hushem_1x_beit_base_rms_0001_fold1

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

• Loss: 1.7892
• Accuracy: 0.4222

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: 0.0001
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• 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
No log	1.0	6	1.3881	0.2444
1.983	2.0	12	1.4040	0.2444
1.983	3.0	18	1.4052	0.2667
1.41	4.0	24	1.3851	0.2444
1.3993	5.0	30	1.3596	0.2667
1.3993	6.0	36	1.5010	0.2444
1.3135	7.0	42	1.4385	0.3778
1.3135	8.0	48	1.3273	0.2222
1.2878	9.0	54	1.7515	0.2444
1.2036	10.0	60	1.4739	0.3111
1.2036	11.0	66	1.4793	0.4444
1.1544	12.0	72	1.6976	0.4444
1.1544	13.0	78	1.5051	0.3778
1.1611	14.0	84	2.0887	0.2444
1.0944	15.0	90	1.7507	0.3778
1.0944	16.0	96	1.5983	0.4
1.1053	17.0	102	1.5239	0.3333
1.1053	18.0	108	1.7239	0.3333
0.9531	19.0	114	1.7796	0.3778
0.9208	20.0	120	1.7000	0.4
0.9208	21.0	126	1.5682	0.3556
0.9119	22.0	132	1.6947	0.2889
0.9119	23.0	138	1.9309	0.3111
0.8438	24.0	144	1.7778	0.4
0.7982	25.0	150	1.3358	0.4889
0.7982	26.0	156	1.8930	0.3778
0.7528	27.0	162	1.5978	0.4444
0.7528	28.0	168	1.7048	0.4
0.7372	29.0	174	1.4976	0.4
0.6872	30.0	180	1.5193	0.4222
0.6872	31.0	186	1.5712	0.3778
0.6257	32.0	192	1.6492	0.4
0.6257	33.0	198	1.6572	0.4444
0.6115	34.0	204	1.7617	0.4222
0.502	35.0	210	1.7836	0.4
0.502	36.0	216	1.7245	0.4222
0.5351	37.0	222	1.8523	0.3778
0.5351	38.0	228	1.8752	0.3778
0.4239	39.0	234	1.7739	0.4222
0.4397	40.0	240	1.8121	0.4
0.4397	41.0	246	1.7942	0.4222
0.3888	42.0	252	1.7892	0.4222
0.3888	43.0	258	1.7892	0.4222
0.3836	44.0	264	1.7892	0.4222
0.3564	45.0	270	1.7892	0.4222
0.3564	46.0	276	1.7892	0.4222
0.3801	47.0	282	1.7892	0.4222
0.3801	48.0	288	1.7892	0.4222
0.316	49.0	294	1.7892	0.4222
0.3933	50.0	300	1.7892	0.4222

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0