This model is a fine-tuned version of nvidia/mit-b0 on the LeninGF/uff-thermography-multiview-segmentation dataset. It achieves the following results on the evaluation set:

Loss: 0.0878
Mean Iou: 0.9213
Mean Accuracy: 0.9619
Overall Accuracy: 0.9605
Accuracy Background: 0.9553
Accuracy Roi: 0.9685
Iou Background: 0.9359
Iou Roi: 0.9067
Roi Precision: 0.9343
Roi Recall: 0.9685
Roi F1: 0.9511
Roi Dice: 0.9511
Roi Specificity: 0.9553

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: 1e-05
train_batch_size: 4
eval_batch_size: 4
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Background	Accuracy Roi	Iou Background	Iou Roi	Roi Precision	Roi Recall	Roi F1	Roi Dice	Roi Specificity
0.3274	1.0	124	0.3163	0.6952	0.8467	0.8206	0.7209	0.9725	0.7081	0.6822	0.6957	0.9725	0.8111	0.8111	0.7209
0.2583	2.0	248	0.2382	0.7916	0.8983	0.8853	0.8356	0.9611	0.8148	0.7685	0.7932	0.9611	0.8691	0.8691	0.8356
0.2154	3.0	372	0.2100	0.8052	0.9083	0.8934	0.8366	0.9800	0.8257	0.7846	0.7973	0.9800	0.8793	0.8793	0.8366
0.2206	4.0	496	0.1878	0.8413	0.9254	0.9155	0.8775	0.9733	0.8624	0.8202	0.8391	0.9733	0.9012	0.9012	0.8775
0.2005	5.0	620	0.1602	0.8769	0.9401	0.9363	0.9220	0.9582	0.8974	0.8564	0.8896	0.9582	0.9226	0.9226	0.9220
0.1672	6.0	744	0.1496	0.8751	0.9411	0.9351	0.9121	0.9701	0.8946	0.8556	0.8787	0.9701	0.9222	0.9222	0.9121
0.1584	7.0	868	0.1457	0.8733	0.9413	0.9340	0.9062	0.9765	0.8924	0.8543	0.8722	0.9765	0.9214	0.9214	0.9062
0.1650	8.0	992	0.1349	0.8847	0.9448	0.9406	0.9246	0.9651	0.9039	0.8656	0.8936	0.9651	0.9280	0.9280	0.9246
0.1242	9.0	1116	0.1344	0.8822	0.9457	0.9390	0.9132	0.9782	0.9004	0.8640	0.8809	0.9782	0.9270	0.9270	0.9132
0.1423	10.0	1240	0.1251	0.8923	0.9491	0.9448	0.9284	0.9698	0.9103	0.8743	0.8988	0.9698	0.9329	0.9329	0.9284
0.1359	11.0	1364	0.1216	0.8948	0.9499	0.9462	0.9319	0.9679	0.9127	0.8769	0.9031	0.9679	0.9344	0.9344	0.9319
0.1415	12.0	1488	0.1207	0.8931	0.9503	0.9451	0.9251	0.9755	0.9105	0.8756	0.8953	0.9755	0.9337	0.9337	0.9251
