hushem_5x_deit_small_sgd_0001_fold3

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

• Loss: 1.4264
• Accuracy: 0.2093

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
1.5059	1.0	28	1.5906	0.2093
1.4969	2.0	56	1.5795	0.2093
1.4242	3.0	84	1.5692	0.2093
1.4628	4.0	112	1.5599	0.2093
1.5851	5.0	140	1.5506	0.2093
1.4881	6.0	168	1.5418	0.2093
1.4744	7.0	196	1.5338	0.2326
1.427	8.0	224	1.5266	0.2558
1.4274	9.0	252	1.5192	0.2558
1.4168	10.0	280	1.5125	0.2558
1.4418	11.0	308	1.5064	0.2558
1.4361	12.0	336	1.5007	0.2558
1.4139	13.0	364	1.4950	0.2558
1.3932	14.0	392	1.4898	0.2558
1.4041	15.0	420	1.4850	0.2326
1.3745	16.0	448	1.4806	0.2326
1.3653	17.0	476	1.4764	0.2093
1.3841	18.0	504	1.4723	0.2093
1.3735	19.0	532	1.4687	0.2093
1.3391	20.0	560	1.4653	0.2093
1.3879	21.0	588	1.4620	0.2093
1.3861	22.0	616	1.4589	0.2093
1.3726	23.0	644	1.4561	0.2093
1.3725	24.0	672	1.4534	0.2093
1.3587	25.0	700	1.4508	0.2093
1.3359	26.0	728	1.4485	0.2093
1.3627	27.0	756	1.4462	0.2326
1.3855	28.0	784	1.4442	0.2326
1.353	29.0	812	1.4424	0.2093
1.301	30.0	840	1.4407	0.2093
1.3248	31.0	868	1.4390	0.2093
1.3654	32.0	896	1.4375	0.2093
1.364	33.0	924	1.4361	0.2093
1.322	34.0	952	1.4347	0.2093
1.3619	35.0	980	1.4335	0.2093
1.3562	36.0	1008	1.4324	0.2093
1.4034	37.0	1036	1.4314	0.2093
1.3401	38.0	1064	1.4304	0.2093
1.3307	39.0	1092	1.4297	0.2093
1.3736	40.0	1120	1.4290	0.2093
1.3675	41.0	1148	1.4284	0.2093
1.3234	42.0	1176	1.4279	0.2093
1.3321	43.0	1204	1.4274	0.2093
1.3436	44.0	1232	1.4270	0.2093
1.3719	45.0	1260	1.4268	0.2093
1.3462	46.0	1288	1.4266	0.2093
1.3448	47.0	1316	1.4265	0.2093
1.3465	48.0	1344	1.4264	0.2093
1.2951	49.0	1372	1.4264	0.2093
1.3665	50.0	1400	1.4264	0.2093

Framework versions

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