|
--- |
|
base_model: openbmb/MiniCPM-V-2_6 |
|
library_name: peft |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: MiniCPM-V-2_6_lora_20240916_155621 |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# MiniCPM-V-2_6_lora_20240916_155621 |
|
|
|
This model is a fine-tuned version of [openbmb/MiniCPM-V-2_6](https://huggingface.co./openbmb/MiniCPM-V-2_6) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.2347 |
|
|
|
## 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: 5e-05 |
|
- train_batch_size: 2 |
|
- eval_batch_size: 2 |
|
- seed: 42 |
|
- distributed_type: multi-GPU |
|
- num_devices: 4 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 32 |
|
- total_eval_batch_size: 8 |
|
- optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine_with_restarts |
|
- lr_scheduler_warmup_ratio: 0.05 |
|
- training_steps: 1000 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:------:|:----:|:---------------:| |
|
| 0.1221 | 0.2430 | 100 | 0.2809 | |
|
| 0.1534 | 0.4860 | 200 | 0.2695 | |
|
| 0.1055 | 0.7290 | 300 | 0.2456 | |
|
| 0.0828 | 0.9721 | 400 | 0.2336 | |
|
| 0.1028 | 1.2151 | 500 | 0.2355 | |
|
| 0.0444 | 1.4581 | 600 | 0.2294 | |
|
| 0.054 | 1.7011 | 700 | 0.2347 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.12.0 |
|
- Transformers 4.40.0 |
|
- Pytorch 2.1.2+cu121 |
|
- Tokenizers 0.19.1 |
