videomae-base-short-ssv2-finetuned-ct_cpc

This model is a fine-tuned version of MCG-NJU/videomae-base-short-ssv2 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.1547
• Accuracy: 0.6508

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: 1e-05
• train_batch_size: 6
• eval_batch_size: 6
• 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: 2300

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.4832	0.02	47	1.1203	0.6667
1.32	1.02	94	1.0505	0.6667
1.1539	2.02	141	1.0525	0.6667
1.2399	3.02	188	1.0354	0.6667
0.9928	4.02	235	1.0516	0.6667
1.0054	5.02	282	1.0297	0.6667
1.1166	6.02	329	1.0247	0.6667
1.2715	7.02	376	1.0098	0.6667
1.0317	8.02	423	1.0404	0.6667
1.1258	9.02	470	1.0054	0.6667
0.8991	10.02	517	1.0477	0.6667
0.8957	11.02	564	1.0608	0.6667
1.078	12.02	611	1.0166	0.6667
0.9067	13.02	658	1.2110	0.6667
1.0383	14.02	705	1.0380	0.6
1.2244	15.02	752	1.1041	0.6667
1.0364	16.02	799	1.0123	0.6
0.9662	17.02	846	1.1803	0.6333
0.9299	18.02	893	1.0910	0.6333
0.877	19.02	940	1.0183	0.65
0.8649	20.02	987	1.1415	0.6167
0.8841	21.02	1034	1.0975	0.6333
0.6276	22.02	1081	1.0728	0.6
0.7974	23.02	1128	1.1676	0.5833
0.7381	24.02	1175	1.1624	0.55
0.4807	25.02	1222	1.1930	0.55
0.6812	26.02	1269	1.1511	0.6
0.5802	27.02	1316	1.2315	0.6167
0.5943	28.02	1363	1.1641	0.5667
0.5416	29.02	1410	1.0963	0.6
0.6676	30.02	1457	1.1131	0.5667
0.4085	31.02	1504	1.2948	0.5833
0.4548	32.02	1551	1.2788	0.6333
0.4351	33.02	1598	1.1922	0.5667
0.5641	34.02	1645	1.2677	0.55
0.4471	35.02	1692	1.2766	0.6167
0.361	36.02	1739	1.2525	0.55
0.5668	37.02	1786	1.2928	0.55
0.5262	38.02	1833	1.2429	0.6333
0.4901	39.02	1880	1.1457	0.5667
0.481	40.02	1927	1.2137	0.5833
0.3566	41.02	1974	1.1830	0.6167
0.5147	42.02	2021	1.1629	0.6
0.4013	43.02	2068	1.2735	0.6
0.3895	44.02	2115	1.2758	0.6167
0.3727	45.02	2162	1.2519	0.5833
0.2379	46.02	2209	1.2318	0.5833
0.2985	47.02	2256	1.2594	0.5833
0.2991	48.02	2300	1.2510	0.5833

Framework versions

• Transformers 4.37.2
• Pytorch 2.0.1+cu117
• Datasets 3.0.1
• Tokenizers 0.15.1