font-identifier

This model is a fine-tuned version of microsoft/resnet-18 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.1706
• Accuracy: 0.9455

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: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• 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
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
3.2323	1.0	14	3.1711	0.05
3.1019	2.0	28	2.9173	0.1227
2.779	3.0	42	2.5695	0.2409
2.5	4.0	56	2.1142	0.4091
1.8064	5.0	70	1.6804	0.4864
1.575	6.0	84	1.2757	0.5773
1.348	7.0	98	1.0973	0.6864
1.0483	8.0	112	0.8965	0.7273
0.9753	9.0	126	0.7025	0.7682
0.7763	10.0	140	0.6220	0.8091
0.7392	11.0	154	0.5169	0.8636
0.7077	12.0	168	0.4815	0.8682
0.5433	13.0	182	0.4650	0.8455
0.565	14.0	196	0.3828	0.8773
0.4204	15.0	210	0.3493	0.8864
0.4798	16.0	224	0.2847	0.9045
0.4353	17.0	238	0.3370	0.8773
0.3871	18.0	252	0.2797	0.9045
0.3779	19.0	266	0.2671	0.9045
0.3819	20.0	280	0.2575	0.9
0.3216	21.0	294	0.2516	0.9227
0.3461	22.0	308	0.2368	0.9045
0.3116	23.0	322	0.2651	0.9136
0.3244	24.0	336	0.2820	0.9
0.2725	25.0	350	0.2320	0.9045
0.3377	26.0	364	0.2309	0.9318
0.2556	27.0	378	0.2361	0.9136
0.2654	28.0	392	0.1988	0.9364
0.2578	29.0	406	0.2322	0.9227
0.2262	30.0	420	0.1686	0.9409
0.2298	31.0	434	0.2148	0.9091
0.2259	32.0	448	0.1982	0.9318
0.2155	33.0	462	0.2340	0.9227
0.213	34.0	476	0.1359	0.9545
0.1812	35.0	490	0.1522	0.9409
0.1793	36.0	504	0.1553	0.9409
0.2391	37.0	518	0.1149	0.9636
0.1755	38.0	532	0.1627	0.9273
0.1907	39.0	546	0.1555	0.95
0.1814	40.0	560	0.1832	0.9409
0.1879	41.0	574	0.2046	0.9318
0.1953	42.0	588	0.1722	0.9364
0.1814	43.0	602	0.2270	0.9455
0.1932	44.0	616	0.1651	0.9318
0.1813	45.0	630	0.1752	0.9318
0.1691	46.0	644	0.1681	0.9636
0.1396	47.0	658	0.1604	0.9545
0.1647	48.0	672	0.1575	0.95
0.1501	49.0	686	0.1360	0.9545
0.1534	50.0	700	0.1706	0.9455

Framework versions

• Transformers 4.54.1
• Pytorch 2.7.1
• Datasets 4.0.0
• Tokenizers 0.21.4