preds, "target": out_label_ids}
def _eval_end(self, outputs):
"Task specific validation"
val_loss_mean = torch.stack([x["val_loss"] for x in outputs]).mean()
preds = np.concatenate([x["pred"] for x in outputs], axis=0)
preds = np.argmax(preds, axis=2)
out_label_ids = np.concatenate([x["target"] for x in outputs], axis=0)
label_map = {i: label for i, label in enumerate(self.labels)}
out_label_list = [[] for _ in range(out_label_ids.shape[0])]
preds_list = [[] for _ in range(out_label_ids.shape[0])]
for i in range(out_label_ids.shape[0]):
for j in range(out_label_ids.shape[1]):
if out_label_ids[i, j] != self.pad_token_label_id:
out_label_list[i].append(label_map[out_label_ids[i][j]])
preds_list[i].append(label_map[preds[i][j]])
results = {
"val_loss": val_loss_mean,
"precision": precision_score(out_label_list, preds_list),
"recall": recall_score(out_label_list, preds_list),
"f1": f1_score(out_label_list, preds_list),
}
if self.is_logger():
logger.info(self.proc_rank)
logger.info("***** Eval results *****")
for key in sorted(results.keys()):
logger.info(" %s = %s", key, str(results[key]))
tensorboard_logs = results
ret = {k: v for k, v in results.items()}
ret["log"] = tensorboard_logs
return ret, preds_list, out_label_list
def validation_end(self, outputs):
ret, preds, targets = self._eval_end(outputs)
return ret
def test_end(self, outputs):
ret, predictions, targets = self._eval_end(outputs)
if self.is_logger():
# Write output to a file:
# Save results
output_test_results_file = os.path.join(self.hparams.output_dir, "test_results.txt")
with open(output_test_results_file, "w") as writer:
for key in sorted(ret.keys()):
if key != "log":
writer.write("{} = {}\n".format(key, str(ret[key])))
# Save predictions
output_test_predictions_file = os.path.join(self.hparams.output_dir, "test_predictions.txt")
with open(output_test_predictions_file, "w") as writer:
with open(os.path.join(self.hparams.data_dir, "test.txt"), "r") as f:
example_id = 0
for line in f:
if line.startswith("-DOCSTART-") or line == "" or line == "\n":
writer.write(line)
if not predictions[example_id]:
example_id += 1
elif predictions[example_id]:
output_line = line.split()[0] + " " + predictions[example_id].pop(0) + "\n"
writer.write(output_line)
else:
logger.warning(
"Maximum sequence length exceeded: No prediction for '%s'.", line.split()[0]
)
return ret
def load_and_cache_examples(self, labels, pad_token_label_id, mode):
args = self.hparams
tokenizer = self.tokenizer
if self.proc_rank not in [-1, 0] and mode == "train":
torch.distributed.barrier() # Make sure only the first process in distributed training process the dataset, and the others will use the cache
# Load data features from cache or dataset file
cached_features_file = os.path.join(
args.data_dir,
"cached_{}_{}_{}".format(
mode, list(filter(None, args.model_name_or_path.split("/"))).pop(), str(args.max_seq_length)
),
)
if os.path.exists(cached_features_file) and not args.overwrite_cache:
logger.info("Loading features from cached file %s", cached_features_file)
features = torch.load(cached_features_file)
else:
logger.info("Creating features from dataset file at %s", args.data_dir)
examples = read_examples_from_file(args.data_dir, mode)
features = convert_examples_to_features(
examples,
labels,
args.max_seq_length,
tokenizer,
cls_token_at_end=bool(args.model_type in ["xlnet"]),
cls_token=tokenizer.cls_token,
cls_token_segment_id=2 if args.model_type in ["xlnet"] else 0,
sep_token=tokenizer.sep_token,
sep_token_extra=bool(args.model_type in ["roberta"]),
pad_on_left=bool(args.model_type in ["xlnet"]),
pad_token=tokenizer.convert_tokens_to_ids([tokenizer.pad_token])[0],
pad_token_segment_id=4 if args.model_type in ["xlnet"] else 0,
pad_token_label_id=pad_token_label_id,
)
if self.proc_rank in [-1, 0]:
logger.info("Saving features into cached file %s", cached_features_file)
torch.save(features, cached_features_file)
if self.proc_rank == 0 and mode == "train":
torch.distributed.barrier() # Make sure only the first process in distributed training process the dataset, and the others will use the cache
# Convert to Tensors and build dataset
all_input_ids = torch.tensor([f.input_ids for f in features], dtype=torch.long)
all_input_mask = torch.tensor([f.input_mask for f in features], dtype=torch.long)
all_segment_ids = torch.tensor([f.segment_ids for f in features], dtype=torch.long)
all_label_ids = torch.tensor([f.label_ids for f in features], dtype=torch.long)
dataset = TensorDataset(all_input_ids, all_input_mask, all_segment_ids, all_label_ids)
return dataset
@staticmethod
def add_model_specific_args(parser, root_dir):
# Add NER specific options
BaseTransformer.add_model_specific_args(parser, root_dir)
parser.add_argument(
"--max_seq_length",
default=128,
type=int,
help="The maximum total input sequence length after tokenization. Sequences longer "
"than this will be truncated, sequences shorter will be padded.",
)
parser.add_argument(
"--labels",
default="",
