Include land cover colormaps of ESAWC and ECOSG

LBaudoux · Sep 4, 2023 · 3e0711e · 3e0711e
1 parent cfbd05a
commit 3e0711e
Showing 1 changed file with 175 additions and 4 deletions.
diff --git a/mmt/datasets/landcover_to_landcover.py b/mmt/datasets/landcover_to_landcover.py
@@ -151,6 +151,91 @@
         (0, 50, 200),
         (0, 0, 128),
     ],
+    "ecosg.hdf5": [
+        (0, 0, 0),
+        (0, 0, 128),
+        (0, 0, 205),
+        (0, 0, 255),
+        (211, 211, 211),
+        (169, 169, 169),
+        (255, 250, 250),
+        (240, 255, 240),
+        (85, 107, 47),
+        (154, 205, 50),
+        (0, 128, 0),
+        (255, 127, 80),
+        (160, 82, 45),
+        (34, 139, 34),
+        (188, 143, 143),
+        (205, 133, 63),
+        (222, 184, 135),
+        (50, 205, 50),
+        (255, 215, 0),
+        (32, 178, 170),
+        (173, 255, 47),
+        (189, 183, 107),
+        (102, 102, 0),
+        (46, 139, 87),
+        (138, 2, 0),
+        (206, 0, 0),
+        (252, 1, 1),
+        (255, 90, 0),
+        (255, 120, 0),
+        (255, 150, 0),
+        (255, 180, 0),
+        (255, 210, 0),
+        (255, 240, 0),
+        (128, 128, 128),
+    ],
+    "esgp.hdf5": [
+        (0, 0, 0),
+        (0, 0, 128),
+        (0, 0, 205),
+        (0, 0, 255),
+        (211, 211, 211),
+        (169, 169, 169),
+        (255, 250, 250),
+        (240, 255, 240),
+        (85, 107, 47),
+        (154, 205, 50),
+        (0, 128, 0),
+        (255, 127, 80),
+        (160, 82, 45),
+        (34, 139, 34),
+        (188, 143, 143),
+        (205, 133, 63),
+        (222, 184, 135),
+        (50, 205, 50),
+        (255, 215, 0),
+        (32, 178, 170),
+        (173, 255, 47),
+        (189, 183, 107),
+        (102, 102, 0),
+        (46, 139, 87),
+        (138, 2, 0),
+        (206, 0, 0),
+        (252, 1, 1),
+        (255, 90, 0),
+        (255, 120, 0),
+        (255, 150, 0),
+        (255, 180, 0),
+        (255, 210, 0),
+        (255, 240, 0),
+        (128, 128, 128),
+    ],
+    "esawc.hdf5": [
+        (0, 100, 0),
+        (255, 187, 34),
+        (255, 255, 76),
+        (240, 150, 255),
+        (250, 0, 0),
+        (180, 180, 180),
+        (240, 240, 240),
+        (0, 100, 200),
+        (0, 150, 160),
+        (0, 207, 117),
+        (250, 230, 160),
+    ],
 }
 label_dict = {
     "clc.hdf5": [
@@ -289,6 +374,91 @@
         "water",
         "ocean",
     ],
+    "ecosg.hdf5": [
+        "0. no data",
+        "1. sea and oceans",
+        "2. lakes",
+        "3. rivers",
+        "4. bare land",
+        "5. bare rock",
+        "6. permanent snow",
+        "7. boreal broadleaf deciduous",
+        "8. temperate broadleaf deciduous",
+        "9. tropical broadleaf deciduous",
+        "10. temperate broadleaf evergreen",
+        "11. tropical broadleaf evergreen",
+        "12. boreal needleleaf evergreen",
+        "13. temperate needleleaf evergreen",
+        "14. boreal needleleaf deciduous",
+        "15. shrubs",
+        "16. boreal grassland",
+        "17. temperate grassland",
+        "18. tropical grassland",
+        "19. winter C3 crops",
+        "20. summer C3 crops",
+        "21. C4 crops",
+        "22. flooded trees",
+        "23. flooded grassland",
+        "24. LCZ1: compact high-rise",
+        "25. LCZ2: compact midrise",
+        "26. LCZ3: compact low-rise",
