CTMAE2_CS_V7_5

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: 0.3647
• 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: 4
• eval_batch_size: 4
• 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: 9700

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6793	0.0201	195	0.7616	0.4667
0.5591	1.0201	390	0.7040	0.4667
0.7211	2.0201	585	0.4916	0.8222
0.5544	3.0201	780	0.7590	0.5556
0.6032	4.0201	975	0.5508	0.6667
0.518	5.0201	1170	0.8928	0.4667
0.4857	6.0201	1365	0.5889	0.6222
0.3634	7.0201	1560	0.8523	0.6444
0.4082	8.0201	1755	0.3647	0.8667
0.4897	9.0201	1950	0.5648	0.7778
0.389	10.0201	2145	0.5736	0.7778
0.3753	11.0201	2340	1.0849	0.5778
0.3118	12.0201	2535	1.0598	0.6222
0.5823	13.0201	2730	0.7086	0.7333
0.2604	14.0201	2925	1.4168	0.6222
0.5767	15.0201	3120	0.7966	0.8
0.4844	16.0201	3315	1.0488	0.7333
0.0729	17.0201	3510	1.0075	0.7333
0.4188	18.0201	3705	1.2724	0.7111
0.247	19.0201	3900	1.3884	0.7556
0.8041	20.0201	4095	1.1552	0.7333
0.168	21.0201	4290	1.7924	0.6889
0.2564	22.0201	4485	1.5682	0.7333
0.2034	23.0201	4680	1.6061	0.6889
0.436	24.0201	4875	1.5508	0.6889
0.379	25.0201	5070	1.6198	0.7111
0.0726	26.0201	5265	2.3293	0.6
0.0099	27.0201	5460	1.7658	0.7333
0.0346	28.0201	5655	1.5937	0.7111
0.0058	29.0201	5850	2.3511	0.6444
0.1163	30.0201	6045	1.7068	0.7333
0.0962	31.0201	6240	1.8767	0.6889
0.2826	32.0201	6435	2.1657	0.7111
0.1249	33.0201	6630	1.7385	0.7333
0.2191	34.0201	6825	2.1789	0.7111
0.0958	35.0201	7020	2.4722	0.6444
0.0006	36.0201	7215	1.9177	0.7111
0.0036	37.0201	7410	1.9591	0.6889
0.0009	38.0201	7605	2.3993	0.6222
0.0005	39.0201	7800	1.7378	0.7778
0.0014	40.0201	7995	2.4454	0.6889
0.1203	41.0201	8190	2.1138	0.7333
0.0138	42.0201	8385	2.1769	0.7333
0.3569	43.0201	8580	2.6946	0.6222
0.0002	44.0201	8775	2.1566	0.7111
0.0924	45.0201	8970	2.4636	0.6667
0.0004	46.0201	9165	2.1076	0.7333
0.0645	47.0201	9360	2.1503	0.7111
0.1121	48.0201	9555	2.2611	0.7111
0.1268	49.0149	9700	2.1728	0.7111

Framework versions

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