type=str,
help="Path to a file containing all labels. If not specified, CoNLL-2003 labels are used.",
)
parser.add_argument(
"--data_dir",
default=None,
type=str,
required=True,
help="The input data dir. Should contain the training files for the CoNLL-2003 NER task.",
)
parser.add_argument(
"--overwrite_cache", action="store_true", help="Overwrite the cached training and evaluation sets"
)
return parser
if __name__ == "__main__":
parser = argparse.ArgumentParser()
add_generic_args(parser, os.getcwd())
parser = NERTransformer.add_model_specific_args(parser, os.getcwd())
args = parser.parse_args()
model = NERTransformer(args)
trainer = generic_train(model, args)
if args.do_predict:
checkpoints = list(sorted(glob.glob(args.output_dir + "/checkpoint_*.ckpt", recursive=True)))
NERTransformer.load_from_checkpoint(checkpoints[-1])
trainer.test(model)

655
examples/ner/run_tf_ner.py Normal file
View File

@@ -0,0 +1,655 @@
# coding=utf-8
import collections
import datetime
import glob
import math
import os
import re
import numpy as np
import tensorflow as tf
from absl import app, flags, logging
from seqeval import metrics
from transformers import (
TF2_WEIGHTS_NAME,
BertConfig,
BertTokenizer,
DistilBertConfig,
DistilBertTokenizer,
GradientAccumulator,
RobertaConfig,
RobertaTokenizer,
TFBertForTokenClassification,
TFDistilBertForTokenClassification,
TFRobertaForTokenClassification,
create_optimizer,
)
from utils_ner import convert_examples_to_features, get_labels, read_examples_from_file
try:
from fastprogress import master_bar, progress_bar
except ImportError:
from fastprogress.fastprogress import master_bar, progress_bar
ALL_MODELS = sum(
(tuple(conf.pretrained_config_archive_map.keys()) for conf in (BertConfig, RobertaConfig, DistilBertConfig)), ()
)
MODEL_CLASSES = {
"bert": (BertConfig, TFBertForTokenClassification, BertTokenizer),
"roberta": (RobertaConfig, TFRobertaForTokenClassification, RobertaTokenizer),
"distilbert": (DistilBertConfig, TFDistilBertForTokenClassification, DistilBertTokenizer),
}
flags.DEFINE_string(
"data_dir", None, "The input data dir. Should contain the .conll files (or other data files) " "for the task."
)
flags.DEFINE_string("model_type", None, "Model type selected in the list: " + ", ".join(MODEL_CLASSES.keys()))
flags.DEFINE_string(
"model_name_or_path",
None,
"Path to pre-trained model or shortcut name selected in the list: " + ", ".join(ALL_MODELS),
)
flags.DEFINE_string("output_dir", None, "The output directory where the model checkpoints will be written.")
flags.DEFINE_string(
"labels", "", "Path to a file containing all labels. If not specified, CoNLL-2003 labels are used."
)
flags.DEFINE_string("config_name", "", "Pretrained config name or path if not the same as model_name")
flags.DEFINE_string("tokenizer_name", "", "Pretrained tokenizer name or path if not the same as model_name")
flags.DEFINE_string("cache_dir", "", "Where do you want to store the pre-trained models downloaded from s3")
flags.DEFINE_integer(
"max_seq_length",
128,
"The maximum total input sentence length after tokenization. "
"Sequences longer than this will be truncated, sequences shorter "
"will be padded.",
)
flags.DEFINE_string(
"tpu",
None,
"The Cloud TPU to use for training. This should be either the name "
"used when creating the Cloud TPU, or a grpc://ip.address.of.tpu:8470 "
"url.",
)
flags.DEFINE_integer("num_tpu_cores", 8, "Total number of TPU cores to use.")
flags.DEFINE_boolean("do_train", False, "Whether to run training.")
flags.DEFINE_boolean("do_eval", False, "Whether to run eval on the dev set.")
flags.DEFINE_boolean("do_predict", False, "Whether to run predictions on the test set.")
flags.DEFINE_boolean(
"evaluate_during_training", False, "Whether to run evaluation during training at each logging step."
)
flags.DEFINE_boolean("do_lower_case", False, "Set this flag if you are using an uncased model.")
flags.DEFINE_integer("per_device_train_batch_size", 8, "Batch size per GPU/CPU/TPU for training.")
flags.DEFINE_integer("per_device_eval_batch_size", 8, "Batch size per GPU/CPU/TPU for evaluation.")
flags.DEFINE_integer(
"gradient_accumulation_steps", 1, "Number of updates steps to accumulate before performing a backward/update pass."
)
flags.DEFINE_float("learning_rate", 5e-5, "The initial learning rate for Adam.")
flags.DEFINE_float("weight_decay", 0.0, "Weight decay if we apply some.")
flags.DEFINE_float("adam_epsilon", 1e-8, "Epsilon for Adam optimizer.")
flags.DEFINE_float("max_grad_norm", 1.0, "Max gradient norm.")
flags.DEFINE_integer("num_train_epochs", 3, "Total number of training epochs to perform.")
flags.DEFINE_integer(
"max_steps", -1, "If > 0: set total number of training steps to perform. Override num_train_epochs."
)
flags.DEFINE_integer("warmup_steps", 0, "Linear warmup over warmup_steps.")
flags.DEFINE_integer("logging_steps", 50, "Log every X updates steps.")
flags.DEFINE_integer("save_steps", 50, "Save checkpoint every X updates steps.")
