videomae-base-finetuned-numbers-augmented2

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: 0.1392
• Accuracy: 0.9681
• F1: 0.9680
• Precision: 0.9692
• Recall: 0.9678

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: 4
• eval_batch_size: 4
• 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: 6756

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
1.3906	0.0833	563	1.2362	0.5628	0.5561	0.6171	0.5623
0.7242	1.0833	1126	0.8677	0.6867	0.6787	0.7247	0.6847
0.5625	2.0833	1689	0.6211	0.7883	0.7865	0.8163	0.7871
0.5111	3.0833	2252	0.4169	0.8623	0.8621	0.8814	0.8631
0.1224	4.0833	2815	0.2908	0.9036	0.9030	0.9077	0.9038
0.0561	5.0833	3378	0.2836	0.9208	0.9207	0.9252	0.9210
0.0028	6.0833	3941	0.2256	0.9466	0.9470	0.9488	0.9471
0.0009	7.0833	4504	0.1670	0.9673	0.9676	0.9702	0.9665
0.0336	8.0833	5067	0.1362	0.9656	0.9656	0.9674	0.9650
0.0004	9.0833	5630	0.1192	0.9776	0.9778	0.9793	0.9771
0.0004	10.0833	6193	0.1204	0.9725	0.9725	0.9745	0.9719
0.0003	11.0833	6756	0.1268	0.9725	0.9723	0.9738	0.9719

Framework versions

• Transformers 4.40.0
• Pytorch 2.1.0+cu121
• Datasets 2.18.0
• Tokenizers 0.19.1