metadata
library_name: transformers
license: apache-2.0
base_model: HuggingFaceTB/SmolVLM2-500M-Video-Instruct
tags:
- generated_from_trainer
model-index:
- name: SmolVLM2-500M-Video-Instruct-coco-kaggle
results: []
SmolVLM2-500M-Video-Instruct-coco-kaggle
This model is a fine-tuned version of HuggingFaceTB/SmolVLM2-500M-Video-Instruct on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3318
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: 0.0001
- train_batch_size: 4
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 50
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.397 | 0.1390 | 50 | 0.3987 |
| 0.341 | 0.2780 | 100 | 0.3579 |
| 0.3324 | 0.4170 | 150 | 0.3434 |
| 0.3503 | 0.5559 | 200 | 0.3383 |
| 0.3481 | 0.6949 | 250 | 0.3340 |
| 0.3298 | 0.8339 | 300 | 0.3320 |
| 0.3248 | 0.9729 | 350 | 0.3318 |
Framework versions
- Transformers 4.52.0.dev0
- Pytorch 2.7.0+cu126
- Datasets 3.6.0
- Tokenizers 0.21.1