flags.DEFINE_boolean(
"eval_all_checkpoints",
False,
"Evaluate all checkpoints starting with the same prefix as model_name ending and ending with step number",
)
flags.DEFINE_boolean("no_cuda", False, "Avoid using CUDA when available")
flags.DEFINE_boolean("overwrite_output_dir", False, "Overwrite the content of the output directory")
flags.DEFINE_boolean("overwrite_cache", False, "Overwrite the cached training and evaluation sets")
flags.DEFINE_integer("seed", 42, "random seed for initialization")
flags.DEFINE_boolean("fp16", False, "Whether to use 16-bit (mixed) precision instead of 32-bit")
flags.DEFINE_string(
"gpus",
"0",
"Comma separated list of gpus devices. If only one, switch to single "
"gpu strategy, if None takes all the gpus available.",
)
def train(
args, strategy, train_dataset, tokenizer, model, num_train_examples, labels, train_batch_size, pad_token_label_id
):
if args["max_steps"] > 0:
num_train_steps = args["max_steps"] * args["gradient_accumulation_steps"]
args["num_train_epochs"] = 1
else:
num_train_steps = (
math.ceil(num_train_examples / train_batch_size)
// args["gradient_accumulation_steps"]
* args["num_train_epochs"]
)
writer = tf.summary.create_file_writer("/tmp/mylogs")
with strategy.scope():
loss_fct = tf.keras.losses.SparseCategoricalCrossentropy(reduction=tf.keras.losses.Reduction.NONE)
optimizer = create_optimizer(args["learning_rate"], num_train_steps, args["warmup_steps"])
if args["fp16"]:
optimizer = tf.keras.mixed_precision.experimental.LossScaleOptimizer(optimizer, "dynamic")
loss_metric = tf.keras.metrics.Mean(name="loss", dtype=tf.float32)
gradient_accumulator = GradientAccumulator()
logging.info("***** Running training *****")
logging.info(" Num examples = %d", num_train_examples)
logging.info(" Num Epochs = %d", args["num_train_epochs"])
logging.info(" Instantaneous batch size per device = %d", args["per_device_train_batch_size"])
logging.info(
" Total train batch size (w. parallel, distributed & accumulation) = %d",
train_batch_size * args["gradient_accumulation_steps"],
)
logging.info(" Gradient Accumulation steps = %d", args["gradient_accumulation_steps"])
logging.info(" Total training steps = %d", num_train_steps)
model.summary()
@tf.function
def apply_gradients():
grads_and_vars = []
for gradient, variable in zip(gradient_accumulator.gradients, model.trainable_variables):
if gradient is not None:
scaled_gradient = gradient / (args["n_device"] * args["gradient_accumulation_steps"])
grads_and_vars.append((scaled_gradient, variable))
else:
grads_and_vars.append((gradient, variable))
optimizer.apply_gradients(grads_and_vars, args["max_grad_norm"])
gradient_accumulator.reset()
@tf.function
def train_step(train_features, train_labels):
def step_fn(train_features, train_labels):
inputs = {"attention_mask": train_features["input_mask"], "training": True}
if args["model_type"] != "distilbert":
inputs["token_type_ids"] = (
train_features["segment_ids"] if args["model_type"] in ["bert", "xlnet"] else None
)
with tf.GradientTape() as tape:
logits = model(train_features["input_ids"], **inputs)[0]
logits = tf.reshape(logits, (-1, len(labels) + 1))
active_loss = tf.reshape(train_features["input_mask"], (-1,))
active_logits = tf.boolean_mask(logits, active_loss)
train_labels = tf.reshape(train_labels, (-1,))
active_labels = tf.boolean_mask(train_labels, active_loss)
cross_entropy = loss_fct(active_labels, active_logits)
loss = tf.reduce_sum(cross_entropy) * (1.0 / train_batch_size)
grads = tape.gradient(loss, model.trainable_variables)
gradient_accumulator(grads)
return cross_entropy
per_example_losses = strategy.experimental_run_v2(step_fn, args=(train_features, train_labels))
mean_loss = strategy.reduce(tf.distribute.ReduceOp.MEAN, per_example_losses, axis=0)
return mean_loss
current_time = datetime.datetime.now()
train_iterator = master_bar(range(args["num_train_epochs"]))
global_step = 0
logging_loss = 0.0
for epoch in train_iterator:
epoch_iterator = progress_bar(
train_dataset, total=num_train_steps, parent=train_iterator, display=args["n_device"] > 1
)
step = 1
with strategy.scope():
for train_features, train_labels in epoch_iterator:
loss = train_step(train_features, train_labels)
if step % args["gradient_accumulation_steps"] == 0:
strategy.experimental_run_v2(apply_gradients)
loss_metric(loss)
global_step += 1
if args["logging_steps"] > 0 and global_step % args["logging_steps"] == 0:
# Log metrics
if (
args["n_device"] == 1 and args["evaluate_during_training"]
): # Only evaluate when single GPU otherwise metrics may not average well
y_true, y_pred, eval_loss = evaluate(
args, strategy, model, tokenizer, labels, pad_token_label_id, mode="dev"
)
report = metrics.classification_report(y_true, y_pred, digits=4)
logging.info("Eval at step " + str(global_step) + "\n" + report)
logging.info("eval_loss: " + str(eval_loss))
precision = metrics.precision_score(y_true, y_pred)
recall = metrics.recall_score(y_true, y_pred)
f1 = metrics.f1_score(y_true, y_pred)
with writer.as_default():
tf.summary.scalar("eval_loss", eval_loss, global_step)
tf.summary.scalar("precision", precision, global_step)
tf.summary.scalar("recall", recall, global_step)
tf.summary.scalar("f1", f1, global_step)
lr = optimizer.learning_rate
learning_rate = lr(step)
with writer.as_default():
tf.summary.scalar("lr", learning_rate, global_step)
