tsf-gs-rot-flip-wtoken-DRPT-r128-f198-4.4-h768-i3072-p32-b4-e60

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: 1.4736
• Accuracy: 0.7487

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: 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: 13020
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.0904	0.0167	217	1.1132	0.3262
1.1105	1.0167	434	1.1408	0.3262
1.2079	2.0167	651	1.1040	0.3422
1.1561	3.0167	868	1.1015	0.3316
1.1399	4.0167	1085	1.1023	0.3262
1.1942	5.0167	1302	1.1053	0.3743
1.09	6.0167	1519	1.0816	0.3690
1.1052	7.0167	1736	1.1205	0.3369
0.9883	8.0167	1953	1.1582	0.3904
1.0268	9.0167	2170	0.9488	0.5775
0.8247	10.0167	2387	0.9106	0.6417
0.841	11.0167	2604	0.8541	0.6471
0.8026	12.0167	2821	0.7250	0.6898
0.942	13.0167	3038	0.7447	0.6578
0.7162	14.0167	3255	0.8518	0.6043
1.0489	15.0167	3472	0.9505	0.6096
0.8231	16.0167	3689	1.0083	0.5882
0.8702	17.0167	3906	0.7928	0.6684
0.5014	18.0167	4123	0.7290	0.7433
0.5914	19.0167	4340	0.6092	0.7701
0.6259	20.0167	4557	1.0359	0.7487
0.7433	21.0167	4774	1.0086	0.7326
0.2929	22.0167	4991	1.7072	0.6364
0.3763	23.0167	5208	0.9161	0.7487
0.4253	24.0167	5425	1.1377	0.7219
0.7882	25.0167	5642	1.2351	0.7059
0.5644	26.0167	5859	1.5853	0.6150
0.6372	27.0167	6076	0.8027	0.7540
0.3907	28.0167	6293	1.6891	0.6096
0.7374	29.0167	6510	1.2730	0.6791
0.6292	30.0167	6727	1.1093	0.7326
0.2892	31.0167	6944	1.2396	0.7219
0.6779	32.0167	7161	1.6986	0.6524
0.2337	33.0167	7378	1.7567	0.6898
0.3919	34.0167	7595	1.1040	0.7380
0.6117	35.0167	7812	1.0982	0.7701
0.5463	36.0167	8029	1.1052	0.7380
0.3559	37.0167	8246	1.6846	0.6578
0.3356	38.0167	8463	0.9712	0.7487
0.0266	39.0167	8680	1.7524	0.6684
0.3919	40.0167	8897	1.6011	0.7112
0.1367	41.0167	9114	1.1935	0.7433
0.1936	42.0167	9331	1.5143	0.7326
0.7236	43.0167	9548	1.1621	0.7861
0.5877	44.0167	9765	1.5272	0.7326
0.3265	45.0167	9982	1.2536	0.7754
0.4684	46.0167	10199	1.1489	0.7647
0.3436	47.0167	10416	1.3433	0.7701
0.0272	48.0167	10633	1.5171	0.7380
0.3487	49.0167	10850	1.1279	0.7647
0.024	50.0167	11067	1.6780	0.7273
0.4644	51.0167	11284	1.3860	0.7914
0.5177	52.0167	11501	1.7174	0.7219
0.263	53.0167	11718	1.4811	0.7326
0.3323	54.0167	11935	1.4279	0.7807
0.2919	55.0167	12152	1.4855	0.7166
0.1878	56.0167	12369	1.4966	0.7594
0.1307	57.0167	12586	1.3280	0.7594
0.4798	58.0167	12803	1.6149	0.7487
0.0599	59.0167	13020	1.3786	0.7540

Framework versions

• Transformers 4.41.2
• Pytorch 1.13.0+cu117
• Datasets 2.20.0
• Tokenizers 0.19.1