videomae-base-finetuned-camera_move-subset

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

• Loss: 2.0146
• Accuracy: 0.7285

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: 4e-05
• train_batch_size: 8
• eval_batch_size: 8
• 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: 13000

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.4019	0.01	131	1.0423	0.5960
1.0428	1.01	262	0.9115	0.6291
0.8694	2.01	393	0.8362	0.6424
0.6894	3.01	524	0.7107	0.7086
0.6741	4.01	655	0.6441	0.7219
0.5955	5.01	786	0.8368	0.6821
0.5898	6.01	917	0.7774	0.7152
0.5252	7.01	1048	0.6022	0.7682
0.453	8.01	1179	0.6388	0.7881
0.4625	9.01	1310	0.8417	0.7152
0.465	10.01	1441	0.8575	0.6954
0.5106	11.01	1572	0.7849	0.7616
0.4138	12.01	1703	0.9293	0.7086
0.3526	13.01	1834	1.2684	0.6225
0.3645	14.01	1965	0.9428	0.7285
0.2778	15.01	2096	1.0217	0.6821
0.2821	16.01	2227	1.3365	0.6623
0.2654	17.01	2358	1.0170	0.7550
0.2961	18.01	2489	1.3952	0.7152
0.1987	19.01	2620	1.4083	0.6954
0.1832	20.01	2751	1.5808	0.6689
0.2035	21.01	2882	1.2864	0.7483
0.1608	22.01	3013	1.7499	0.6755
0.2171	23.01	3144	1.2574	0.7417
0.1824	24.01	3275	1.4203	0.7483
0.152	25.01	3406	1.4163	0.7351
0.1512	26.01	3537	1.5550	0.7086
0.1635	27.01	3668	1.2334	0.7682
0.1371	28.01	3799	1.5798	0.7417
0.1048	29.01	3930	1.6993	0.7219
0.097	30.01	4061	1.8505	0.6887
0.1662	31.01	4192	1.5091	0.7417
0.0957	32.01	4323	1.5388	0.7285
0.0967	33.01	4454	1.6766	0.7219
0.1581	34.01	4585	1.5444	0.7550
0.0967	35.01	4716	1.6108	0.7219
0.1013	36.01	4847	1.6734	0.7417
0.0691	37.01	4978	1.5116	0.7616
0.1161	38.01	5109	1.6343	0.7285
0.0847	39.01	5240	1.6119	0.7616
0.0769	40.01	5371	1.8494	0.7285
0.0604	41.01	5502	1.9854	0.7020
0.0845	42.01	5633	1.9722	0.7020
0.0808	43.01	5764	1.9046	0.7219
0.091	44.01	5895	1.8344	0.7152
0.1249	45.01	6026	1.7306	0.7219
0.0602	46.01	6157	1.8982	0.7550
0.0901	47.01	6288	1.9487	0.6954
0.081	48.01	6419	1.6702	0.7417
0.0336	49.01	6550	1.5591	0.7682
0.0975	50.01	6681	1.9009	0.7086
0.0697	51.01	6812	1.6135	0.7616
0.0474	52.01	6943	1.5886	0.7682
0.0427	53.01	7074	1.6915	0.7616
0.0774	54.01	7205	1.7434	0.7417
0.0685	55.01	7336	1.9552	0.7152
0.0531	56.01	7467	1.7448	0.7550
0.0371	57.01	7598	1.8175	0.7483
0.0244	58.01	7729	1.9252	0.7351
0.0317	59.01	7860	2.1868	0.6954
0.0813	60.01	7991	1.8846	0.7417
0.0857	61.01	8122	1.7532	0.7285
0.0424	62.01	8253	1.8003	0.7550
0.0481	63.01	8384	1.7935	0.7351
0.0495	64.01	8515	2.0369	0.7219
0.0217	65.01	8646	2.0552	0.7219
0.0303	66.01	8777	1.8629	0.7417
0.0439	67.01	8908	1.9953	0.7417
0.0518	68.01	9039	1.9958	0.7351
0.0295	69.01	9170	1.8984	0.7616
0.0485	70.01	9301	1.7583	0.7483
0.0149	71.01	9432	1.8815	0.7351
0.0107	72.01	9563	1.8341	0.7616
0.0475	73.01	9694	1.7961	0.7351
0.0318	74.01	9825	2.0999	0.7285
0.0108	75.01	9956	2.0441	0.7219
0.0085	76.01	10087	1.9867	0.7351
0.0278	77.01	10218	1.7968	0.7682
0.0106	78.01	10349	1.8662	0.7682
0.0241	79.01	10480	1.8388	0.7682
0.0303	80.01	10611	1.8758	0.7682
0.0199	81.01	10742	1.9275	0.7483
0.0052	82.01	10873	1.8741	0.7417
0.007	83.01	11004	2.0243	0.7219
0.0101	84.01	11135	1.9947	0.7285
0.0184	85.01	11266	1.9342	0.7417
0.0062	86.01	11397	2.0611	0.7351
0.018	87.01	11528	1.9867	0.7550
0.0194	88.01	11659	1.8631	0.7483
0.0396	89.01	11790	1.9126	0.7417
0.0048	90.01	11921	1.8788	0.7351
0.0044	91.01	12052	1.8962	0.7351
0.0033	92.01	12183	1.9455	0.7351
0.0052	93.01	12314	1.9772	0.7351
0.0068	94.01	12445	2.0047	0.7351
0.0046	95.01	12576	2.0135	0.7351
0.0158	96.01	12707	2.0176	0.7351
0.0074	97.01	12838	2.0148	0.7285
0.0016	98.01	12969	2.0152	0.7285
0.0051	99.0	13000	2.0146	0.7285

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.13.3