9_mae_1

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: 0.5411
• Accuracy: 0.9348

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
0.6763	0.0201	195	0.8186	0.4565
0.5598	1.0201	390	0.6849	0.4565
0.5409	2.0201	585	0.4701	0.7826
0.4927	3.0201	780	0.4634	0.7826
0.4862	4.0201	975	0.3988	0.8261
0.4784	5.0201	1170	0.6177	0.7391
0.3523	6.0201	1365	0.7854	0.6087
0.4808	7.0201	1560	1.5676	0.4783
0.9314	8.0201	1755	0.3879	0.8478
0.4025	9.0201	1950	0.4501	0.7826
0.47	10.0201	2145	0.4795	0.7826
0.3665	11.0201	2340	0.4920	0.8043
0.3445	12.0201	2535	0.5613	0.7391
0.3533	13.0201	2730	0.4344	0.8261
0.5705	14.0201	2925	1.2464	0.6087
0.3866	15.0201	3120	0.4108	0.8261
0.7164	16.0201	3315	0.3473	0.8696
0.9217	17.0201	3510	0.3302	0.8478
0.8893	18.0201	3705	0.5586	0.8043
0.4097	19.0201	3900	0.5201	0.7826
0.7003	20.0201	4095	0.6116	0.8478
0.4199	21.0201	4290	0.7902	0.7391
0.2288	22.0201	4485	0.7856	0.8261
0.205	23.0201	4680	0.5854	0.8478
0.312	24.0201	4875	0.6283	0.8696
0.4002	25.0201	5070	0.8241	0.8261
0.4637	26.0201	5265	1.0216	0.7826
0.1429	27.0201	5460	0.9300	0.8043
0.4351	28.0201	5655	1.0207	0.8043
0.2324	29.0201	5850	1.3275	0.7391
0.3545	30.0201	6045	1.1007	0.8261
0.1156	31.0201	6240	1.2325	0.8261
0.3738	32.0201	6435	1.0279	0.8261
0.2009	33.0201	6630	1.5226	0.7391
0.1003	34.0201	6825	0.6896	0.8478
0.2604	35.0201	7020	0.6089	0.8913
0.0003	36.0201	7215	0.7792	0.8696
0.1185	37.0201	7410	0.5411	0.9348
0.1388	38.0201	7605	0.5653	0.9130
0.713	39.0201	7800	0.5521	0.9348
0.3258	40.0201	7995	0.5768	0.9348
0.0009	41.0201	8190	0.7341	0.8696
0.0802	42.0201	8385	0.7582	0.8478
0.2027	43.0201	8580	1.2214	0.8261
0.1575	44.0201	8775	1.0244	0.8696
0.0298	45.0201	8970	0.9467	0.8696
0.0002	46.0201	9165	0.9423	0.8696
0.0785	47.0201	9360	0.9311	0.8478
0.0056	48.0201	9555	0.9431	0.8478
0.0178	49.0149	9700	0.9325	0.8478

Framework versions

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