Spaces:
Running
Running
Added error handling for model prediction
Browse files
app.py
CHANGED
@@ -2,19 +2,53 @@ import gradio as gr
|
|
2 |
from deepforest import main
|
3 |
import cv2
|
4 |
|
5 |
-
|
6 |
def show_trees(img_path):
|
7 |
-
|
8 |
-
|
9 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
10 |
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
|
|
|
11 |
return img
|
12 |
|
13 |
def show_birds(img_path):
|
14 |
-
|
15 |
-
|
16 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
17 |
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
|
|
|
18 |
return img
|
19 |
|
20 |
with gr.Blocks() as demo:
|
@@ -36,6 +70,7 @@ with gr.Blocks() as demo:
|
|
36 |
input_image=gr.Image(label="Input image",type="filepath")
|
37 |
with gr.Column():
|
38 |
output_image=gr.Image(label="Predicted Image")
|
|
|
39 |
submit_button_birds = gr.Button("Predict birds")
|
40 |
submit_button_birds.click(show_birds,inputs=input_image,outputs=output_image)
|
41 |
|
|
|
2 |
from deepforest import main
|
3 |
import cv2
|
4 |
|
|
|
5 |
def show_trees(img_path):
|
6 |
+
try:
|
7 |
+
model = main.deepforest()
|
8 |
+
model.use_release()
|
9 |
+
except Exception as model_error:
|
10 |
+
raise gr.Error(f"Error initializing the deepforest model: {model_error}")
|
11 |
+
|
12 |
+
img=cv2.imread(img_path)
|
13 |
+
|
14 |
+
if img is None:
|
15 |
+
raise gr.Error(f"Image path is not valid.")
|
16 |
+
|
17 |
+
if img.shape[0]<1000 and img.shape[1]<1000:
|
18 |
+
img = model.predict_image(path=img_path, return_plot=True)
|
19 |
+
elif img.shape[0]>1000 or img.shape[1]>1000:
|
20 |
+
img=model.predict_image(path=img_path,return_plot=True,thickness=2)
|
21 |
+
else:
|
22 |
+
img = model.predict_image(path=img_path, return_plot=True,thickness=3)
|
23 |
+
|
24 |
+
if img is None:
|
25 |
+
raise gr.Error("No predictions were made. Check your test image. Ensure it conists")
|
26 |
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
|
27 |
+
|
28 |
return img
|
29 |
|
30 |
def show_birds(img_path):
|
31 |
+
try:
|
32 |
+
model = main.deepforest()
|
33 |
+
model.use_bird_release()
|
34 |
+
except Exception as model_error:
|
35 |
+
raise gr.Error(f"Error initializing the deepforest model: {model_error}")
|
36 |
+
img=cv2.imread(img_path)
|
37 |
+
|
38 |
+
if img is None:
|
39 |
+
raise gr.Error(f"Image path is not valid.")
|
40 |
+
|
41 |
+
if img.shape[0]<1000 and img.shape[1]<1000:
|
42 |
+
img = model.predict_image(path=img_path, return_plot=True)
|
43 |
+
elif img.shape[0]>1000 or img.shape[1]>1000:
|
44 |
+
img=model.predict_image(path=img_path,return_plot=True,thickness=2)
|
45 |
+
else:
|
46 |
+
img = model.predict_image(path=img_path, return_plot=True,thickness=3)
|
47 |
+
|
48 |
+
if img is None:
|
49 |
+
raise gr.Error("No predictions were made. Check your test image. Ensure it conists")
|
50 |
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
|
51 |
+
|
52 |
return img
|
53 |
|
54 |
with gr.Blocks() as demo:
|
|
|
70 |
input_image=gr.Image(label="Input image",type="filepath")
|
71 |
with gr.Column():
|
72 |
output_image=gr.Image(label="Predicted Image")
|
73 |
+
|
74 |
submit_button_birds = gr.Button("Predict birds")
|
75 |
submit_button_birds.click(show_birds,inputs=input_image,outputs=output_image)
|
76 |
|