metadata

license: cc-by-nc-4.0
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: videomae-base-finetuned-basketball-subset-v3-25epoch
    results: []

videomae-base-finetuned-basketball-subset-v3-25epoch

This model is a fine-tuned version of MCG-NJU/videomae-base on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.8922
Accuracy: 0.9

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: 5100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.2489	0.04	202	0.7252	0.7
0.5366	1.04	404	1.2745	0.6
0.9659	2.04	606	0.7013	0.85
0.0226	3.04	808	1.3065	0.7
0.5437	4.04	1010	2.0397	0.7
0.0002	5.04	1212	1.8936	0.75
0.0003	6.04	1414	1.4473	0.8
0.0193	7.04	1616	1.1602	0.75
0.0001	8.04	1818	0.8922	0.9
0.0001	9.04	2020	1.0781	0.85
0.0	10.04	2222	1.1948	0.85
0.0	11.04	2424	1.2431	0.85
0.0	12.04	2626	1.2794	0.85
0.0	13.04	2828	1.3082	0.85
0.0	14.04	3030	1.3332	0.85
0.0	15.04	3232	1.3539	0.85
0.0	16.04	3434	1.3793	0.85
0.0	17.04	3636	1.4510	0.8
0.0	18.04	3838	1.5646	0.8
0.0	19.04	4040	1.6535	0.8
0.0	20.04	4242	1.7017	0.8
0.0	21.04	4444	1.7366	0.8
0.0	22.04	4646	1.7639	0.8
0.0	23.04	4848	1.7792	0.8
0.0	24.04	5050	1.7855	0.8
0.0	25.01	5100	1.7857	0.8

Framework versions

Transformers 4.26.1
Pytorch 1.12.1+cu116
Datasets 2.4.0
Tokenizers 0.12.1