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_50
results: []
Compcap_cosi_0_50
This model is a fine-tuned version of llava-hf/llava-v1.6-mistral-7b-hf on the Compcap_cosi_0_50 dataset. It achieves the following results on the evaluation set:
- Loss: 0.8178
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 |
|---|---|---|---|
| 0.9375 | 0.2706 | 50 | 0.9355 |
| 0.8824 | 0.5413 | 100 | 0.8776 |
| 0.8549 | 0.8119 | 150 | 0.8518 |
| 0.7825 | 1.0825 | 200 | 0.8398 |
| 0.759 | 1.3532 | 250 | 0.8301 |
| 0.7676 | 1.6238 | 300 | 0.8219 |
| 0.7624 | 1.8945 | 350 | 0.8160 |
| 0.6976 | 2.1651 | 400 | 0.8193 |
| 0.7004 | 2.4357 | 450 | 0.8186 |
| 0.7133 | 2.7064 | 500 | 0.8176 |
| 0.7 | 2.9770 | 550 | 0.8178 |
Framework versions
- Transformers 4.45.2
- Pytorch 2.4.0+cu121
- Datasets 2.21.0
- Tokenizers 0.20.3