videomae-base-finetuned-ucf101-subset-SBDtoy

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: 1.8346
• Accuracy: 0.6275

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: 1200

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.7634	0.03	41	0.6320	0.5882
0.7445	1.03	82	1.4198	0.5686
0.4931	2.03	123	1.3689	0.5686
0.7206	3.03	164	1.3282	0.5882
0.735	4.03	205	2.0849	0.5686
1.769	5.03	246	1.0332	0.6078
0.9494	6.03	287	2.1307	0.6078
0.3857	7.03	328	2.5985	0.6078
0.3639	8.03	369	2.2300	0.6078
0.6456	9.03	410	1.8992	0.6078
0.9483	10.03	451	1.9556	0.6078
0.4518	11.03	492	1.8346	0.6275
0.9109	12.03	533	1.9937	0.6078
0.5441	13.03	574	1.4013	0.6078
0.4035	14.03	615	1.9078	0.6078
0.1713	15.03	656	2.0803	0.5882
0.0542	16.03	697	2.5432	0.6078
0.3084	17.03	738	2.5753	0.6078
0.2476	18.03	779	2.4253	0.6275
0.006	19.03	820	2.4320	0.6078
0.0021	20.03	861	2.7182	0.6078
0.008	21.03	902	2.7711	0.5882
0.0011	22.03	943	2.8089	0.5882
0.7546	23.03	984	2.8142	0.6078
0.0006	24.03	1025	2.8796	0.6078
0.0008	25.03	1066	2.8486	0.6078
0.1745	26.03	1107	2.8475	0.6078
0.5421	27.03	1148	2.8462	0.6078
0.1233	28.03	1189	2.8324	0.6078
0.1298	29.01	1200	2.8332	0.6078

Framework versions

• Transformers 4.38.2
• Pytorch 2.0.1
• Datasets 2.18.0
• Tokenizers 0.15.2