Huggingface Transformer教程(一)

本系列文章介绍Huggingface Transformer的用法。

简介

简介

from transformers import BertForSequenceClassification, BertTokenizerFast, Trainer, TrainingArguments
from nlp import load_dataset
import torch
import numpy as np
from sklearn.metrics import accuracy_score, precision_recall_fscore_support

model = BertForSequenceClassification.from_pretrained('bert-base-uncased')
tokenizer = BertTokenizerFast.from_pretrained('bert-base-uncased')


def tokenize(batch):
    return tokenizer(batch['text'], padding=True, truncation=True)

train_dataset, test_dataset = load_dataset('imdb', split=['train', 'test'])
train_dataset = train_dataset.map(tokenize, batched=True, batch_size=len(train_dataset))
test_dataset = test_dataset.map(tokenize, batched=True, batch_size=len(train_dataset))
train_dataset.set_format('torch', columns=['input_ids', 'attention_mask', 'label'])
test_dataset.set_format('torch', columns=['input_ids', 'attention_mask', 'label'])


def compute_metrics(pred):
    labels = pred.label_ids
    preds = pred.predictions.argmax(-1)
    precision, recall, f1, _ = precision_recall_fscore_support(labels, preds, average='binary')
    acc = accuracy_score(labels, preds)
    return {
        'accuracy': acc,
        'f1': f1,
        'precision': precision,
        'recall': recall
    }

training_args = TrainingArguments(
    output_dir='./results',
    num_train_epochs=1,
    per_device_train_batch_size=2,
    per_device_eval_batch_size=2,
    warmup_steps=500,
    weight_decay=0.01,
    #evaluate_during_training=True,
    evaluation_strategy="epoch",
    save_total_limit=3,
    logging_dir='./logs',
    load_best_model_at_end=True
)

trainer = Trainer(
    model=model,
    args=training_args,
    compute_metrics=compute_metrics,
    train_dataset=train_dataset,
    eval_dataset=test_dataset
)


trainer.train()

trainer.evaluate()

结果：

{'eval_loss': 0.40767231583595276, 'eval_accuracy': 0.92072, 'eval_f1': 0.9225599749941392, 'eval_precision': 0.9016343363372537, 'eval_recall': 0.94448, 'eval_runtime': 1495.5637, 'eval_samples_per_second': 16.716, 'epoch': 1.0}

FEATURED TAGS

人工智能深度学习 chatbot PyTorch Java BERT git 编程 OCR 汪曾祺语音识别 Kaldi Linux XLNet 情感分析 sentiment analysis 语法纠错 Transformer Tensorflow Huggingface Ubuntu TensorFlow 深度学习框架 Tensor2Tensor 机器翻译微信 wechat automation selenium webdriver pywinauto CentOS GPU Appium t2t 代码阅读中英翻译公众号爬虫 ocr tesseract pytesseract python 默认参数位置参数 VPN JSON Jackson huggingface RoPE PagedAttention vLLM Pre-training LLM CPT weather forecasting graph neural networks qlora quantization transformers cmake pip pipenv conda padding vscode debug source code build deep learning Speech ASR linux pytorch c++ extension Deep Learning DeepSeek Attention Positional Embedding MoE Reinforcement Learning cs336 bpe tokenizer

FRIENDS

Li Li