0.1076	13.0	1612	0.1222	0.8909	0.9494	0.9439	0.9227	0.9761	0.9085	0.8733	0.8923	0.9761	0.9323	0.9323	0.9227
0.1299	14.0	1736	0.1111	0.9029	0.9538	0.9506	0.9382	0.9695	0.9198	0.8860	0.9114	0.9695	0.9396	0.9396	0.9382
0.1018	15.0	1860	0.1128	0.8991	0.9532	0.9484	0.9301	0.9763	0.9159	0.8823	0.9017	0.9763	0.9375	0.9375	0.9301
0.1195	16.0	1984	0.1177	0.8946	0.9510	0.9460	0.9269	0.9750	0.9120	0.8773	0.8974	0.9750	0.9346	0.9346	0.9269
0.1092	17.0	2108	0.1087	0.9019	0.9544	0.9500	0.9330	0.9759	0.9184	0.8854	0.9052	0.9759	0.9392	0.9392	0.9330
0.1051	18.0	2232	0.1052	0.9058	0.9554	0.9522	0.9400	0.9708	0.9223	0.8894	0.9138	0.9708	0.9415	0.9415	0.9400
0.0945	19.0	2356	0.1076	0.9037	0.9544	0.9510	0.9381	0.9706	0.9204	0.8870	0.9115	0.9706	0.9401	0.9401	0.9381
0.1093	20.0	2480	0.1021	0.9080	0.9575	0.9533	0.9372	0.9777	0.9237	0.8923	0.9108	0.9777	0.9431	0.9431	0.9372
0.1270	21.0	2604	0.1002	0.9137	0.9578	0.9565	0.9513	0.9644	0.9296	0.8978	0.9285	0.9644	0.9461	0.9461	0.9513
0.0953	22.0	2728	0.0973	0.9141	0.9585	0.9567	0.9498	0.9673	0.9298	0.8985	0.9267	0.9673	0.9465	0.9465	0.9498
0.1226	23.0	2852	0.0976	0.9158	0.9578	0.9577	0.9574	0.9581	0.9318	0.8997	0.9366	0.9581	0.9472	0.9472	0.9574
0.0806	24.0	2976	0.0965	0.9129	0.9592	0.9559	0.9432	0.9753	0.9281	0.8976	0.9185	0.9753	0.9460	0.9460	0.9432
0.0877	25.0	3100	0.0964	0.9132	0.9587	0.9561	0.9464	0.9710	0.9287	0.8976	0.9223	0.9710	0.9461	0.9461	0.9464
0.0982	26.0	3224	0.1004	0.9063	0.9574	0.9523	0.9328	0.9819	0.9219	0.8907	0.9056	0.9819	0.9422	0.9422	0.9328
0.1049	27.0	3348	0.0974	0.9117	0.9585	0.9553	0.9433	0.9737	0.9273	0.8962	0.9184	0.9737	0.9453	0.9453	0.9433
0.0758	28.0	3472	0.0938	0.9143	0.9600	0.9567	0.9440	0.9759	0.9293	0.8992	0.9196	0.9759	0.9469	0.9469	0.9440
0.0989	29.0	3596	0.0922	0.9179	0.9602	0.9587	0.9533	0.9670	0.9331	0.9028	0.9314	0.9670	0.9489	0.9489	0.9533
0.0779	30.0	3720	0.0929	0.9166	0.9600	0.9580	0.9506	0.9693	0.9318	0.9014	0.9279	0.9693	0.9482	0.9482	0.9506
0.0661	31.0	3844	0.0902	0.9190	0.9612	0.9592	0.9519	0.9704	0.9338	0.9041	0.9297	0.9704	0.9497	0.9497	0.9519
