segformer-ore-minerals-v1

This model is a fine-tuned version of nvidia/segformer-b0-finetuned-ade-512-512 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3128

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: 6e-05
• train_batch_size: 6
• eval_batch_size: 6
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 12
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
1.4449	1.0	36	1.2149
0.9787	2.0	72	0.9250
0.8349	3.0	108	0.7390
0.6549	4.0	144	0.6332
0.5758	5.0	180	0.5384
0.4905	6.0	216	0.4754
0.4525	7.0	252	0.4232
0.4037	8.0	288	0.3887
0.3464	9.0	324	0.3776
0.3394	10.0	360	0.3582
0.3105	11.0	396	0.3278
0.2821	12.0	432	0.3397
0.2691	13.0	468	0.3194
0.2642	14.0	504	0.3303
0.2784	15.0	540	0.3271
0.2477	16.0	576	0.3144
0.2472	17.0	612	0.3330
0.2339	18.0	648	0.3263
0.2492	19.0	684	0.3235
0.2226	20.0	720	0.3046
0.2414	21.0	756	0.3087
0.2228	22.0	792	0.3016
0.2023	23.0	828	0.2947
0.1925	24.0	864	0.3021
0.2405	25.0	900	0.3051
0.2327	26.0	936	0.3307
0.2014	27.0	972	0.2994
0.2169	28.0	1008	0.3049
0.1995	29.0	1044	0.3144
0.2151	30.0	1080	0.3128

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu126
• Datasets 4.4.1
• Tokenizers 0.22.1