ekstrakcja-bert/bert.py
2021-06-22 22:15:01 +02:00

103 lines
3.0 KiB
Python

import torch
from transformers.file_utils import is_tf_available, is_torch_available
from transformers import BertTokenizerFast, BertForSequenceClassification
from transformers import Trainer, TrainingArguments
import numpy as np
import random
from sklearn.metrics import accuracy_score
import pandas as pd
class Dataset(torch.utils.data.Dataset):
def __init__(self, encodings, labels):
self.encodings = encodings
self.labels = labels
def __getitem__(self, idx):
item = {k: torch.tensor(v[idx]) for k, v in self.encodings.items()}
item["labels"] = torch.tensor([self.labels[idx]])
return item
def __len__(self):
return len(self.labels)
def set_seed(seed: int):
random.seed(seed)
np.random.seed(seed)
if is_torch_available():
torch.manual_seed(seed)
torch.cuda.manual_seed_all(seed)
if is_tf_available():
import tensorflow as tf
tf.random.set_seed(seed)
def compute_metrics(pred):
labels = pred.label_ids
preds = pred.predictions.argmax(-1)
acc = accuracy_score(labels, preds)
return {
'accuracy': acc,
}
def get_prediction(text):
inputs = tokenizer(text, padding=True, truncation=True, max_length=max_length, return_tensors="pt").to("cuda")
outputs = model(**inputs)
return outputs[0].softmax(1).argmax()
set_seed(1)
SAMPLES = 2000
train_texts = \
pd.read_csv('train/in.tsv.xz', compression='xz',
sep='\t', header=None, error_bad_lines=False, quoting=3)[0][:SAMPLES].tolist()
train_labels = pd.read_csv('train/expected.tsv', sep='\t', header=None, quoting=3)[0][:SAMPLES].tolist()
dev_texts = pd.read_csv('dev-0/in.tsv.xz', compression='xz', sep='\t', header=None, quoting=3)[0].tolist()
dev_labels = pd.read_csv('dev-0/expected.tsv', sep='\t', header=None, quoting=3)[0].tolist()
model_name = "bert-base-uncased"
max_length = 512
tokenizer = BertTokenizerFast.from_pretrained(model_name, do_lower_case=True)
train_encodings = tokenizer(train_texts, truncation=True, padding=True, max_length=max_length)
valid_encodings = tokenizer(dev_texts, truncation=True, padding=True, max_length=max_length)
train_dataset = Dataset(train_encodings, train_labels)
valid_dataset = Dataset(valid_encodings, dev_labels)
model = BertForSequenceClassification.from_pretrained(
model_name, num_labels=len(pd.unique(train_labels))).to("cuda")
training_args = TrainingArguments(
output_dir='./results',
num_train_epochs=1,
per_device_train_batch_size=4,
per_device_eval_batch_size=4,
warmup_steps=500,
weight_decay=0.005,
logging_dir='./logs',
load_best_model_at_end=True,
logging_steps=250,
evaluation_strategy="steps",
)
trainer = Trainer(
model=model,
args=training_args,
train_dataset=train_dataset,
eval_dataset=valid_dataset,
compute_metrics=compute_metrics,
)
trainer.train()
trainer.evaluate()
model_path = "bert-base-uncased-2k"
model.save_pretrained(model_path)
tokenizer.save_pretrained(model_path)