app.py

#made by gpt 

from datasets import load_dataset
from transformers import DistilBertTokenizerFast, DistilBertForSequenceClassification, Trainer, TrainingArguments
import torch

# Load a small dataset (IMDB with just a few samples for quick testing)
dataset = load_dataset("imdb", split='train[:2%]').train_test_split(test_size=0.2)

# Tokenizer and model
tokenizer = DistilBertTokenizerFast.from_pretrained("distilbert-base-uncased")
model = DistilBertForSequenceClassification.from_pretrained("distilbert-base-uncased", num_labels=2)

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

tokenized_dataset = dataset.map(tokenize, batched=True)
tokenized_dataset = tokenized_dataset.rename_column("label", "labels")

# Training arguments
training_args = TrainingArguments(
    output_dir="./results",
    evaluation_strategy="epoch",
    per_device_train_batch_size=4,
    per_device_eval_batch_size=4,
    num_train_epochs=1,
    logging_steps=10,
    save_steps=10,
    report_to="none"
)

# Trainer
trainer = Trainer(
    model=model,
    args=training_args,
    train_dataset=tokenized_dataset["train"],
    eval_dataset=tokenized_dataset["test"]
)

# Train the model
trainer.train()

# Save model
trainer.save_model("my-simple-sentiment-model")