0.0701	32.0	3968	0.0953	0.9130	0.9591	0.9560	0.9440	0.9743	0.9283	0.8977	0.9194	0.9743	0.9461	0.9461	0.9440
0.0952	33.0	4092	0.0924	0.9178	0.9593	0.9587	0.9563	0.9624	0.9333	0.9023	0.9353	0.9624	0.9486	0.9486	0.9563
0.0834	34.0	4216	0.0915	0.9160	0.9604	0.9576	0.9470	0.9737	0.9310	0.9010	0.9234	0.9737	0.9479	0.9479	0.9470
0.0877	35.0	4340	0.0897	0.9186	0.9611	0.9590	0.9513	0.9708	0.9334	0.9037	0.9290	0.9708	0.9494	0.9494	0.9513
0.0650	36.0	4464	0.0900	0.9189	0.9610	0.9592	0.9525	0.9695	0.9338	0.9041	0.9306	0.9695	0.9496	0.9496	0.9525
0.0811	37.0	4588	0.0908	0.9181	0.9606	0.9588	0.9516	0.9697	0.9330	0.9031	0.9293	0.9697	0.9491	0.9491	0.9516
0.0823	38.0	4712	0.0891	0.9203	0.9614	0.9600	0.9544	0.9684	0.9350	0.9055	0.9330	0.9684	0.9504	0.9504	0.9544
0.0631	39.0	4836	0.0919	0.9156	0.9604	0.9574	0.9458	0.9751	0.9305	0.9006	0.9218	0.9751	0.9477	0.9477	0.9458
0.0731	40.0	4960	0.0892	0.9195	0.9612	0.9595	0.9530	0.9695	0.9343	0.9047	0.9312	0.9695	0.9499	0.9499	0.9530
0.0804	41.0	5084	0.0883	0.9207	0.9617	0.9602	0.9545	0.9688	0.9354	0.9060	0.9332	0.9688	0.9507	0.9507	0.9545
0.0765	42.0	5208	0.0904	0.9184	0.9609	0.9589	0.9516	0.9701	0.9333	0.9034	0.9293	0.9701	0.9493	0.9493	0.9516
0.0706	43.0	5332	0.0886	0.9210	0.9614	0.9604	0.9564	0.9664	0.9358	0.9061	0.9357	0.9664	0.9508	0.9508	0.9564
0.0753	44.0	5456	0.0890	0.9198	0.9612	0.9597	0.9538	0.9687	0.9346	0.9049	0.9322	0.9687	0.9501	0.9501	0.9538
0.0834	45.0	5580	0.0897	0.9194	0.9610	0.9595	0.9536	0.9685	0.9343	0.9045	0.9319	0.9685	0.9498	0.9498	0.9536
0.0833	46.0	5704	0.0905	0.9180	0.9611	0.9587	0.9496	0.9726	0.9328	0.9031	0.9268	0.9726	0.9491	0.9491	0.9496
0.0812	47.0	5828	0.0886	0.9207	0.9616	0.9602	0.9550	0.9681	0.9354	0.9060	0.9338	0.9681	0.9507	0.9507	0.9550
0.0959	48.0	5952	0.0886	0.9202	0.9616	0.9599	0.9531	0.9702	0.9348	0.9055	0.9314	0.9702	0.9504	0.9504	0.9531
0.0757	49.0	6076	0.0877	0.9215	0.9618	0.9606	0.9559	0.9678	0.9361	0.9068	0.9350	0.9678	0.9511	0.9511	0.9559
0.0732	50.0	6200	0.0878	0.9213	0.9619	0.9605	0.9553	0.9685	0.9359	0.9067	0.9343	0.9685	0.9511	0.9511	0.9553

