metadata
license: mit
base_model: google/vivit-b-16x2-kinetics400
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: vivit-surf-analytics-runpod
results: []
vivit-surf-analytics-runpod
This model is a fine-tuned version of google/vivit-b-16x2-kinetics400 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.7570
- Accuracy: 0.9023
- F1: 0.9016
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: 1
- eval_batch_size: 1
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- training_steps: 741
Training results
| Training Loss | Epoch | Step | Accuracy | F1 | Validation Loss |
|---|---|---|---|---|---|
| 1.3533 | 0.05 | 741 | 0.8393 | 0.8382 | 0.7031 |
| 0.0028 | 1.05 | 1482 | 0.8482 | 0.8460 | 0.7500 |
| 0.0021 | 2.05 | 2223 | 0.8839 | 0.8821 | 0.5604 |
| 0.0002 | 3.05 | 2964 | 0.9018 | 0.9001 | 0.3880 |
| 0.0001 | 4.05 | 3705 | 0.9286 | 0.9284 | 0.4309 |
| 0.0001 | 5.05 | 4446 | 0.9107 | 0.9105 | 0.7365 |
| 0.8987 | 6.05 | 5187 | 0.8393 | 0.8294 | 0.9310 |
| 0.4888 | 7.05 | 5928 | 0.875 | 0.8703 | 0.8563 |
| 0.0001 | 8.05 | 6669 | 0.8929 | 0.8894 | 0.6909 |
| 0.0018 | 9.05 | 7410 | 0.8929 | 0.8917 | 0.9169 |
| 0.0 | 10.05 | 8151 | 0.8929 | 0.8928 | 0.6104 |
| 0.0 | 11.05 | 8892 | 0.9196 | 0.9207 | 0.6125 |
| 0.0 | 12.05 | 9633 | 0.9286 | 0.9281 | 0.5644 |
| 0.0 | 13.05 | 10374 | 0.9286 | 0.9286 | 0.5062 |
| 0.0 | 14.05 | 11115 | 0.9375 | 0.9373 | 0.5186 |
| 0.0 | 15.0013 | 11116 | 0.7569 | 0.9023 | 0.9016 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.1+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1