metadata

license: cc-by-nc-4.0
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: videomae-base-finetuned-ucf101-subset
    results: []

videomae-base-finetuned-ucf101-subset

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: 0.1051
Accuracy: 0.9677

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: 3750

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.3017	0.02	75	2.3034	0.1286
1.9471	1.02	150	1.9157	0.3714
0.9752	2.02	225	0.9294	0.6714
0.4076	3.02	300	0.3997	0.8143
0.4284	4.02	375	0.6502	0.7714
0.1713	5.02	450	0.3226	0.8714
0.4188	6.02	525	0.6348	0.8429
0.0101	7.02	600	0.4137	0.9143
0.0126	8.02	675	0.1349	0.9429
0.0444	9.02	750	1.0845	0.8
0.0023	10.02	825	0.1856	0.9
0.0018	11.02	900	0.2485	0.9143
0.0017	12.02	975	0.0677	0.9714
0.0429	13.02	1050	0.1233	0.9286
0.0035	14.02	1125	0.3002	0.9571
0.0013	15.02	1200	0.1211	0.9857
0.001	16.02	1275	0.1264	0.9714
0.0013	17.02	1350	0.2979	0.9429
0.0714	18.02	1425	0.4232	0.9143
0.0011	19.02	1500	0.3415	0.9286
0.001	20.02	1575	0.1736	0.9286
0.0007	21.02	1650	0.2461	0.9429
0.0008	22.02	1725	0.3369	0.9286
0.0007	23.02	1800	0.1453	0.9571
0.0007	24.02	1875	0.1013	0.9857
0.0006	25.02	1950	0.1100	0.9857
0.0006	26.02	2025	0.1088	0.9857
0.0006	27.02	2100	0.1165	0.9714
0.0006	28.02	2175	0.0660	0.9857
0.0007	29.02	2250	0.2951	0.9429
0.0005	30.02	2325	0.0896	0.9857
0.0007	31.02	2400	0.1059	0.9857
0.0005	32.02	2475	0.0989	0.9857
0.0005	33.02	2550	0.0771	0.9857
0.0005	34.02	2625	0.0759	0.9857
0.0005	35.02	2700	0.0803	0.9857
0.0004	36.02	2775	0.0892	0.9857
0.0004	37.02	2850	0.1913	0.9571
0.0004	38.02	2925	0.4228	0.9429
0.0004	39.02	3000	0.4060	0.9429
0.0004	40.02	3075	0.3824	0.9429
0.0951	41.02	3150	0.4202	0.9429
0.0004	42.02	3225	0.1987	0.9571
0.0004	43.02	3300	0.1764	0.9571
0.0004	44.02	3375	0.1509	0.9571
0.0004	45.02	3450	0.1498	0.9571
0.0004	46.02	3525	0.1441	0.9714
0.0004	47.02	3600	0.1332	0.9714
0.0004	48.02	3675	0.1573	0.9714
0.0003	49.02	3750	0.1616	0.9714

Framework versions

Transformers 4.24.0
Pytorch 1.8.0+cu111
Datasets 2.7.1
Tokenizers 0.13.2