tf.summary.scalar(
"loss", (loss_metric.result() - logging_loss) / args["logging_steps"], global_step
)
logging_loss = loss_metric.result()
with writer.as_default():
tf.summary.scalar("loss", loss_metric.result(), step=step)
if args["save_steps"] > 0 and global_step % args["save_steps"] == 0:
# Save model checkpoint
output_dir = os.path.join(args["output_dir"], "checkpoint-{}".format(global_step))
if not os.path.exists(output_dir):
os.makedirs(output_dir)
model.save_pretrained(output_dir)
logging.info("Saving model checkpoint to %s", output_dir)
train_iterator.child.comment = f"loss : {loss_metric.result()}"
step += 1
train_iterator.write(f"loss epoch {epoch + 1}: {loss_metric.result()}")
loss_metric.reset_states()
logging.info(" Training took time = {}".format(datetime.datetime.now() - current_time))
def evaluate(args, strategy, model, tokenizer, labels, pad_token_label_id, mode):
eval_batch_size = args["per_device_eval_batch_size"] * args["n_device"]
eval_dataset, size = load_and_cache_examples(
args, tokenizer, labels, pad_token_label_id, eval_batch_size, mode=mode
)
eval_dataset = strategy.experimental_distribute_dataset(eval_dataset)
preds = None
num_eval_steps = math.ceil(size / eval_batch_size)
master = master_bar(range(1))
eval_iterator = progress_bar(eval_dataset, total=num_eval_steps, parent=master, display=args["n_device"] > 1)
loss_fct = tf.keras.losses.SparseCategoricalCrossentropy(reduction=tf.keras.losses.Reduction.NONE)
loss = 0.0
logging.info("***** Running evaluation *****")
logging.info(" Num examples = %d", size)
logging.info(" Batch size = %d", eval_batch_size)
for eval_features, eval_labels in eval_iterator:
inputs = {"attention_mask": eval_features["input_mask"], "training": False}
if args["model_type"] != "distilbert":
inputs["token_type_ids"] = (
eval_features["segment_ids"] if args["model_type"] in ["bert", "xlnet"] else None
)
with strategy.scope():
logits = model(eval_features["input_ids"], **inputs)[0]
tmp_logits = tf.reshape(logits, (-1, len(labels) + 1))
active_loss = tf.reshape(eval_features["input_mask"], (-1,))
active_logits = tf.boolean_mask(tmp_logits, active_loss)
tmp_eval_labels = tf.reshape(eval_labels, (-1,))
active_labels = tf.boolean_mask(tmp_eval_labels, active_loss)
cross_entropy = loss_fct(active_labels, active_logits)
loss += tf.reduce_sum(cross_entropy) * (1.0 / eval_batch_size)
if preds is None:
preds = logits.numpy()
label_ids = eval_labels.numpy()
else:
preds = np.append(preds, logits.numpy(), axis=0)
label_ids = np.append(label_ids, eval_labels.numpy(), axis=0)
preds = np.argmax(preds, axis=2)
y_pred = [[] for _ in range(label_ids.shape[0])]
y_true = [[] for _ in range(label_ids.shape[0])]
loss = loss / num_eval_steps
for i in range(label_ids.shape[0]):
for j in range(label_ids.shape[1]):
if label_ids[i, j] != pad_token_label_id:
y_pred[i].append(labels[preds[i, j] - 1])
y_true[i].append(labels[label_ids[i, j] - 1])
return y_true, y_pred, loss.numpy()
def load_cache(cached_file, max_seq_length):
name_to_features = {
"input_ids": tf.io.FixedLenFeature([max_seq_length], tf.int64),
"input_mask": tf.io.FixedLenFeature([max_seq_length], tf.int64),
"segment_ids": tf.io.FixedLenFeature([max_seq_length], tf.int64),
"label_ids": tf.io.FixedLenFeature([max_seq_length], tf.int64),
}
def _decode_record(record):
example = tf.io.parse_single_example(record, name_to_features)
features = {}
features["input_ids"] = example["input_ids"]
features["input_mask"] = example["input_mask"]
features["segment_ids"] = example["segment_ids"]
return features, example["label_ids"]
d = tf.data.TFRecordDataset(cached_file)
d = d.map(_decode_record, num_parallel_calls=4)
count = d.reduce(0, lambda x, _: x + 1)
return d, count.numpy()
def save_cache(features, cached_features_file):
writer = tf.io.TFRecordWriter(cached_features_file)
for (ex_index, feature) in enumerate(features):
if ex_index % 5000 == 0:
logging.info("Writing example %d of %d" % (ex_index, len(features)))
def create_int_feature(values):
f = tf.train.Feature(int64_list=tf.train.Int64List(value=list(values)))
return f
record_feature = collections.OrderedDict()
record_feature["input_ids"] = create_int_feature(feature.input_ids)
record_feature["input_mask"] = create_int_feature(feature.input_mask)
record_feature["segment_ids"] = create_int_feature(feature.segment_ids)
record_feature["label_ids"] = create_int_feature(feature.label_ids)
tf_example = tf.train.Example(features=tf.train.Features(feature=record_feature))
writer.write(tf_example.SerializeToString())
writer.close()
def load_and_cache_examples(args, tokenizer, labels, pad_token_label_id, batch_size, mode):
drop_remainder = True if args["tpu"] or mode == "train" else False
# Load data features from cache or dataset file
cached_features_file = os.path.join(
args["data_dir"],
"cached_{}_{}_{}.tf_record".format(
mode, list(filter(None, args["model_name_or_path"].split("/"))).pop(), str(args["max_seq_length"])
),
)
if os.path.exists(cached_features_file) and not args["overwrite_cache"]:
logging.info("Loading features from cached file %s", cached_features_file)
dataset, size = load_cache(cached_features_file, args["max_seq_length"])
else:
logging.info("Creating features from dataset file at %s", args["data_dir"])
examples = read_examples_from_file(args["data_dir"], mode)
