Spaces:
Sleeping
Sleeping
wip
Browse files
app.py
CHANGED
@@ -13,6 +13,8 @@ def get_index_of_element_containing_word(lst, word):
|
|
13 |
return indices[0] if indices else -1
|
14 |
|
15 |
pred_global = None
|
|
|
|
|
16 |
|
17 |
stl_preds = np.load("stl_species.npy")
|
18 |
df = pd.read_csv("gbif_full_filtered.csv")
|
@@ -36,30 +38,45 @@ def update_fn(val):
|
|
36 |
return gr.Dropdown(label="Name", choices=obs, interactive=True)
|
37 |
|
38 |
def text_fn(taxon, name):
|
39 |
-
global pred_global
|
40 |
|
41 |
species_index = get_index_of_element_containing_word(obs, name)
|
42 |
preds = np.flip(stl_preds[:, species_index].reshape(510, 510), 1)
|
43 |
|
44 |
pred_global = preds
|
|
|
45 |
cmap = plt.get_cmap('plasma')
|
46 |
|
47 |
rgba_img = cmap(preds)
|
48 |
rgb_img = np.delete(rgba_img, 3, 2)
|
49 |
-
blend = Image.blend(stl_base, Image.fromarray((rgb_img * 255).astype(np.uint8)),
|
50 |
rgb_img = np.array(blend)
|
51 |
#return gr.Image(preds, label="Predicted Heatmap", visible=True)
|
52 |
return rgb_img
|
53 |
|
54 |
def thresh_fn(val):
|
55 |
-
global pred_global
|
56 |
preds = deepcopy(pred_global)
|
57 |
preds[preds<val] = 0
|
58 |
preds[preds>=val] = 1
|
|
|
59 |
cmap = plt.get_cmap('plasma')
|
60 |
|
61 |
rgba_img = cmap(preds)
|
62 |
rgb_img = np.delete(rgba_img, 3, 2)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
63 |
return rgb_img
|
64 |
|
65 |
with gr.Blocks() as demo:
|
@@ -79,10 +96,14 @@ with gr.Blocks() as demo:
|
|
79 |
with gr.Row():
|
80 |
slider = gr.Slider(minimum=0, maximum=1, step=0.01, value=0.5, label="Confidence Threshold")
|
81 |
|
|
|
|
|
|
|
82 |
with gr.Row():
|
83 |
pred = gr.Image(label="Predicted Heatmap", visible=True)
|
84 |
|
85 |
check_button.click(text_fn, inputs=[inp, out], outputs=[pred])
|
86 |
slider.change(thresh_fn, slider, outputs=pred)
|
|
|
87 |
|
88 |
demo.launch()
|
|
|
13 |
return indices[0] if indices else -1
|
14 |
|
15 |
pred_global = None
|
16 |
+
alpha_global = 0.5
|
17 |
+
alpha_image = None
|
18 |
|
19 |
stl_preds = np.load("stl_species.npy")
|
20 |
df = pd.read_csv("gbif_full_filtered.csv")
|
|
|
38 |
return gr.Dropdown(label="Name", choices=obs, interactive=True)
|
39 |
|
40 |
def text_fn(taxon, name):
|
41 |
+
global pred_global, alpha_global, alpha_image
|
42 |
|
43 |
species_index = get_index_of_element_containing_word(obs, name)
|
44 |
preds = np.flip(stl_preds[:, species_index].reshape(510, 510), 1)
|
45 |
|
46 |
pred_global = preds
|
47 |
+
alpha_image = preds
|
48 |
cmap = plt.get_cmap('plasma')
|
49 |
|
50 |
rgba_img = cmap(preds)
|
51 |
rgb_img = np.delete(rgba_img, 3, 2)
|
52 |
+
blend = Image.blend(stl_base, Image.fromarray((rgb_img * 255).astype(np.uint8)), alpha_global)
|
53 |
rgb_img = np.array(blend)
|
54 |
#return gr.Image(preds, label="Predicted Heatmap", visible=True)
|
55 |
return rgb_img
|
56 |
|
57 |
def thresh_fn(val):
|
58 |
+
global pred_global, alpha_global, alpha_image
|
59 |
preds = deepcopy(pred_global)
|
60 |
preds[preds<val] = 0
|
61 |
preds[preds>=val] = 1
|
62 |
+
alpha_image = deepcopy(preds)
|
63 |
cmap = plt.get_cmap('plasma')
|
64 |
|
65 |
rgba_img = cmap(preds)
|
66 |
rgb_img = np.delete(rgba_img, 3, 2)
|
67 |
+
blend = Image.blend(stl_base, Image.fromarray((rgb_img * 255).astype(np.uint8)), alpha_global)
|
68 |
+
rgb_img = np.array(blend)
|
69 |
+
return rgb_img
|
70 |
+
|
71 |
+
def alpha_fn(val):
|
72 |
+
global pred_global, alpha_global, alpha_image
|
73 |
+
alpha_global = val
|
74 |
+
preds = deepcopy(alpha_image)
|
75 |
+
cmap = plt.get_cmap('plasma')
|
76 |
+
rgba_img = cmap(preds)
|
77 |
+
rgb_img = np.delete(rgba_img, 3, 2)
|
78 |
+
blend = Image.blend(stl_base, Image.fromarray((rgb_img * 255).astype(np.uint8)), alpha_global)
|
79 |
+
rgb_img = np.array(blend)
|
80 |
return rgb_img
|
81 |
|
82 |
with gr.Blocks() as demo:
|
|
|
96 |
with gr.Row():
|
97 |
slider = gr.Slider(minimum=0, maximum=1, step=0.01, value=0.5, label="Confidence Threshold")
|
98 |
|
99 |
+
with gr.Row():
|
100 |
+
alpha = gr.Slider(minimum=0, maximum=1, step=0.01, value=0.5, label="Image Transparency")
|
101 |
+
|
102 |
with gr.Row():
|
103 |
pred = gr.Image(label="Predicted Heatmap", visible=True)
|
104 |
|
105 |
check_button.click(text_fn, inputs=[inp, out], outputs=[pred])
|
106 |
slider.change(thresh_fn, slider, outputs=pred)
|
107 |
+
alpha.change(alpha_fn, alpha, outputs=pred)
|
108 |
|
109 |
demo.launch()
|