htlou/backup_0202_llamafactory_Compcap_cooccur_0_40-llava-mistral

Compcap_cooccur_0_40

This model is a fine-tuned version of llava-hf/llava-v1.6-mistral-7b-hf on the Compcap_cooccur_0_40 dataset. It achieves the following results on the evaluation set:

• Loss: 0.8389

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: 1e-06
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 4
• total_train_batch_size: 256
• total_eval_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 10
• num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss
0.9458	0.3384	50	0.9426
0.8779	0.6768	100	0.8877
0.82	1.0152	150	0.8625
0.7763	1.3536	200	0.8518
0.7719	1.6920	250	0.8421
0.7347	2.0305	300	0.8381
0.7232	2.3689	350	0.8402
0.721	2.7073	400	0.8391

Framework versions

• Transformers 4.45.2
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.20.3