features = convert_examples_to_features(
examples,
labels,
args["max_seq_length"],
tokenizer,
cls_token_at_end=bool(args["model_type"] in ["xlnet"]),
# xlnet has a cls token at the end
cls_token=tokenizer.cls_token,
cls_token_segment_id=2 if args["model_type"] in ["xlnet"] else 0,
sep_token=tokenizer.sep_token,
sep_token_extra=bool(args["model_type"] in ["roberta"]),
# roberta uses an extra separator b/w pairs of sentences, cf. github.com/pytorch/fairseq/commit/1684e166e3da03f5b600dbb7855cb98ddfcd0805
pad_on_left=bool(args["model_type"] in ["xlnet"]),
# pad on the left for xlnet
pad_token=tokenizer.convert_tokens_to_ids([tokenizer.pad_token])[0],
pad_token_segment_id=4 if args["model_type"] in ["xlnet"] else 0,
pad_token_label_id=pad_token_label_id,
)
logging.info("Saving features into cached file %s", cached_features_file)
save_cache(features, cached_features_file)
dataset, size = load_cache(cached_features_file, args["max_seq_length"])
if mode == "train":
dataset = dataset.repeat()
dataset = dataset.shuffle(buffer_size=8192, seed=args["seed"])
dataset = dataset.batch(batch_size, drop_remainder)
dataset = dataset.prefetch(buffer_size=batch_size)
return dataset, size
def main(_):
logging.set_verbosity(logging.INFO)
args = flags.FLAGS.flag_values_dict()
if (
os.path.exists(args["output_dir"])
and os.listdir(args["output_dir"])
and args["do_train"]
and not args["overwrite_output_dir"]
):
raise ValueError(
"Output directory ({}) already exists and is not empty. Use --overwrite_output_dir to overcome.".format(
args["output_dir"]
)
)
if args["fp16"]:
tf.config.optimizer.set_experimental_options({"auto_mixed_precision": True})
if args["tpu"]:
resolver = tf.distribute.cluster_resolver.TPUClusterResolver(tpu=args["tpu"])
tf.config.experimental_connect_to_cluster(resolver)
tf.tpu.experimental.initialize_tpu_system(resolver)
strategy = tf.distribute.experimental.TPUStrategy(resolver)
args["n_device"] = args["num_tpu_cores"]
elif len(args["gpus"].split(",")) > 1:
args["n_device"] = len([f"/gpu:{gpu}" for gpu in args["gpus"].split(",")])
strategy = tf.distribute.MirroredStrategy(devices=[f"/gpu:{gpu}" for gpu in args["gpus"].split(",")])
elif args["no_cuda"]:
args["n_device"] = 1
strategy = tf.distribute.OneDeviceStrategy(device="/cpu:0")
else:
args["n_device"] = len(args["gpus"].split(","))
strategy = tf.distribute.OneDeviceStrategy(device="/gpu:" + args["gpus"].split(",")[0])
logging.warning(
"n_device: %s, distributed training: %s, 16-bits training: %s",
args["n_device"],
bool(args["n_device"] > 1),
args["fp16"],
)
labels = get_labels(args["labels"])
num_labels = len(labels) + 1
pad_token_label_id = 0
config_class, model_class, tokenizer_class = MODEL_CLASSES[args["model_type"]]
config = config_class.from_pretrained(
args["config_name"] if args["config_name"] else args["model_name_or_path"],
num_labels=num_labels,
cache_dir=args["cache_dir"] if args["cache_dir"] else None,
)
logging.info("Training/evaluation parameters %s", args)
# Training
if args["do_train"]:
tokenizer = tokenizer_class.from_pretrained(
args["tokenizer_name"] if args["tokenizer_name"] else args["model_name_or_path"],
do_lower_case=args["do_lower_case"],
cache_dir=args["cache_dir"] if args["cache_dir"] else None,
)
with strategy.scope():
model = model_class.from_pretrained(
args["model_name_or_path"],
from_pt=bool(".bin" in args["model_name_or_path"]),
config=config,
cache_dir=args["cache_dir"] if args["cache_dir"] else None,
)
model.layers[-1].activation = tf.keras.activations.softmax
train_batch_size = args["per_device_train_batch_size"] * args["n_device"]
train_dataset, num_train_examples = load_and_cache_examples(
args, tokenizer, labels, pad_token_label_id, train_batch_size, mode="train"
)
train_dataset = strategy.experimental_distribute_dataset(train_dataset)
train(
args,
strategy,
train_dataset,
tokenizer,
model,
num_train_examples,
labels,
train_batch_size,
pad_token_label_id,
)
if not os.path.exists(args["output_dir"]):
os.makedirs(args["output_dir"])
logging.info("Saving model to %s", args["output_dir"])
model.save_pretrained(args["output_dir"])
tokenizer.save_pretrained(args["output_dir"])
# Evaluation
if args["do_eval"]:
tokenizer = tokenizer_class.from_pretrained(args["output_dir"], do_lower_case=args["do_lower_case"])
checkpoints = []
results = []
if args["eval_all_checkpoints"]:
checkpoints = list(
os.path.dirname(c)
for c in sorted(
glob.glob(args["output_dir"] + "/**/" + TF2_WEIGHTS_NAME, recursive=True),
key=lambda f: int("".join(filter(str.isdigit, f)) or -1),
)
)
logging.info("Evaluate the following checkpoints: %s", checkpoints)
if len(checkpoints) == 0:
checkpoints.append(args["output_dir"])
for checkpoint in checkpoints:
global_step = checkpoint.split("-")[-1] if re.match(".*checkpoint-[0-9]", checkpoint) else "final"
with strategy.scope():
model = model_class.from_pretrained(checkpoint)
y_true, y_pred, eval_loss = evaluate(
args, strategy, model, tokenizer, labels, pad_token_label_id, mode="dev"
)
report = metrics.classification_report(y_true, y_pred, digits=4)
if global_step:
results.append({global_step + "_report": report, global_step + "_loss": eval_loss})
output_eval_file = os.path.join(args["output_dir"], "eval_results.txt")
with tf.io.gfile.GFile(output_eval_file, "w") as writer:
for res in results:
for key, val in res.items():
