videomae-base-finetuned-ucf101-subset-V3KILLER

This model is a fine-tuned version of abduldattijo/videomae-base-finetuned-ucf101-subset on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2181
• Accuracy: 0.9615

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: 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: 5960

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.3447	0.03	150	0.1339	0.9579
0.3161	1.03	300	0.1538	0.9465
0.3386	2.03	450	0.3260	0.9019
0.3572	3.03	600	0.1967	0.9311
0.3699	4.03	750	0.1661	0.9505
0.3125	5.03	900	0.3292	0.9205
0.4785	6.03	1050	0.2029	0.9324
0.3477	7.03	1200	0.1534	0.9385
0.2909	8.03	1350	0.1265	0.9571
0.2646	9.03	1500	0.1239	0.9586
0.3339	10.03	1650	0.1341	0.9628
0.0954	11.03	1800	0.1835	0.9423
0.3861	12.03	1950	0.2241	0.9467
0.248	13.03	2100	0.1258	0.9620
0.2513	14.03	2250	0.2217	0.9357
0.1133	15.03	2400	0.2129	0.9406
0.1421	16.03	2550	0.3006	0.9264
0.0248	17.03	2700	0.3868	0.9142
0.0166	18.03	2850	0.2594	0.9518
0.0874	19.03	3000	0.3652	0.9252
0.0889	20.03	3150	0.2249	0.9533
0.0804	21.03	3300	0.2027	0.9628
0.0019	22.03	3450	0.4682	0.9212
0.0405	23.03	3600	0.2425	0.9493
0.0847	24.03	3750	0.2456	0.9558
0.1656	25.03	3900	0.2623	0.9505
0.1007	26.03	4050	0.2389	0.9484
0.0616	27.03	4200	0.2529	0.9543
0.0005	28.03	4350	0.1521	0.9732
0.0006	29.03	4500	0.4115	0.9165
0.0007	30.03	4650	0.4279	0.9220
0.0004	31.03	4800	0.3572	0.9372
0.0003	32.03	4950	0.3314	0.9419
0.0002	33.03	5100	0.4008	0.9347
0.0611	34.03	5250	0.4632	0.9239
0.0003	35.03	5400	0.3756	0.9368
0.0003	36.03	5550	0.3745	0.9429
0.163	37.03	5700	0.3967	0.9383
0.0059	38.03	5850	0.3808	0.9389
0.0003	39.02	5960	0.3824	0.9395

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.1+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0