2D_hgg_lgg_classification

This model is a fine-tuned version of microsoft/swin-tiny-patch4-window7-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.7560
• Accuracy: 0.8203

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: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• 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
0.7243	0.9655	7	0.6272	0.7656
0.5807	1.9310	14	0.5266	0.7812
0.556	2.8966	21	0.5086	0.7812
0.4675	4.0	29	0.4844	0.7812
0.4992	4.9655	36	0.4664	0.7812
0.4562	5.9310	43	0.4430	0.7344
0.4344	6.8966	50	0.4726	0.7109
0.3778	8.0	58	0.4302	0.7656
0.3922	8.9655	65	0.4350	0.8125
0.3864	9.9310	72	0.4259	0.7656
0.3388	10.8966	79	0.4462	0.7656
0.3071	12.0	87	0.5272	0.7969
0.3233	12.9655	94	0.4723	0.7188
0.3103	13.9310	101	0.4494	0.7656
0.2818	14.8966	108	0.4279	0.8047
0.2341	16.0	116	0.4069	0.7891
0.2103	16.9655	123	0.4237	0.7969
0.219	17.9310	130	0.4467	0.8047
0.21	18.8966	137	0.4380	0.7812
0.1994	20.0	145	0.4629	0.7969
0.1865	20.9655	152	0.5012	0.7891
0.1872	21.9310	159	0.5055	0.8203
0.2144	22.8966	166	0.6089	0.8125
0.1737	24.0	174	0.4914	0.7969
0.1633	24.9655	181	0.5137	0.7812
0.1624	25.9310	188	0.5985	0.7812
0.1525	26.8966	195	0.5090	0.8047
0.136	28.0	203	0.5170	0.8125
0.1451	28.9655	210	0.6165	0.8203
0.1405	29.9310	217	0.6124	0.7969
0.1384	30.8966	224	0.5578	0.8047
0.1246	32.0	232	0.5967	0.8125
0.1371	32.9655	239	0.6135	0.7812
0.1111	33.9310	246	0.6878	0.8047
0.1305	34.8966	253	0.7300	0.8125
0.1124	36.0	261	0.6687	0.8203
0.1214	36.9655	268	0.6692	0.8047
0.1065	37.9310	275	0.7058	0.8125
0.1183	38.8966	282	0.6884	0.7969
0.0928	40.0	290	0.7104	0.7969
0.1248	40.9655	297	0.6961	0.7969
0.0949	41.9310	304	0.7265	0.8203
0.1048	42.8966	311	0.7430	0.8281
0.0887	44.0	319	0.7627	0.8047
0.0866	44.9655	326	0.7483	0.8203
0.0978	45.9310	333	0.7515	0.8125
0.0901	46.8966	340	0.7518	0.8125
0.0785	48.0	348	0.7557	0.8203
0.0747	48.2759	350	0.7560	0.8203

Framework versions

• Transformers 4.45.1
• Pytorch 2.4.0
• Datasets 3.0.1
• Tokenizers 0.20.0