5c_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: 4.4509
• Accuracy: 0.48

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: 1
• eval_batch_size: 1
• 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: 23400

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.9641	0.01	234	1.4838	0.4
1.5439	1.01	468	3.7125	0.4
1.2944	2.01	702	3.6749	0.4
0.9419	3.01	936	3.0422	0.4
2.4333	4.01	1170	2.6803	0.4
1.4646	5.01	1404	3.5355	0.4
2.1201	6.01	1638	3.0479	0.4
2.9021	7.01	1872	2.8181	0.4
2.1527	8.01	2106	2.7605	0.4
1.9428	9.01	2340	2.4513	0.4
1.6949	10.01	2574	3.2310	0.4
0.7839	11.01	2808	3.2372	0.4
0.3228	12.01	3042	4.4588	0.4
3.5377	13.01	3276	2.8621	0.4
0.509	14.01	3510	2.7460	0.4
0.1437	15.01	3744	2.9698	0.4
1.0039	16.01	3978	1.9415	0.44
0.0062	17.01	4212	3.7041	0.4
0.6038	18.01	4446	3.2141	0.4
1.1687	19.01	4680	2.4072	0.44
0.8397	20.01	4914	3.4212	0.4
1.1147	21.01	5148	2.5115	0.44
0.2286	22.01	5382	2.4343	0.44
0.8939	23.01	5616	3.0712	0.4
0.3871	24.01	5850	3.2394	0.4
0.3649	25.01	6084	3.9466	0.44
1.2601	26.01	6318	2.9586	0.44
0.852	27.01	6552	4.6464	0.4
0.6269	28.01	6786	3.1292	0.44
1.0013	29.01	7020	4.6319	0.4
0.02	30.01	7254	4.2514	0.4
0.1333	31.01	7488	4.3310	0.4
0.0005	32.01	7722	4.5354	0.4
0.004	33.01	7956	4.5970	0.4
0.3017	34.01	8190	4.5879	0.44
0.2014	35.01	8424	4.2809	0.4
0.1573	36.01	8658	4.6822	0.44
0.0041	37.01	8892	5.1673	0.4
0.0001	38.01	9126	5.4005	0.4
0.1066	39.01	9360	4.4509	0.48
0.0001	40.01	9594	5.0906	0.44
1.3235	41.01	9828	4.4093	0.48
0.4313	42.01	10062	4.0898	0.48
0.0002	43.01	10296	4.7817	0.44
0.0001	44.01	10530	4.8667	0.48
0.0007	45.01	10764	4.5619	0.48
0.0009	46.01	10998	5.0250	0.44
0.0001	47.01	11232	4.4129	0.48
0.0001	48.01	11466	5.5987	0.44
0.0003	49.01	11700	5.4567	0.44
0.0468	50.01	11934	5.0218	0.48
0.187	51.01	12168	5.3269	0.4
0.0002	52.01	12402	5.4364	0.44
0.0001	53.01	12636	5.7307	0.44
0.0	54.01	12870	5.9781	0.44
0.0001	55.01	13104	4.8221	0.44
0.0001	56.01	13338	5.5808	0.4
0.0	57.01	13572	5.7662	0.44
0.0001	58.01	13806	5.4463	0.44
0.0021	59.01	14040	5.9576	0.44
0.5042	60.01	14274	5.9419	0.4
0.0053	61.01	14508	5.2977	0.48
0.0	62.01	14742	5.8541	0.4
0.1555	63.01	14976	6.5367	0.4
0.0081	64.01	15210	5.4808	0.4
0.0008	65.01	15444	5.8818	0.4
0.0	66.01	15678	6.4378	0.4
0.0	67.01	15912	5.6597	0.4
0.0	68.01	16146	5.8197	0.44
0.0061	69.01	16380	6.0141	0.4
0.0001	70.01	16614	6.2449	0.4
0.0001	71.01	16848	6.2530	0.4
0.0	72.01	17082	5.7655	0.4
0.0	73.01	17316	6.1521	0.4
0.0	74.01	17550	6.1597	0.44
0.6123	75.01	17784	6.4786	0.4
0.0	76.01	18018	6.5528	0.4
0.0	77.01	18252	5.5426	0.44
0.0	78.01	18486	6.4276	0.4
0.0	79.01	18720	6.8676	0.4
0.0	80.01	18954	6.6693	0.4
0.0	81.01	19188	6.7919	0.4
0.0	82.01	19422	6.7520	0.4
0.0	83.01	19656	6.7565	0.4
0.0	84.01	19890	6.8186	0.4
0.0	85.01	20124	6.5549	0.4
0.0	86.01	20358	6.7223	0.4
0.0	87.01	20592	6.9096	0.4
0.0	88.01	20826	6.9918	0.4
0.0	89.01	21060	7.2247	0.4
0.0001	90.01	21294	7.2267	0.4
0.0	91.01	21528	6.9826	0.4
0.0	92.01	21762	6.6385	0.4
0.792	93.01	21996	6.4020	0.4
0.0	94.01	22230	6.4453	0.4
0.0	95.01	22464	6.9102	0.4
0.0	96.01	22698	6.9262	0.4
0.0	97.01	22932	6.7757	0.4
0.0	98.01	23166	6.8298	0.4
0.0	99.01	23400	6.8317	0.4

Framework versions

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