videomae-base-finetuned-ucf101-subset-finetuned-subset-0401

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

• Loss: 0.7418
• Accuracy: 0.7269

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: 2775

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.6151	0.02	56	1.5567	0.2949
1.3399	1.02	112	1.2929	0.3825
1.2751	2.02	168	1.4618	0.3134
1.3725	3.02	224	0.9080	0.6498
1.0782	4.02	280	1.1473	0.5300
1.1514	5.02	336	0.8953	0.6359
1.0593	6.02	392	1.3372	0.4608
1.1193	7.02	448	0.9655	0.6313
0.719	8.02	504	0.8527	0.6728
1.0157	9.02	560	1.2763	0.5023
0.6991	10.02	616	0.8840	0.6406
0.9019	11.02	672	0.8941	0.6636
0.7456	12.02	728	1.0455	0.6037
0.6631	13.02	784	0.6456	0.7558
0.7143	14.02	840	0.8887	0.6682
0.6639	15.02	896	0.6863	0.7604
0.5195	16.02	952	0.9475	0.6221
0.9211	17.02	1008	0.7339	0.7373
0.5328	18.02	1064	0.9085	0.6544
0.6818	19.02	1120	0.7977	0.7097
0.6132	20.02	1176	0.7116	0.7373
0.4113	21.02	1232	1.0191	0.5853
0.7443	22.02	1288	1.2705	0.5714
0.6904	23.02	1344	0.8419	0.6636
0.5888	24.02	1400	0.7481	0.6959
0.6739	25.02	1456	0.9970	0.7097
0.6595	26.02	1512	1.3474	0.5806
0.5574	27.02	1568	0.6245	0.7926
0.5627	28.02	1624	0.7718	0.7143
0.6417	29.02	1680	0.6506	0.7604
0.3854	30.02	1736	0.9524	0.6544
0.354	31.02	1792	1.0662	0.5945
0.7568	32.02	1848	0.7329	0.7604
0.5359	33.02	1904	0.8958	0.6774
0.5946	34.02	1960	0.8312	0.6912
0.5673	35.02	2016	0.7245	0.7051
0.4291	36.02	2072	0.8294	0.6912
0.5245	37.02	2128	0.8931	0.7005
0.4113	38.02	2184	0.7470	0.6912
0.456	39.02	2240	0.8325	0.6728
0.6955	40.02	2296	0.6941	0.7788
0.6283	41.02	2352	0.9662	0.6636
0.6465	42.02	2408	1.1286	0.6129
0.4387	43.02	2464	0.9525	0.6175
0.2879	44.02	2520	1.0277	0.6313
0.5188	45.02	2576	1.0631	0.6359
0.4464	46.02	2632	0.9313	0.6359
0.6155	47.02	2688	0.9699	0.6267
0.3921	48.02	2744	1.0027	0.6313
0.3345	49.01	2775	1.0278	0.6221

Framework versions

• Transformers 4.35.2
• Pytorch 1.13.1
• Datasets 2.15.0
• Tokenizers 0.15.0