Created
March 26, 2024 14:28
-
-
Save mrmaheshrajput/6054f5b3a2830490556256e78fece925 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| import evaluate | |
| from transformers import AutoTokenizer, AutoModelForSeq2SeqLM | |
| from datasets import load_dataset | |
| # Load the fine-tuned model and tokenizer | |
| model_name = "your-fine-tuned-model-name" | |
| tokenizer = AutoTokenizer.from_pretrained(model_name) | |
| model = AutoModelForSeq2SeqLM.from_pretrained(model_name) | |
| # Load the test dataset | |
| test_dataset = load_dataset("your-test-dataset-name") | |
| # Function to generate predictions from model | |
| def generate_predictions(batch): | |
| inputs = tokenizer(batch["input"], padding="max_length", truncation=True, return_tensors="pt") | |
| with torch.no_grad(): | |
| outputs = model.generate(**inputs) | |
| return tokenizer.batch_decode(outputs, skip_special_tokens=True) | |
| # Generate predictions on the test dataset | |
| predictions = test_dataset.map(generate_predictions, batched=True, batch_size=4) | |
| # Convert predictions and references to lists | |
| predictions_list = predictions["predictions"] | |
| references_list = predictions["references"] | |
| # Compute ROUGE scores | |
| rouge_scores = evaluate.compute(predictions_list, references_list) | |
| # Print ROUGE scores | |
| print(rouge_scores) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment