CTMAE2_CS_V7_7

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

• Loss: 1.4462
• Accuracy: 0.8222

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: 5
• eval_batch_size: 5
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 7750

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.675	0.0201	156	0.8126	0.4667
0.6306	1.0201	312	0.8617	0.4667
0.5345	2.0201	468	0.6569	0.6222
0.36	3.0201	624	0.6944	0.6444
0.4708	4.0201	780	1.5217	0.5333
0.274	5.0201	936	0.7516	0.6
0.4109	6.0201	1092	0.7190	0.5778
0.639	7.0201	1248	0.8166	0.6444
0.5028	8.0201	1404	0.6623	0.7778
0.3825	9.0201	1560	0.5929	0.7111
0.4497	10.0201	1716	0.6088	0.7556
0.5095	11.0201	1872	1.6115	0.5556
0.3282	12.0201	2028	1.2386	0.5778
0.5552	13.0201	2184	1.5737	0.6
0.3099	14.0201	2340	1.4204	0.6
0.2961	15.0201	2496	1.5614	0.5333
0.3876	16.0201	2652	0.9776	0.7778
0.2942	17.0201	2808	1.1406	0.7778
0.4692	18.0201	2964	1.2078	0.7778
0.1124	19.0201	3120	1.3382	0.7333
0.3766	20.0201	3276	1.0293	0.8
0.2437	21.0201	3432	1.3097	0.7111
0.3116	22.0201	3588	1.4258	0.7111
0.0411	23.0201	3744	1.1758	0.7556
0.2582	24.0201	3900	1.5001	0.7333
0.0351	25.0201	4056	1.4352	0.7778
0.2065	26.0201	4212	1.3168	0.7556
0.4435	27.0201	4368	1.2661	0.8
0.0559	28.0201	4524	1.2387	0.7778
0.3598	29.0201	4680	1.4478	0.7333
0.0103	30.0201	4836	1.8970	0.6444
0.1988	31.0201	4992	1.3777	0.8
0.1377	32.0201	5148	1.8340	0.7333
0.281	33.0201	5304	1.5446	0.7778
0.1181	34.0201	5460	1.5648	0.7778
0.194	35.0201	5616	1.5953	0.7333
0.0012	36.0201	5772	1.6369	0.7556
0.2422	37.0201	5928	1.8253	0.7556
0.2209	38.0201	6084	1.4462	0.8222
0.2681	39.0201	6240	1.9831	0.6889
0.3073	40.0201	6396	1.5197	0.7556
0.1418	41.0201	6552	1.9243	0.7333
0.0021	42.0201	6708	1.6636	0.7556
0.0029	43.0201	6864	2.1923	0.6667
0.2099	44.0201	7020	1.7219	0.7556
0.0003	45.0201	7176	2.0043	0.6889
0.0269	46.0201	7332	1.8475	0.7333
0.0008	47.0201	7488	1.8957	0.7333
0.0003	48.0201	7644	1.8716	0.7556
0.1757	49.0137	7750	1.8668	0.7556

Framework versions

• Transformers 4.46.2
• Pytorch 2.0.1+cu117
• Datasets 3.0.1
• Tokenizers 0.20.0