This model is a fine-tuned version of llava-hf/llava-v1.6-mistral-7b-hf on the Compcap_cosi_0_70 dataset. It achieves the following results on the evaluation set:

Loss: 0.7919

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
0.9353	0.1934	50	0.9258
0.8526	0.3868	100	0.8687
0.8383	0.5803	150	0.8432
0.8264	0.7737	200	0.8269
0.825	0.9671	250	0.8141
0.7583	1.1605	300	0.8098
0.7459	1.3540	350	0.8036
0.7534	1.5474	400	0.7978
0.7429	1.7408	450	0.7935
0.7409	1.9342	500	0.7897
0.6927	2.1277	550	0.7949
0.6889	2.3211	600	0.7934
0.7015	2.5145	650	0.7928
0.6837	2.7079	700	0.7921
0.687	2.9014	750	0.7919

Framework versions

Transformers 4.45.2
Pytorch 2.4.0+cu121
Datasets 2.21.0
Tokenizers 0.20.3

htlou/backup_0202_llamafactory_Compcap_cosi_0_70-llava-mistral

Compcap_cosi_0_70

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

htlou/
backup_0202_llamafactory_Compcap_cosi_0_70-llava-mistral