CTMAE2_CS_V7_9

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: 1.0853
• Accuracy: 0.8

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
0.6691	0.0201	156	0.7815	0.4667
0.6103	1.0201	312	0.8786	0.4667
0.5756	2.0201	468	0.9522	0.4667
0.5278	3.0201	624	0.7274	0.5556
0.4309	4.0201	780	0.8362	0.6
0.4808	5.0201	936	0.8308	0.5333
0.373	6.0201	1092	0.6670	0.5778
0.8291	7.0201	1248	0.5672	0.7333
0.5016	8.0201	1404	0.5802	0.7111
0.2954	9.0201	1560	0.8878	0.6222
0.2576	10.0201	1716	0.5753	0.7778
0.4279	11.0201	1872	0.9186	0.6889
0.2818	12.0201	2028	1.3718	0.6
0.4623	13.0201	2184	1.2370	0.7111
0.3438	14.0201	2340	1.6716	0.5778
0.4862	15.0201	2496	0.9000	0.7778
0.2198	16.0201	2652	1.2056	0.6667
0.1955	17.0201	2808	1.0853	0.8
0.2232	18.0201	2964	2.0170	0.6444
0.6413	19.0201	3120	1.3823	0.7111
0.3481	20.0201	3276	0.9786	0.7111
0.1043	21.0201	3432	1.3672	0.7778
0.3651	22.0201	3588	1.5295	0.6444
0.1562	23.0201	3744	1.0870	0.7778
0.137	24.0201	3900	1.9200	0.6889
0.446	25.0201	4056	1.9016	0.6667
0.0023	26.0201	4212	1.6282	0.7333
0.3529	27.0201	4368	1.7609	0.6444
0.384	28.0201	4524	1.6187	0.7333
0.2808	29.0201	4680	2.1017	0.6
0.1582	30.0201	4836	1.7994	0.7333
0.1092	31.0201	4992	1.3328	0.7556
0.2964	32.0201	5148	2.8201	0.5778
0.2952	33.0201	5304	1.4045	0.7778
0.3124	34.0201	5460	1.6641	0.7111
0.0011	35.0201	5616	1.7113	0.7111
0.1375	36.0201	5772	2.0553	0.6667
0.1252	37.0201	5928	2.2712	0.6444
0.1576	38.0201	6084	2.0950	0.6889
0.0008	39.0201	6240	2.9621	0.6
0.0003	40.0201	6396	1.9269	0.7333
0.0002	41.0201	6552	2.2481	0.6889
0.0111	42.0201	6708	2.1604	0.6667
0.3096	43.0201	6864	2.9864	0.6
0.0006	44.0201	7020	2.0592	0.7111
0.0001	45.0201	7176	2.2161	0.7111
0.0002	46.0201	7332	2.2129	0.7111
0.0002	47.0201	7488	2.3814	0.6889
0.0001	48.0201	7644	2.2948	0.7111
0.2534	49.0137	7750	2.3587	0.6889

Framework versions

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