+        "27. LCZ4: open high-rise",
+        "28. LCZ5: open midrise",
+        "29: LCZ6: open low-rise",
+        "30: LCZ7: lightweight low-rise",
+        "31: LCZ8: large low-rise",
+        "32: LCZ9: sparsely built",
+        "33: LCZ10: heavy industry",
+    ],
+    "esgp.hdf5": [
+        "0. no data",
+        "1. sea and oceans",
+        "2. lakes",
+        "3. rivers",
+        "4. bare land",
+        "5. bare rock",
+        "6. permanent snow",
+        "7. boreal broadleaf deciduous",
+        "8. temperate broadleaf deciduous",
+        "9. tropical broadleaf deciduous",
+        "10. temperate broadleaf evergreen",
+        "11. tropical broadleaf evergreen",
+        "12. boreal needleleaf evergreen",
+        "13. temperate needleleaf evergreen",
+        "14. boreal needleleaf deciduous",
+        "15. shrubs",
+        "16. boreal grassland",
+        "17. temperate grassland",
+        "18. tropical grassland",
+        "19. winter C3 crops",
+        "20. summer C3 crops",
+        "21. C4 crops",
+        "22. flooded trees",
+        "23. flooded grassland",
+        "24. LCZ1: compact high-rise",
+        "25. LCZ2: compact midrise",
+        "26. LCZ3: compact low-rise",
+        "27. LCZ4: open high-rise",
+        "28. LCZ5: open midrise",
+        "29: LCZ6: open low-rise",
+        "30: LCZ7: lightweight low-rise",
+        "31: LCZ8: large low-rise",
+        "32: LCZ9: sparsely built",
+        "33: LCZ10: heavy industry",
+    ],
+    "esawc.hdf5": [
+        "Tree cover",
+        "Shrubland",
+        "Grassland",
+        "Cropland",
+        "Built-up",
+        "Bare/sparse veg.",
+        "Snow and ice",
+        "Permanent water bodies",
+        "Herbaceous wetland",
+        "Mangroves",
+        "Moss and lichen",
+    ],
 }
 
 
@@ -625,7 +795,7 @@ def plot_samples_per_epoch(
                 vmin=0 - 0.5,
                 vmax=self.n_classes[dataset_tgt] + 0.5,
             )
-            ax[0][1].set_title("target")
+            ax[0][1].set_title("Target")
             m3 = ax[1][0].imshow(
                 outputs.cpu().long().numpy(),
                 cmap=cmap_tgt,
@@ -660,8 +830,8 @@ def plot_samples_per_epoch(
             emb = color_vector.reshape((n_pix, n_pix, 3), order="F").transpose(1, 0, 2)
             # emb = np.mean(embedding[0].cpu().numpy(), axis=0)
             # m4=  ax[1][1].imshow(emb/np.max(emb),vmin=np.percentile(emb,5),vmax=np.percentile(emb,95),cmap="jet")
-            m4 = ax[1][1].imshow(emb)
-            ax[1][1].set_title("embedding")
+            m4 = ax[1][1].imshow((emb * 255).astype(np.uint8))
+            ax[1][1].set_title("Embedding")
             # tell the colorbar to tick at integers
             f.colorbar(
                 m1, ticks=np.arange(0, self.n_classes[dataset_src] + 1), ax=ax[0][0]
@@ -673,8 +843,9 @@ def plot_samples_per_epoch(
                 m3, ticks=np.arange(0, self.n_classes[dataset_tgt] + 1), ax=ax[1][0]
             )
             # f.colorbar(m4,ax=ax[1][1])
+            acc = np.round((outputs == targets).sum().item()/targets.numel(), 3)
             f.suptitle(
-                "x={},y={}".format(coordinate[0][0].item(), coordinate[1][0].item())
+                f"x={coordinate[0][0].item()}, y={coordinate[1][0].item()}, accuracy={acc}"
             )
         f.savefig(title)
         plt.close(f)