if "loss" in key:
logging.info(key + " = " + str(val))
writer.write(key + " = " + str(val))
writer.write("\n")
else:
logging.info(key)
logging.info("\n" + report)
writer.write(key + "\n")
writer.write(report)
writer.write("\n")
if args["do_predict"]:
tokenizer = tokenizer_class.from_pretrained(args["output_dir"], do_lower_case=args["do_lower_case"])
model = model_class.from_pretrained(args["output_dir"])
eval_batch_size = args["per_device_eval_batch_size"] * args["n_device"]
predict_dataset, _ = load_and_cache_examples(
args, tokenizer, labels, pad_token_label_id, eval_batch_size, mode="test"
)
y_true, y_pred, pred_loss = evaluate(args, strategy, model, tokenizer, labels, pad_token_label_id, mode="test")
output_test_results_file = os.path.join(args["output_dir"], "test_results.txt")
output_test_predictions_file = os.path.join(args["output_dir"], "test_predictions.txt")
report = metrics.classification_report(y_true, y_pred, digits=4)
with tf.io.gfile.GFile(output_test_results_file, "w") as writer:
report = metrics.classification_report(y_true, y_pred, digits=4)
logging.info("\n" + report)
writer.write(report)
writer.write("\n\nloss = " + str(pred_loss))
with tf.io.gfile.GFile(output_test_predictions_file, "w") as writer:
with tf.io.gfile.GFile(os.path.join(args["data_dir"], "test.txt"), "r") as f:
example_id = 0
for line in f:
if line.startswith("-DOCSTART-") or line == "" or line == "\n":
writer.write(line)
if not y_pred[example_id]:
example_id += 1
elif y_pred[example_id]:
output_line = line.split()[0] + " " + y_pred[example_id].pop(0) + "\n"
writer.write(output_line)
else:
logging.warning("Maximum sequence length exceeded: No prediction for '%s'.", line.split()[0])
if __name__ == "__main__":
flags.mark_flag_as_required("data_dir")
flags.mark_flag_as_required("output_dir")
flags.mark_flag_as_required("model_name_or_path")
flags.mark_flag_as_required("model_type")
app.run(main)

264
examples/ner/transformer_base.py Normal file
View File

@@ -0,0 +1,264 @@
import os
import random
import numpy as np
import pytorch_lightning as pl
import torch
from transformers import (
AdamW,
BertConfig,
BertForTokenClassification,
BertTokenizer,
CamembertConfig,
CamembertForTokenClassification,
CamembertTokenizer,
DistilBertConfig,
DistilBertForTokenClassification,
DistilBertTokenizer,
RobertaConfig,
RobertaForTokenClassification,
RobertaTokenizer,
XLMRobertaConfig,
XLMRobertaForTokenClassification,
XLMRobertaTokenizer,
get_linear_schedule_with_warmup,
)
ALL_MODELS = sum(
(
tuple(conf.pretrained_config_archive_map.keys())
for conf in (BertConfig, RobertaConfig, DistilBertConfig, CamembertConfig, XLMRobertaConfig)
),
(),
)
MODEL_CLASSES = {
"bert": (BertConfig, BertForTokenClassification, BertTokenizer),
"roberta": (RobertaConfig, RobertaForTokenClassification, RobertaTokenizer),
"distilbert": (DistilBertConfig, DistilBertForTokenClassification, DistilBertTokenizer),
"camembert": (CamembertConfig, CamembertForTokenClassification, CamembertTokenizer),
"xlmroberta": (XLMRobertaConfig, XLMRobertaForTokenClassification, XLMRobertaTokenizer),
}
def set_seed(args):
random.seed(args.seed)
np.random.seed(args.seed)
torch.manual_seed(args.seed)
if args.n_gpu > 0:
torch.cuda.manual_seed_all(args.seed)
class BaseTransformer(pl.LightningModule):
def __init__(self, hparams, num_labels=None):
"Initialize a model."
super(BaseTransformer, self).__init__()
self.hparams = hparams
self.hparams.model_type = self.hparams.model_type.lower()
config_class, model_class, tokenizer_class = MODEL_CLASSES[self.hparams.model_type]
config = config_class.from_pretrained(
self.hparams.config_name if self.hparams.config_name else self.hparams.model_name_or_path,
num_labels=num_labels,
cache_dir=self.hparams.cache_dir if self.hparams.cache_dir else None,
)
tokenizer = tokenizer_class.from_pretrained(
self.hparams.tokenizer_name if self.hparams.tokenizer_name else self.hparams.model_name_or_path,
do_lower_case=self.hparams.do_lower_case,
cache_dir=self.hparams.cache_dir if self.hparams.cache_dir else None,
)
model = model_class.from_pretrained(
self.hparams.model_name_or_path,
from_tf=bool(".ckpt" in self.hparams.model_name_or_path),
config=config,
cache_dir=self.hparams.cache_dir if self.hparams.cache_dir else None,
)
self.config, self.tokenizer, self.model = config, tokenizer, model
self.proc_rank = -1
def is_logger(self):
return self.proc_rank <= 0
def configure_optimizers(self):
"Prepare optimizer and schedule (linear warmup and decay)"
model = self.model
t_total = (
len(self.train_dataloader())
// self.hparams.gradient_accumulation_steps
* float(self.hparams.num_train_epochs)
)
no_decay = ["bias", "LayerNorm.weight"]
optimizer_grouped_parameters = [
{
"params": [p for n, p in model.named_parameters() if not any(nd in n for nd in no_decay)],
"weight_decay": self.hparams.weight_decay,
},
{
"params": [p for n, p in model.named_parameters() if any(nd in n for nd in no_decay)],
"weight_decay": 0.0,
},
]
optimizer = AdamW(optimizer_grouped_parameters, lr=self.hparams.learning_rate, eps=self.hparams.adam_epsilon)
scheduler = get_linear_schedule_with_warmup(
optimizer, num_warmup_steps=self.hparams.warmup_steps, num_training_steps=t_total
)
self.lr_scheduler = scheduler
return [optimizer]
def optimizer_step(self, epoch, batch_idx, optimizer, optimizer_idx, second_order_closure=None):
