MAE-CT-CPC-Dicotomized-v6-Day1

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: 1.2706
• Accuracy: 0.7333

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 2750

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.638	0.02	56	0.6488	0.64
0.7032	1.02	112	0.6394	0.64
0.6243	2.02	168	0.5943	0.64
0.6339	3.02	224	0.6504	0.6
0.6566	4.02	280	0.6405	0.64
0.4056	5.02	336	0.6274	0.64
0.3523	6.02	392	0.9109	0.64
0.4581	7.02	448	0.4675	0.8
0.3648	8.02	504	0.5780	0.76
0.4622	9.02	560	1.1120	0.64
0.5836	10.02	616	0.6369	0.76
0.316	11.02	672	0.8769	0.64
0.135	12.02	728	0.6827	0.72
0.0817	13.02	784	0.9667	0.64
0.5254	14.02	840	0.7442	0.76
0.2692	15.02	896	0.5944	0.84
0.0177	16.02	952	1.0163	0.76
0.0386	17.02	1008	0.8789	0.76
0.2142	18.02	1064	1.0580	0.68
0.0653	19.02	1120	0.9189	0.72
0.0004	20.02	1176	1.1913	0.76
0.0006	21.02	1232	1.1668	0.72
0.0006	22.02	1288	1.2782	0.76
0.0003	23.02	1344	1.2591	0.76
0.0004	24.02	1400	1.5768	0.72
0.0431	25.02	1456	1.1632	0.8
0.008	26.02	1512	1.4113	0.76
0.0003	27.02	1568	1.2239	0.76
0.003	28.02	1624	1.7195	0.72
0.3129	29.02	1680	1.9161	0.68
0.0001	30.02	1736	1.4177	0.76
0.0001	31.02	1792	1.4688	0.68
0.0314	32.02	1848	1.4026	0.68
0.0001	33.02	1904	1.5846	0.72
0.0001	34.02	1960	1.4021	0.64
0.0002	35.02	2016	1.5994	0.72
0.0001	36.02	2072	1.4027	0.72
0.0002	37.02	2128	1.3608	0.76
0.0001	38.02	2184	1.3569	0.8
0.0001	39.02	2240	1.4026	0.8
0.0001	40.02	2296	1.4563	0.72
0.0028	41.02	2352	1.3389	0.76
0.0001	42.02	2408	1.3355	0.8
0.0001	43.02	2464	1.3445	0.8
0.0001	44.02	2520	1.3526	0.8
0.0001	45.02	2576	1.3842	0.72
0.0001	46.02	2632	1.3895	0.72
0.0025	47.02	2688	1.4828	0.76
0.0001	48.02	2744	1.4092	0.76
0.2131	49.0	2750	1.4071	0.76

Framework versions

• Transformers 4.37.2
• Pytorch 2.0.1+cu117
• Datasets 3.0.1
• Tokenizers 0.15.1