convnext-tiny-224-finetuned-eurosat-albumentations

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

• Loss: 0.2746
• Accuracy: 0.9589

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
0.0418	0.9362	11	0.2497	0.9041
0.0349	1.9574	23	0.2377	0.9041
0.0231	2.9787	35	0.2695	0.8904
0.0154	4.0	47	0.2185	0.9041
0.0061	4.9362	58	0.2810	0.9178
0.0023	5.9574	70	0.2905	0.9315
0.0029	6.9787	82	0.2052	0.9315
0.0013	8.0	94	0.2333	0.9178
0.0009	8.9362	105	0.2262	0.9315
0.008	9.9574	117	0.2247	0.9178
0.0014	10.9787	129	0.3200	0.9041
0.0005	12.0	141	0.2643	0.9178
0.0006	12.9362	152	0.2911	0.9178
0.0007	13.9574	164	0.2567	0.9178
0.0009	14.9787	176	0.3170	0.9178
0.0052	16.0	188	0.2435	0.9315
0.0005	16.9362	199	0.2746	0.9589
0.0004	17.9574	211	0.2347	0.9315
0.002	18.9787	223	0.2999	0.9178
0.0021	20.0	235	0.2648	0.9178
0.0003	20.9362	246	0.2609	0.9178
0.0002	21.9574	258	0.2709	0.9315
0.0004	22.9787	270	0.2359	0.9315
0.005	24.0	282	0.2484	0.9178
0.0011	24.9362	293	0.3019	0.9178
0.0012	25.9574	305	0.2715	0.9178
0.0003	26.9787	317	0.2486	0.9315
0.0009	28.0	329	0.2472	0.9315
0.0002	28.9362	340	0.2449	0.9315
0.0002	29.9574	352	0.2480	0.9315
0.0002	30.9787	364	0.2520	0.9315
0.0002	32.0	376	0.2528	0.9315
0.0001	32.9362	387	0.2520	0.9315
0.0002	33.9574	399	0.2503	0.9315
0.0001	34.9787	411	0.2508	0.9315
0.0001	36.0	423	0.2493	0.9315
0.0008	36.9362	434	0.2558	0.9315
0.0001	37.9574	446	0.2616	0.9315
0.0001	38.9787	458	0.2623	0.9315
0.0011	40.0	470	0.2617	0.9315
0.0002	40.9362	481	0.2532	0.9315
0.0001	41.9574	493	0.2495	0.9315
0.0001	42.9787	505	0.2478	0.9315
0.0001	44.0	517	0.2479	0.9315
0.0001	44.9362	528	0.2481	0.9315
0.0001	45.9574	540	0.2481	0.9315
0.002	46.8085	550	0.2475	0.9315

Framework versions

• Transformers 4.44.2
• Pytorch 2.5.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1