CTMAE-P2-V3-3G-S3

This model is a fine-tuned version of MCG-NJU/videomae-large-finetuned-kinetics on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7936
• Accuracy: 0.7609

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: 1e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 3250

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6412	0.02	65	0.7701	0.3913
0.6226	1.02	130	0.8575	0.3913
0.5791	2.02	195	0.8565	0.3913
0.5009	3.02	260	0.8441	0.4348
0.4851	4.02	325	0.6869	0.6522
0.3977	5.02	390	1.2020	0.4783
0.4922	6.02	455	0.6499	0.6739
0.6226	7.02	520	0.7404	0.6522
0.4225	8.02	585	0.9433	0.5217
0.5705	9.02	650	1.1601	0.4565
0.4147	10.02	715	0.9161	0.5870
0.4679	11.02	780	1.3047	0.4565
0.4552	12.02	845	0.6094	0.7391
0.2635	13.02	910	1.1579	0.6087
0.3768	14.02	975	1.6995	0.4130
0.3825	15.02	1040	1.1773	0.5652
0.3682	16.02	1105	1.3407	0.6304
0.3113	17.02	1170	0.7936	0.7609
0.7703	18.02	1235	1.4047	0.5652
0.2917	19.02	1300	1.3266	0.5435
0.1025	20.02	1365	1.6342	0.6087
0.2367	21.02	1430	1.5501	0.5870
0.561	22.02	1495	1.4234	0.6304
0.2348	23.02	1560	1.4827	0.6957
0.1334	24.02	1625	1.4610	0.6739
0.1724	25.02	1690	1.5800	0.6957
0.7534	26.02	1755	1.6274	0.6957
0.1041	27.02	1820	1.5118	0.6739
0.3881	28.02	1885	1.9602	0.5870
0.2522	29.02	1950	1.3981	0.6957
0.3325	30.02	2015	1.3942	0.7174
0.0227	31.02	2080	1.5405	0.6957
0.0993	32.02	2145	1.8304	0.6957
0.0054	33.02	2210	2.0809	0.6522
0.1713	34.02	2275	2.0062	0.6739
0.1629	35.02	2340	1.8752	0.6739
0.2571	36.02	2405	2.1408	0.6304
0.1893	37.02	2470	2.0235	0.6304
0.164	38.02	2535	1.9872	0.7174
0.0382	39.02	2600	1.9095	0.6739
0.0006	40.02	2665	1.9461	0.6739
0.2606	41.02	2730	1.4566	0.7174
0.0276	42.02	2795	1.5791	0.7391
0.091	43.02	2860	1.8797	0.6739
0.0031	44.02	2925	1.8557	0.6739
0.0093	45.02	2990	1.6521	0.7391
0.0004	46.02	3055	1.8300	0.6739
0.0278	47.02	3120	1.7774	0.6739
0.0034	48.02	3185	1.8724	0.6739
0.0363	49.02	3250	1.9296	0.6739

Framework versions

• Transformers 4.46.2
• Pytorch 2.0.1+cu117
• Datasets 3.0.1
• Tokenizers 0.20.0