deit-base-patch16-224-finetuned-ind-17-imbalanced-aadhaarmask-14687

This model is a fine-tuned version of facebook/deit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

Loss: 0.3950
Accuracy: 0.8310
Recall: 0.8310
F1: 0.8298
Precision: 0.8360

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
gradient_accumulation_steps: 4
total_train_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Recall	F1	Precision
0.8293	0.9974	293	0.7793	0.7680	0.7680	0.7403	0.7277
0.5921	1.9983	587	0.5663	0.7940	0.7940	0.7843	0.7839
0.4308	2.9991	881	0.4589	0.8208	0.8208	0.8161	0.8213
0.3999	4.0	1175	0.4772	0.8263	0.8263	0.8216	0.8337
0.4801	4.9974	1468	0.4258	0.8378	0.8378	0.8306	0.8463
0.4201	5.9983	1762	0.4120	0.8246	0.8246	0.8213	0.8394
0.3233	6.9991	2056	0.3989	0.8306	0.8306	0.8268	0.8445
0.3954	8.0	2350	0.3794	0.8365	0.8365	0.8341	0.8383
0.2835	8.9974	2643	0.4438	0.8318	0.8318	0.8278	0.8434
0.2913	9.9983	2937	0.3799	0.8416	0.8416	0.8404	0.8451
0.3261	10.9991	3231	0.3694	0.8297	0.8297	0.8272	0.8306
0.3299	12.0	3525	0.3637	0.8442	0.8442	0.8425	0.8529
0.3273	12.9974	3818	0.3649	0.8421	0.8421	0.8411	0.8482
0.2596	13.9983	4112	0.4152	0.8259	0.8259	0.8213	0.8281
0.2813	14.9991	4406	0.3578	0.8429	0.8429	0.8409	0.8491
0.2406	16.0	4700	0.3813	0.8323	0.8323	0.8285	0.8362
0.2263	16.9974	4993	0.3808	0.8318	0.8318	0.8275	0.8377
0.3192	17.9983	5287	0.3625	0.8412	0.8412	0.8372	0.8484
0.2003	18.9991	5581	0.3549	0.8438	0.8438	0.8430	0.8462
0.2431	20.0	5875	0.3620	0.8425	0.8425	0.8408	0.8467
0.2654	20.9974	6168	0.3865	0.8340	0.8340	0.8320	0.8338
0.2989	21.9983	6462	0.3632	0.8463	0.8463	0.8449	0.8498
0.2403	22.9991	6756	0.3824	0.8301	0.8301	0.8267	0.8304
0.2393	24.0	7050	0.3607	0.8489	0.8489	0.8473	0.8519
0.2305	24.9974	7343	0.3758	0.8365	0.8365	0.8350	0.8401
0.2654	25.9983	7637	0.3652	0.8421	0.8421	0.8392	0.8415
0.176	26.9991	7931	0.3929	0.8306	0.8306	0.8289	0.8385
0.1893	28.0	8225	0.3794	0.8374	0.8374	0.8365	0.8404
0.2652	28.9974	8518	0.3995	0.8387	0.8387	0.8372	0.8423
0.2029	29.9983	8812	0.3981	0.8433	0.8433	0.8411	0.8430
0.1799	30.9991	9106	0.3554	0.8352	0.8352	0.8340	0.8368
0.2002	32.0	9400	0.3618	0.8310	0.8310	0.8300	0.8322
0.1525	32.9974	9693	0.3629	0.8348	0.8348	0.8343	0.8381
0.1663	33.9983	9987	0.3664	0.8425	0.8425	0.8410	0.8427
0.1728	34.9991	10281	0.3928	0.8429	0.8429	0.8415	0.8468
0.2252	36.0	10575	0.3842	0.8421	0.8421	0.8420	0.8443
0.1554	36.9974	10868	0.3889	0.8301	0.8301	0.8294	0.8349
0.2179	37.9983	11162	0.3775	0.8399	0.8399	0.8389	0.8429
0.1771	38.9991	11456	0.3906	0.8306	0.8306	0.8291	0.8324
0.2167	40.0	11750	0.3870	0.8404	0.8404	0.8382	0.8456
0.1563	40.9974	12043	0.3779	0.8284	0.8284	0.8277	0.8288
0.1419	41.9983	12337	0.4049	0.8340	0.8340	0.8327	0.8360
0.2083	42.9991	12631	0.3800	0.8421	0.8421	0.8410	0.8427
0.2185	44.0	12925	0.3964	0.8433	0.8433	0.8422	0.8441
0.1989	44.9974	13218	0.3870	0.8340	0.8340	0.8339	0.8357
0.1731	45.9983	13512	0.4206	0.8340	0.8340	0.8335	0.8357
0.1831	46.9991	13806	0.4027	0.8429	0.8429	0.8422	0.8439
0.1471	48.0	14100	0.4016	0.8318	0.8318	0.8307	0.8320
0.1879	48.9974	14393	0.3877	0.8438	0.8438	0.8441	0.8468
0.1775	49.8723	14650	0.3984	0.8421	0.8421	0.8408	0.8428

Framework versions

Transformers 4.40.1
Pytorch 2.2.0a0+81ea7a4
Datasets 2.18.0
Tokenizers 0.19.1

Kushagra07/deit-base-patch16-224-finetuned-ind-17-imbalanced-aadhaarmask-14687

deit-base-patch16-224-finetuned-ind-17-imbalanced-aadhaarmask-14687

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

为此模型做种