timesformer-base-finetuned-k400-timesformer-2-10-epochs-v_2sec_others-1735155772.870025

This model is a fine-tuned version of facebook/timesformer-base-finetuned-k400 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7685
• Accuracy: 0.7754

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: 184630

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.0556	0.0250	4616	0.5976	0.7717
0.1338	1.0250	9232	0.4428	0.8530
0.0482	2.0250	13848	0.3892	0.8250
0.0496	3.0250	18464	0.6797	0.8018
0.061	4.0250	23080	0.5792	0.8001
0.0668	5.0250	27696	0.7014	0.8054
0.0653	6.0250	32312	1.0386	0.7756
0.0079	7.0250	36928	0.5399	0.8282
0.0377	8.0250	41544	0.5704	0.8325
0.1408	9.0250	46160	0.4445	0.8596
0.0676	10.0250	50776	0.4410	0.8730
0.019	11.0250	55392	0.7983	0.8138
0.026	12.0250	60008	0.6877	0.8357
0.0095	13.0250	64624	0.9396	0.7877
0.0801	14.0250	69240	0.7813	0.8400
0.0243	15.0250	73856	0.9217	0.8256
0.1769	16.0250	78472	0.7920	0.8380
0.0027	17.0250	83088	0.9414	0.8292
0.0902	18.0250	87704	0.8596	0.8289
0.0715	19.0250	92320	0.7860	0.8357
0.1008	20.0250	96936	0.7183	0.8393
0.0808	21.0250	101552	0.8277	0.8302
0.0324	22.0250	106168	0.7808	0.8485
0.0304	23.0250	110784	0.8359	0.8521
0.0089	24.0250	115400	0.9719	0.8233
0.0371	25.0250	120016	0.9092	0.8380
0.0278	26.0250	124632	0.9338	0.8374
0.0681	27.0250	129248	0.8934	0.8566
0.0224	28.0250	133864	0.9310	0.8472
0.0432	29.0250	138480	0.8325	0.8641
0.019	30.0250	143096	0.8224	0.8357
0.0888	31.0250	147712	0.9873	0.8217
0.0045	32.0250	152328	1.0976	0.8184
0.08	33.0250	156944	0.9634	0.8442
0.0005	34.0250	161560	1.0483	0.8347
0.0421	35.0250	166176	1.0527	0.8250
0.0354	36.0250	170792	1.0301	0.8442
0.0086	37.0250	175408	1.1078	0.8432
0.1433	38.0250	180024	1.1225	0.8396
0.0164	39.0249	184630	1.1233	0.8396

Framework versions

• Transformers 4.45.1
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.20.0