CTMAE2_CS_V7_4

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: 0.3363
• Accuracy: 0.8667

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
0.6652	0.0201	156	0.7404	0.4667
0.5679	1.0201	312	0.6957	0.4667
0.4455	2.0201	468	0.4729	0.7778
0.479	3.0201	624	0.8238	0.6
0.4476	4.0201	780	0.5751	0.7111
0.5195	5.0201	936	0.8028	0.5333
0.4507	6.0201	1092	0.5427	0.6889
0.8518	7.0201	1248	0.5461	0.6889
0.5845	8.0201	1404	0.3363	0.8667
0.3814	9.0201	1560	0.4834	0.7111
0.3419	10.0201	1716	0.9276	0.6444
0.2682	11.0201	1872	1.0908	0.6667
0.272	12.0201	2028	0.7781	0.7333
0.313	13.0201	2184	1.4005	0.6444
0.2476	14.0201	2340	0.9079	0.7111
0.6739	15.0201	2496	0.7558	0.7333
0.3332	16.0201	2652	0.3408	0.8667
0.4414	17.0201	2808	0.4559	0.8667
0.5799	18.0201	2964	0.3860	0.8444
0.3393	19.0201	3120	1.2055	0.6889
0.1268	20.0201	3276	0.5242	0.8444
0.1138	21.0201	3432	0.4627	0.8444
0.6119	22.0201	3588	0.8547	0.7556
0.1607	23.0201	3744	0.7664	0.8222
0.1289	24.0201	3900	0.7742	0.8
0.3687	25.0201	4056	0.7859	0.8
0.1596	26.0201	4212	0.6785	0.8222
0.1539	27.0201	4368	0.7018	0.8
0.0724	28.0201	4524	0.5837	0.8667
0.1998	29.0201	4680	0.9271	0.8222
0.2286	30.0201	4836	1.7057	0.7111
0.192	31.0201	4992	0.7722	0.8222
0.1149	32.0201	5148	0.9803	0.8
0.1723	33.0201	5304	0.9823	0.8
0.0624	34.0201	5460	1.1380	0.8
0.3264	35.0201	5616	0.9571	0.8222
0.0631	36.0201	5772	1.4243	0.7778
0.1043	37.0201	5928	0.8813	0.8444
0.1337	38.0201	6084	1.6177	0.7333
0.271	39.0201	6240	1.1881	0.7778
0.3917	40.0201	6396	1.2265	0.7778
0.3894	41.0201	6552	1.1420	0.8
0.0459	42.0201	6708	1.0437	0.8222
0.1631	43.0201	6864	1.2080	0.8
0.0029	44.0201	7020	1.1999	0.8
0.1862	45.0201	7176	1.2467	0.8
0.0266	46.0201	7332	1.3164	0.7778
0.0013	47.0201	7488	1.2166	0.8
0.0007	48.0201	7644	1.2552	0.8
0.0089	49.0137	7750	1.2637	0.8

Framework versions

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