# Step each time.
optimizer.step()
self.lr_scheduler.step()
optimizer.zero_grad()
def get_tqdm_dict(self):
tqdm_dict = {"loss": "{:.3f}".format(self.trainer.avg_loss), "lr": self.lr_scheduler.get_last_lr()[-1]}
return tqdm_dict
def test_step(self, batch, batch_nb):
return self.validation_step(batch, batch_nb)
def test_end(self, outputs):
return self.validation_end(outputs)
@pl.data_loader
def train_dataloader(self):
return self.load_dataset("train", self.hparams.train_batch_size)
@pl.data_loader
def val_dataloader(self):
return self.load_dataset("dev", self.hparams.eval_batch_size)
@pl.data_loader
def test_dataloader(self):
return self.load_dataset("test", self.hparams.eval_batch_size)
def init_ddp_connection(self, proc_rank, world_size):
self.proc_rank = proc_rank
super(BaseTransformer, self).init_ddp_connection(proc_rank, world_size)
@staticmethod
def add_model_specific_args(parser, root_dir):
parser.add_argument(
"--model_type",
default=None,
type=str,
required=True,
help="Model type selected in the list: " + ", ".join(MODEL_CLASSES.keys()),
)
parser.add_argument(
"--model_name_or_path",
default=None,
type=str,
required=True,
help="Path to pre-trained model or shortcut name selected in the list: " + ", ".join(ALL_MODELS),
)
parser.add_argument(
"--config_name", default="", type=str, help="Pretrained config name or path if not the same as model_name"
)
parser.add_argument(
"--tokenizer_name",
default="",
type=str,
help="Pretrained tokenizer name or path if not the same as model_name",
)
parser.add_argument(
"--cache_dir",
default="",
type=str,
help="Where do you want to store the pre-trained models downloaded from s3",
)
parser.add_argument(
"--do_lower_case", action="store_true", help="Set this flag if you are using an uncased model."
)
parser.add_argument("--learning_rate", default=5e-5, type=float, help="The initial learning rate for Adam.")
parser.add_argument("--weight_decay", default=0.0, type=float, help="Weight decay if we apply some.")
parser.add_argument("--adam_epsilon", default=1e-8, type=float, help="Epsilon for Adam optimizer.")
parser.add_argument("--warmup_steps", default=0, type=int, help="Linear warmup over warmup_steps.")
parser.add_argument(
"--num_train_epochs", default=3, type=int, help="Total number of training epochs to perform."
)
parser.add_argument("--train_batch_size", default=32, type=int)
parser.add_argument("--eval_batch_size", default=32, type=int)
def add_generic_args(parser, root_dir):
parser.add_argument(
"--output_dir",
default=None,
type=str,
required=True,
help="The output directory where the model predictions and checkpoints will be written.",
)
parser.add_argument(
"--fp16",
action="store_true",
help="Whether to use 16-bit (mixed) precision (through NVIDIA apex) instead of 32-bit",
)
parser.add_argument(
"--fp16_opt_level",
type=str,
default="O1",
help="For fp16: Apex AMP optimization level selected in ['O0', 'O1', 'O2', and 'O3']."
"See details at https://nvidia.github.io/apex/amp.html",
)
parser.add_argument("--n_gpu", type=int, default=1)
parser.add_argument("--max_grad_norm", default=1.0, type=float, help="Max gradient norm.")
parser.add_argument("--do_train", action="store_true", help="Whether to run training.")
parser.add_argument("--do_predict", action="store_true", help="Whether to run predictions on the test set.")
parser.add_argument(
"--gradient_accumulation_steps",
type=int,
default=1,
help="Number of updates steps to accumulate before performing a backward/update pass.",
)
parser.add_argument("--server_ip", type=str, default="", help="For distant debugging.")
parser.add_argument("--server_port", type=str, default="", help="For distant debugging.")
parser.add_argument("--seed", type=int, default=42, help="random seed for initialization")
def generic_train(model, args):
# init model
set_seed(args)
# Setup distant debugging if needed
if args.server_ip and args.server_port:
# Distant debugging - see https://code.visualstudio.com/docs/python/debugging#_attach-to-a-local-script
import ptvsd
print("Waiting for debugger attach")
ptvsd.enable_attach(address=(args.server_ip, args.server_port), redirect_output=True)
ptvsd.wait_for_attach()
if os.path.exists(args.output_dir) and os.listdir(args.output_dir) and args.do_train:
raise ValueError("Output directory ({}) already exists and is not empty.".format(args.output_dir))
checkpoint_callback = pl.callbacks.ModelCheckpoint(
filepath=args.output_dir, prefix="checkpoint", monitor="val_loss", mode="min", save_top_k=5
)
trainer = pl.Trainer(
accumulate_grad_batches=args.gradient_accumulation_steps,
gpus=args.n_gpu,
max_epochs=args.num_train_epochs,
use_amp=args.fp16,
amp_level=args.fp16_opt_level,
distributed_backend="ddp",
gradient_clip_val=args.max_grad_norm,
checkpoint_callback=checkpoint_callback,
)
if args.do_train:
trainer.fit(model)
return trainer

207
examples/ner/utils_ner.py Normal file
View File

@@ -0,0 +1,207 @@
# coding=utf-8
# Copyright 2018 The Google AI Language Team Authors and The HuggingFace Inc. team.
# Copyright (c) 2018, NVIDIA CORPORATION. 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.
""" Named entity recognition fine-tuning: utilities to work with CoNLL-2003 task. """
import logging
import os
logger = logging.getLogger(__name__)
class InputExample(object):
"""A single training/test example for token classification."""
def __init__(self, guid, words, labels):
"""Constructs a InputExample.
Args:
guid: Unique id for the example.
words: list. The words of the sequence.
labels: (Optional) list. The labels for each word of the sequence. This should be
specified for train and dev examples, but not for test examples.
