Views
No views yet
Burdenthrive/cloud-detection-segformer-mit-b4segmentation_models_pytorch) with MiT‑B4 encoder.B×3×512×512, bands B04‑B03‑B02).mit_b4 (MiT encoder via segmentation_models_pytorch).B×4×512×512 (apply softmax + argmax).model.py, config.json, and weights.model.py — defines the SegFormer class (wrapper around smp.Segformer).config.json — hyperparameters and class names:
1{
2 "task": "image-segmentation",
3 "model_name": "segformer-mit-b4",
4 "model_kwargs": {
5 "encoder_name": "mit_b4",
6 "encoder_weights": "imagenet",
7 "in_channels": 3,
8 "num_classes": 4,
9 "freeze_encoder": false
10 },
11 "classes": ["clear", "thick cloud", "thin cloud", "cloud shadow"],
12 "id2label": { "0": "clear", "1": "thick cloud", "2": "thin cloud", "3": "cloud shadow" },
13 "label2id": { "clear": 0, "thick cloud": 1, "thin cloud": 2, "cloud shadow": 3 },
14 "input_bands": ["B04", "B03", "B02"]
15}