videomae-base-finetuned-SLT-subset

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: 0.4062
• Accuracy: 1.0

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: 2
• eval_batch_size: 2
• 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: 944

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
3.8603	0.06	59	3.7147	0.025
3.7747	1.06	118	3.6984	0.05
3.83	2.06	177	3.5418	0.075
3.5065	3.06	236	3.3917	0.075
3.6541	4.06	295	3.3558	0.1
3.5419	5.06	354	3.2460	0.15
3.2664	6.06	413	3.0603	0.2
3.2295	7.06	472	2.7967	0.425
2.829	8.06	531	2.3743	0.625
2.5769	9.06	590	1.9349	0.675
2.0383	10.06	649	1.3413	0.875
1.8328	11.06	708	0.9239	0.925
1.0447	12.06	767	0.6702	0.975
0.8312	13.06	826	0.5127	1.0
0.7203	14.06	885	0.4366	1.0
0.6204	15.06	944	0.4062	1.0

Framework versions

• Transformers 4.34.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.14.0