gm-lora-bfloat16-idefics2-8b-xrayvqa-finetuned-mimic-report

This model is a fine-tuned version of HuggingFaceM4/idefics2-8b on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.8581

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 10
• total_train_batch_size: 80
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 50
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss
0.1649	0.1932	50	0.8254
0.1768	0.3864	100	0.7980
0.1822	0.5796	150	0.7919
0.2301	0.7728	200	0.7679
0.2153	0.9660	250	0.7749
0.1555	1.1592	300	0.8227
0.1456	1.3524	350	0.8360
0.1407	1.5456	400	0.8407
0.1356	1.7388	450	0.8560
0.134	1.9320	500	0.8581

Framework versions

• Transformers 4.41.0.dev0
• Pytorch 2.2.1+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1