"""
self.guid = guid
self.words = words
self.labels = labels
class InputFeatures(object):
"""A single set of features of data."""
def __init__(self, input_ids, input_mask, segment_ids, label_ids):
self.input_ids = input_ids
self.input_mask = input_mask
self.segment_ids = segment_ids
self.label_ids = label_ids
def read_examples_from_file(data_dir, mode):
file_path = os.path.join(data_dir, "{}.txt".format(mode))
guid_index = 1
examples = []
with open(file_path, encoding="utf-8") as f:
words = []
labels = []
for line in f:
if line.startswith("-DOCSTART-") or line == "" or line == "\n":
if words:
examples.append(InputExample(guid="{}-{}".format(mode, guid_index), words=words, labels=labels))
guid_index += 1
words = []
labels = []
else:
splits = line.split(" ")
words.append(splits[0])
if len(splits) > 1:
labels.append(splits[-1].replace("\n", ""))
else:
# Examples could have no label for mode = "test"
labels.append("O")
if words:
examples.append(InputExample(guid="{}-{}".format(mode, guid_index), words=words, labels=labels))
return examples
def convert_examples_to_features(
examples,
label_list,
max_seq_length,
tokenizer,
cls_token_at_end=False,
cls_token="[CLS]",
cls_token_segment_id=1,
sep_token="[SEP]",
sep_token_extra=False,
pad_on_left=False,
pad_token=0,
pad_token_segment_id=0,
pad_token_label_id=-100,
sequence_a_segment_id=0,
mask_padding_with_zero=True,
):
""" Loads a data file into a list of `InputBatch`s
`cls_token_at_end` define the location of the CLS token:
- False (Default, BERT/XLM pattern): [CLS] + A + [SEP] + B + [SEP]
- True (XLNet/GPT pattern): A + [SEP] + B + [SEP] + [CLS]
`cls_token_segment_id` define the segment id associated to the CLS token (0 for BERT, 2 for XLNet)
"""
label_map = {label: i for i, label in enumerate(label_list)}
features = []
for (ex_index, example) in enumerate(examples):
if ex_index % 10000 == 0:
logger.info("Writing example %d of %d", ex_index, len(examples))
tokens = []
label_ids = []
for word, label in zip(example.words, example.labels):
word_tokens = tokenizer.tokenize(word)
tokens.extend(word_tokens)
# Use the real label id for the first token of the word, and padding ids for the remaining tokens
label_ids.extend([label_map[label]] + [pad_token_label_id] * (len(word_tokens) - 1))
# Account for [CLS] and [SEP] with "- 2" and with "- 3" for RoBERTa.
special_tokens_count = 3 if sep_token_extra else 2
if len(tokens) > max_seq_length - special_tokens_count:
tokens = tokens[: (max_seq_length - special_tokens_count)]
label_ids = label_ids[: (max_seq_length - special_tokens_count)]
# The convention in BERT is:
# (a) For sequence pairs:
# tokens: [CLS] is this jack ##son ##ville ? [SEP] no it is not . [SEP]
# type_ids: 0 0 0 0 0 0 0 0 1 1 1 1 1 1
# (b) For single sequences:
# tokens: [CLS] the dog is hairy . [SEP]
# type_ids: 0 0 0 0 0 0 0
#
# Where "type_ids" are used to indicate whether this is the first
# sequence or the second sequence. The embedding vectors for `type=0` and
# `type=1` were learned during pre-training and are added to the wordpiece
# embedding vector (and position vector). This is not *strictly* necessary
# since the [SEP] token unambiguously separates the sequences, but it makes
# it easier for the model to learn the concept of sequences.
#
# For classification tasks, the first vector (corresponding to [CLS]) is
# used as as the "sentence vector". Note that this only makes sense because
# the entire model is fine-tuned.
tokens += [sep_token]
label_ids += [pad_token_label_id]
if sep_token_extra:
# roberta uses an extra separator b/w pairs of sentences
tokens += [sep_token]
label_ids += [pad_token_label_id]
segment_ids = [sequence_a_segment_id] * len(tokens)
if cls_token_at_end:
tokens += [cls_token]
label_ids += [pad_token_label_id]
segment_ids += [cls_token_segment_id]
else:
tokens = [cls_token] + tokens
label_ids = [pad_token_label_id] + label_ids
segment_ids = [cls_token_segment_id] + segment_ids
input_ids = tokenizer.convert_tokens_to_ids(tokens)
# The mask has 1 for real tokens and 0 for padding tokens. Only real
# tokens are attended to.
input_mask = [1 if mask_padding_with_zero else 0] * len(input_ids)
# Zero-pad up to the sequence length.
padding_length = max_seq_length - len(input_ids)
if pad_on_left:
input_ids = ([pad_token] * padding_length) + input_ids
input_mask = ([0 if mask_padding_with_zero else 1] * padding_length) + input_mask
segment_ids = ([pad_token_segment_id] * padding_length) + segment_ids
label_ids = ([pad_token_label_id] * padding_length) + label_ids
else:
input_ids += [pad_token] * padding_length
input_mask += [0 if mask_padding_with_zero else 1] * padding_length
segment_ids += [pad_token_segment_id] * padding_length
label_ids += [pad_token_label_id] * padding_length
assert len(input_ids) == max_seq_length
assert len(input_mask) == max_seq_length
assert len(segment_ids) == max_seq_length
assert len(label_ids) == max_seq_length
if ex_index < 5:
logger.info("*** Example ***")
logger.info("guid: %s", example.guid)
logger.info("tokens: %s", " ".join([str(x) for x in tokens]))
logger.info("input_ids: %s", " ".join([str(x) for x in input_ids]))
logger.info("input_mask: %s", " ".join([str(x) for x in input_mask]))
logger.info("segment_ids: %s", " ".join([str(x) for x in segment_ids]))
logger.info("label_ids: %s", " ".join([str(x) for x in label_ids]))
features.append(
InputFeatures(input_ids=input_ids, input_mask=input_mask, segment_ids=segment_ids, label_ids=label_ids)
)
return features
def get_labels(path):
if path:
with open(path, "r") as f:
labels = f.read().splitlines()
if "O" not in labels:
labels = ["O"] + labels
return labels
else:
return ["O", "B-MISC", "I-MISC", "B-PER", "I-PER", "B-ORG", "I-ORG", "B-LOC", "I-LOC"]