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_l0_0_70
results: []
Compcap_l0_0_70
This model is a fine-tuned version of llava-hf/llava-v1.6-mistral-7b-hf on the Compcap_l0_0_70 dataset. It achieves the following results on the evaluation set:
- Loss: 0.7935
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 |
|---|---|---|---|
| 0.9541 | 0.1934 | 50 | 0.9273 |
| 0.8621 | 0.3868 | 100 | 0.8719 |
| 0.8657 | 0.5803 | 150 | 0.8459 |
| 0.8232 | 0.7737 | 200 | 0.8287 |
| 0.8179 | 0.9671 | 250 | 0.8162 |
| 0.7558 | 1.1605 | 300 | 0.8122 |
| 0.7602 | 1.3540 | 350 | 0.8059 |
| 0.771 | 1.5474 | 400 | 0.7998 |
| 0.753 | 1.7408 | 450 | 0.7955 |
| 0.7479 | 1.9342 | 500 | 0.7920 |
| 0.6735 | 2.1277 | 550 | 0.7972 |
| 0.7116 | 2.3211 | 600 | 0.7952 |
| 0.6822 | 2.5145 | 650 | 0.7943 |
| 0.6861 | 2.7079 | 700 | 0.7938 |
| 0.6989 | 2.9014 | 750 | 0.7935 |
Framework versions
- Transformers 4.45.2
- Pytorch 2.4.0+cu121
- Datasets 2.21.0
- Tokenizers 0.20.3