qwen2_5_omni_video_only_backup
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.6012
- Token Acc: 0.8135
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: 2
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- total_eval_batch_size: 8
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.95) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.05
- num_epochs: 3.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Token Acc |
|---|
| 0.9030 | 0.2882 | 50 | 0.9056 | 0.7396 |
| 0.8437 | 0.5764 | 100 | 0.8620 | 0.7503 |
| 0.8076 | 0.8646 | 150 | 0.8290 | 0.7571 |
| 0.6862 | 1.1499 | 200 | 0.7866 | 0.7696 |
| 0.6440 | 1.4380 | 250 | 0.7497 | 0.7784 |
| 0.6254 | 1.7262 | 300 | 0.7001 | 0.7880 |
| 0.5372 | 2.0115 | 350 | 0.6609 | 0.7965 |
| 0.4902 | 2.2997 | 400 | 0.6290 | 0.8059 |
| 0.4615 | 2.5879 | 450 | 0.6076 | 0.8116 |
| 0.4290 | 2.8761 | 500 | 0.6014 | 0.8136 |
| 0.4272 | 3.0 | 522 | 0.6012 | 0.8135 |
Framework versions
- Transformers 4.57.0.dev0
- Pytorch 2.8.0+cu128
- Datasets 4.1.1
- Tokenizers 0.22.1
