| from transformers import AutoTokenizer, AutoModelForSequenceClassification, Trainer, TrainingArguments |
| from datasets import load_dataset |
| import torch |
|
|
| |
| dataset = load_dataset('glue', 'mrpc') |
|
|
| |
| tokenizer = AutoTokenizer.from_pretrained('bert-base-uncased') |
| model = AutoModelForSequenceClassification.from_pretrained('bert-base-uncased', num_labels=2) |
|
|
| |
| def tokenize_function(examples): |
| return tokenizer(examples['sentence1'], examples['sentence2'], truncation=True, padding='max_length', max_length=128) |
|
|
| tokenized_datasets = dataset.map(tokenize_function, batched=True) |
|
|
| |
| training_args = TrainingArguments( |
| output_dir='./results', |
| evaluation_strategy='epoch', |
| learning_rate=2e-5, |
| per_device_train_batch_size=8, |
| per_device_eval_batch_size=8, |
| num_train_epochs=3, |
| weight_decay=0.01, |
| ) |
|
|
| trainer = Trainer( |
| model=model, |
| args=training_args, |
| train_dataset=tokenized_datasets['train'], |
| eval_dataset=tokenized_datasets['validation'], |
| ) |
|
|
| |
| trainer.train() |
|
|
| |
| model.save_pretrained('./my_model') |
| tokenizer.save_pretrained('./my_model') |
|
|
| |
| model = AutoModelForSequenceClassification.from_pretrained('./my_model') |
| tokenizer = AutoTokenizer.from_pretrained('./my_model') |
|
|
| |
| def predict(text1, text2): |
| inputs = tokenizer(text1, text2, return_tensors='pt', truncation=True, padding='max_length', max_length=128) |
| outputs = model(**inputs) |
| logits = outputs.logits |
| predictions = torch.argmax(logits, dim=-1).item() |
| return "Similar" if predictions == 0 else "Not Similar" |
|
|
| |
| if __name__ == "__main__": |
| import gradio as gr |
|
|
| iface = gr.Interface( |
| fn=predict, |
| inputs=["text", "text"], |
| outputs="text", |
| title="MRPC Sentiment Analysis", |
| description="Determine if two sentences are semantically equivalent." |
| ) |
|
|
| iface.launch() |
