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_60
results: []
Compcap_cosi_0_60
This model is a fine-tuned version of llava-hf/llava-v1.6-mistral-7b-hf on the Compcap_cosi_0_60 dataset. It achieves the following results on the evaluation set:
- Loss: 0.8026
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 |
|---|---|---|---|
| 0.9416 | 0.2257 | 50 | 0.9273 |
| 0.876 | 0.4515 | 100 | 0.8720 |
| 0.8328 | 0.6772 | 150 | 0.8474 |
| 0.8288 | 0.9029 | 200 | 0.8303 |
| 0.772 | 1.1287 | 250 | 0.8230 |
| 0.7434 | 1.3544 | 300 | 0.8150 |
| 0.7387 | 1.5801 | 350 | 0.8084 |
| 0.7409 | 1.8059 | 400 | 0.8031 |
| 0.7112 | 2.0316 | 450 | 0.8033 |
| 0.7123 | 2.2573 | 500 | 0.8042 |
| 0.7057 | 2.4831 | 550 | 0.8041 |
| 0.6976 | 2.7088 | 600 | 0.8029 |
| 0.6986 | 2.9345 | 650 | 0.8026 |
Framework versions
- Transformers 4.45.2
- Pytorch 2.4.0+cu121
- Datasets 2.21.0
- Tokenizers 0.20.3