Framework versions

Transformers 5.12.1
Pytorch 2.6.0+cu118
Datasets 5.0.0
Tokenizers 0.22.2

Experimental Setup and Reproducibility Configuration

This section is auto-synced from the notebook run configuration.

Run Summary

saved_at: 2026-07-17T22:34:04.786119-05:00
run_tag: all5-vc1-pc0-cw1-e50-260717-2234
dataset: LeninGF/uff-thermography-multiview-segmentation
selected_view: all_canonical_views
use_view_conditioning: True. When True, the model receives a view ID embedding to condition predictions on the view.
use_pseudo_color: False (map=inferno). When True, the model receives a pseudo-colored image instead of the original grayscale image.
epochs: 50
learning_rate: 1e-05
ce_weight/dice_weight: 0.5/0.5
use_train_augmentation: True. When True, the training dataset is augmented with random transformations to improve generalization.
aug_prob_hflip: 0.0. If 0.0, horizontal flips are disabled; if 1.0, all training images are flipped; if in (0.0, 1.0), flips occur with the given probability.
use_class_weights: True. When True, the loss function uses class weights to handle class imbalance at pixel level.

Full Experiment Artifact

JSON artifact: experiment_conditions.json

JSON Snapshot

json

{
  "saved_at": "2026-07-17T22:34:04.786119-05:00",
  "dataset": {
    "hf_dataset_identifier": "LeninGF/uff-thermography-multiview-segmentation",
    "shuffle_seed": 42,
    "split_seed": 42
  },
  "view_selection": {
    "selected_view": "all_canonical_views",
    "use_view_conditioning": true
  },
  "augmentation_preprocessing": {
    "use_train_augmentation": true,
    "aug_prob_hflip": 0.0,
    "aug_prob_vflip": 0.0,
    "aug_max_rotation_deg": 5.0,
    "aug_max_translate": 0.03,
    "aug_scale_min": 0.95,
    "aug_scale_max": 1.05,
    "aug_brightness": 0.1,
    "aug_contrast": 0.1,
    "use_pseudo_color": false,
    "pseudo_color_map": "inferno"
  },
  "model": {
    "pretrained_model_name": "nvidia/mit-b0",
    "view_embed_dim": 32
  },
  "training": {
    "epochs": 50,
    "lr": 1e-05,
    "use_dynamic_batch_size": true,
    "batch_size_override": null,
    "use_fp16": false,
    "save_total_limit": 2,
    "eval_strategy": "epoch",
    "save_strategy": "epoch",
    "logging_steps": 10,
    "eval_accumulation_steps": 5,
    "load_best_model_at_end": true,
    "greater_is_better": true,
    "max_grad_norm": 1.0
  },
  "loss_metrics": {
    "IGNORE_INDEX": 255,
    "ce_weight": 0.5,
    "dice_weight": 0.5,
    "use_class_weights": true
  },
  "hub_output": {
    "hf_username": "LeninGF",
    "push_to_hub": true,
    "hub_private_repo": true,
    "hub_strategy": "end",
    "load_from_local_best": true,
    "hub_model_id": "segformer-b0-FT-uff-roi-segmentation-all5-vc1-pc0-cw1-e50-260717-2234",
    "output_dir": "segformer-b0-FT-uff-roi-out-all5-vc1-pc0-cw1-e50-260717-2234",
    "best_model_dir": "segformer-b0-FT-uff-roi-out-all5-vc1-pc0-cw1-e50-260717-2234/best-model",
    "images_dir": "segformer-b0-FT-uff-roi-out-all5-vc1-pc0-cw1-e50-260717-2234/images"
  },
  "naming": {
    "name_base": "segformer-b0-FT-uff-roi-segmentation",
    "view_code": "all5",
    "use_view_conditioning(vc_code)": "vc1",
    "use_pseudo_color(pc_code)": "pc0",
    "class_weights_code": "cw1",
    "exp_code": "all5-vc1-pc0-cw1-e50",
    "run_datetime": "260717-2234",
    "run_tag": "all5-vc1-pc0-cw1-e50-260717-2234",
    "max_name_len": 96,
    "max_tag_len": 36
  },
  "benchmark": {
    "benchmark_n_runs": 5,
    "benchmark_samples_per_run": 20,
    "benchmark_show_first_run_panel": true
  },
  "hub_tags": [
    "vision",
    "image-segmentation",
    "medical-imaging",
    "thermography",
    "multiview",
    "breast-cancer",
    "roi-segmentation"
  ]
}

LeninGF/segformer-b0-FT-uff-roi-segmentation-all5-vc1-pc0-cw1-e50-260717-2234

segformer-b0-FT-uff-roi-segmentation-all5-vc1-pc0-cw1-e50-260717-2234

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Experimental Setup and Reproducibility Configuration

Run Summary

Full Experiment Artifact

JSON Snapshot

LeninGF/
segformer-b0-FT-uff-roi-segmentation-all5-vc1-pc0-cw1-e50-260717-2234