videomae-base-finetuned-ucf-crimev2

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: 3.2451
• Accuracy: 0.4182

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: 5e-05
• train_batch_size: 16
• eval_batch_size: 16
• 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
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.7909	1.0	21	1.7588	0.3636
1.646	2.0	42	1.7454	0.3455
1.4793	3.0	63	1.8800	0.2727
1.4107	4.0	84	1.8559	0.3455
1.2787	5.0	105	1.9568	0.3636
1.0594	6.0	126	1.8249	0.3
0.8529	7.0	147	1.7899	0.3273
0.7591	8.0	168	1.8875	0.3
0.6159	9.0	189	2.2923	0.3091
0.3994	10.0	210	1.9616	0.3727
0.3325	11.0	231	2.1035	0.3455
0.2161	12.0	252	2.2052	0.3818
0.1492	13.0	273	2.4497	0.3273
0.1329	14.0	294	2.4025	0.3
0.097	15.0	315	2.8474	0.3091
0.0307	16.0	336	2.5895	0.3818
0.0262	17.0	357	2.8746	0.3182
0.0125	18.0	378	2.6371	0.3636
0.0186	19.0	399	2.9341	0.3727
0.0207	20.0	420	3.1826	0.3364
0.0244	21.0	441	2.9959	0.3364
0.0033	22.0	462	2.8171	0.4
0.0051	23.0	483	2.9913	0.3909
0.0025	24.0	504	3.0294	0.4
0.0022	25.0	525	3.0329	0.4
0.0031	26.0	546	3.0223	0.3818
0.002	27.0	567	3.0312	0.3909
0.002	28.0	588	3.0663	0.4091
0.0017	29.0	609	3.0907	0.4091
0.0016	30.0	630	3.1117	0.4091
0.0016	31.0	651	3.1219	0.4091
0.0015	32.0	672	3.1439	0.4091
0.0015	33.0	693	3.1582	0.4182
0.0014	34.0	714	3.1660	0.4182
0.0015	35.0	735	3.1735	0.4182
0.0014	36.0	756	3.1864	0.4182
0.0013	37.0	777	3.1929	0.4182
0.0013	38.0	798	3.2001	0.4182
0.0013	39.0	819	3.2073	0.4182
0.0013	40.0	840	3.2151	0.4182
0.0012	41.0	861	3.2217	0.4182
0.0012	42.0	882	3.2269	0.4182
0.0012	43.0	903	3.2316	0.4182
0.0012	44.0	924	3.2363	0.4182
0.0012	45.0	945	3.2396	0.4182
0.0012	46.0	966	3.2406	0.4182
0.0011	47.0	987	3.2423	0.4182
0.0012	48.0	1008	3.2436	0.4182
0.0011	49.0	1029	3.2446	0.4182
0.0012	50.0	1050	3.2451	0.4182

Framework versions

• Transformers 4.47.0.dev0
• Pytorch 2.5.1+cu124
• Datasets 3.1.0
• Tokenizers 0.20.3