AnswerDotAI / RAGatouille

Easily use and train state of the art late-interaction retrieval methods (ColBERT) in any RAG pipeline. Designed for modularity and ease-of-use, backed by research.
Apache License 2.0
3.08k stars 210 forks source link

can't access my finetuned model #223

Open GP24CairoUni opened 5 months ago

GP24CairoUni commented 5 months ago

from ragatouille import RAGTrainer

trainer = RAGTrainer(model_name="MyFineTunedColBERT", pretrained_model_name="colbert-ir/colbertv2.0")

trainer.prepare_training_data(raw_data=positive_pairs, data_out_path="./data2/", all_documents=cleaned_documents, mine_hard_negatives=True)

training_path = trainer.train(batch_size=16, nbits=4, maxsteps=500000, use_ib_negatives=True, dim=128, learning_rate=5e-6, doc_maxlen=256, use_relu=False, warmup_steps="auto")

image

I finished finetuning but I cant access and use the model again and no checkpoints on my local storage

abdelkareemkobo commented 3 months ago

may be try to export it into the huggingface hub and then download it again, but didn't find it in the path is extremely weird.