videomae-base-finetuned-ASBD_ESBD

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: 1.4178
• Accuracy: 0.5714

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: 12
• eval_batch_size: 12
• 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: 500

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.504	0.02	11	1.3929	0.3433
1.3539	1.02	22	1.3380	0.1940
1.2957	2.02	33	1.3519	0.2239
1.2368	3.02	44	1.3220	0.3881
1.1561	4.02	55	1.2803	0.3134
1.0195	5.02	66	1.2588	0.5373
0.8594	6.02	77	1.1591	0.5075
0.8756	7.02	88	0.9532	0.6119
0.6488	8.02	99	1.1922	0.5373
0.4427	9.02	110	0.9780	0.6567
0.3975	10.02	121	1.3228	0.5373
0.3978	11.02	132	1.2083	0.6418
0.2859	12.02	143	1.0027	0.7463
0.3441	13.02	154	1.3718	0.5821
0.2239	14.02	165	1.4324	0.5821
0.2275	15.02	176	1.1823	0.6418
0.1734	16.02	187	1.5484	0.6119
0.2451	17.02	198	1.3764	0.5821
0.1317	18.02	209	1.3731	0.6716
0.0778	19.02	220	1.3567	0.7164
0.1963	20.02	231	1.0905	0.7164
0.1474	21.02	242	2.1361	0.4627
0.0487	22.02	253	1.2189	0.7164
0.0699	23.02	264	1.7618	0.5970
0.1576	24.02	275	1.1939	0.7463
0.0377	25.02	286	1.2287	0.7313
0.0674	26.02	297	1.5247	0.6567
0.0188	27.02	308	1.7585	0.6567
0.0681	28.02	319	1.7868	0.6567
0.0341	29.02	330	1.3745	0.6567
0.05	30.02	341	1.8781	0.6418
0.0269	31.02	352	1.9228	0.5970
0.0213	32.02	363	1.8014	0.6119
0.0061	33.02	374	1.4477	0.6866
0.0338	34.02	385	1.5303	0.6567
0.0086	35.02	396	1.5219	0.7015
0.0891	36.02	407	1.8414	0.5821
0.0032	37.02	418	1.8731	0.5821
0.0028	38.02	429	1.6881	0.6418
0.0434	39.02	440	1.7288	0.6567
0.0018	40.02	451	1.8235	0.6119
0.0232	41.02	462	1.8903	0.6119
0.0016	42.02	473	1.9292	0.6119
0.0016	43.02	484	1.9059	0.6119
0.0029	44.02	495	1.8093	0.6418
0.0519	45.01	500	1.8045	0.6418

Framework versions

• Transformers 4.28.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.4
• Tokenizers 0.13.3