metadata
library_name: transformers
license: other
base_model: llava-hf/llava-v1.6-mistral-7b-hf
tags:
- llama-factory
- full
- generated_from_trainer
model-index:
- name: Compcap_cosi_0_80
results: []
Compcap_cosi_0_80
This model is a fine-tuned version of llava-hf/llava-v1.6-mistral-7b-hf on the Compcap_cosi_0_80 dataset. It achieves the following results on the evaluation set:
- Loss: 0.7927
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 |
|---|---|---|---|
| 0.9744 | 0.1692 | 50 | 0.9363 |
| 0.8749 | 0.3384 | 100 | 0.8795 |
| 0.8423 | 0.5076 | 150 | 0.8532 |
| 0.8269 | 0.6768 | 200 | 0.8365 |
| 0.8223 | 0.8460 | 250 | 0.8226 |
| 0.7651 | 1.0152 | 300 | 0.8149 |
| 0.7388 | 1.1844 | 350 | 0.8108 |
| 0.7429 | 1.3536 | 400 | 0.8051 |
| 0.7481 | 1.5228 | 450 | 0.8002 |
| 0.7308 | 1.6920 | 500 | 0.7954 |
| 0.7306 | 1.8613 | 550 | 0.7920 |
| 0.7065 | 2.0305 | 600 | 0.7943 |
| 0.695 | 2.1997 | 650 | 0.7947 |
| 0.7013 | 2.3689 | 700 | 0.7939 |
| 0.6743 | 2.5381 | 750 | 0.7932 |
| 0.6778 | 2.7073 | 800 | 0.7929 |
| 0.6951 | 2.8765 | 850 | 0.7927 |
Framework versions
- Transformers 4.45.2
- Pytorch 2.4.0+cu121
- Datasets 2.21.0
- Tokenizers 0.20.3