videomae-base-finetuned-ucf101-subset-hand

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: 0.5545
• 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: 3
• eval_batch_size: 3
• 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: 1600

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.6621	11.0006	100	1.6555	0.2
1.6996	22.0012	200	1.5660	0.4167
0.886	33.0019	300	1.4698	0.4167
0.6728	44.0025	400	0.7168	0.7833
0.1659	55.0031	500	1.3954	0.6667
0.1471	66.0037	600	1.7320	0.6
0.0085	77.0044	700	1.4034	0.7333
0.0613	88.005	800	1.1479	0.7
0.1256	99.0056	900	1.3657	0.7
0.194	111.0006	1000	1.0879	0.7
0.1336	122.0012	1100	0.9304	0.8
0.0605	133.0019	1200	0.6773	0.85
0.0025	144.0025	1300	0.7631	0.7833
0.0016	155.0031	1400	0.4629	0.9167
0.1698	166.0037	1500	0.5422	0.9167
0.0011	177.0044	1600	0.5545	0.9

Framework versions

• Transformers 4.45.0
• Pytorch 2.4.1+cu118
• Datasets 3.0.0
• Tokenizers 0.20.0