clear working
Browse files
app.py
CHANGED
@@ -8,6 +8,9 @@ from models import Net,NetConv
|
|
8 |
net = torch.load('mnist.pth')
|
9 |
net.eval()
|
10 |
|
|
|
|
|
|
|
11 |
def predict(img):
|
12 |
arr = np.array(img) / 255 # Assuming img is in the range [0, 255]
|
13 |
arr = np.expand_dims(arr, axis=0) # Add batch dimension
|
@@ -16,21 +19,53 @@ def predict(img):
|
|
16 |
topk_values, topk_indices = torch.topk(output, 2) # Get the top 2 classes
|
17 |
return [str(k) for k in topk_indices[0].tolist()]
|
18 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
19 |
with gr.Blocks() as iface:
|
20 |
gr.Markdown("# MNIST + Gradio End to End")
|
21 |
gr.HTML("Shows end to end MNIST training with Gradio interface")
|
22 |
-
with gr.
|
23 |
-
with gr.
|
24 |
-
|
25 |
-
|
26 |
-
with gr.
|
27 |
-
|
28 |
-
|
29 |
-
|
30 |
-
|
31 |
-
|
32 |
-
|
33 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
34 |
pred_button.click(predict, inputs=sp, outputs=[label1,label2])
|
35 |
-
|
|
|
|
|
|
|
|
|
36 |
iface.launch()
|
|
|
8 |
net = torch.load('mnist.pth')
|
9 |
net.eval()
|
10 |
|
11 |
+
net_conv = torch.load('mnist_conv.pth')
|
12 |
+
net_conv.eval()
|
13 |
+
|
14 |
def predict(img):
|
15 |
arr = np.array(img) / 255 # Assuming img is in the range [0, 255]
|
16 |
arr = np.expand_dims(arr, axis=0) # Add batch dimension
|
|
|
19 |
topk_values, topk_indices = torch.topk(output, 2) # Get the top 2 classes
|
20 |
return [str(k) for k in topk_indices[0].tolist()]
|
21 |
|
22 |
+
def predict_conv(img):
|
23 |
+
arr = np.array(img) / 255 # Assuming img is in the range [0, 255]
|
24 |
+
arr = np.expand_dims(arr, axis=0) # Conv needs one more dimension
|
25 |
+
arr = np.expand_dims(arr, axis=0) # Add batch dimension
|
26 |
+
arr = torch.from_numpy(arr).float() # Convert to PyTorch tensor
|
27 |
+
output = net_conv(arr)
|
28 |
+
topk_values, topk_indices = torch.topk(output, 2) # Get the top 2 classes
|
29 |
+
return [str(k) for k in topk_indices[0].tolist()]
|
30 |
+
|
31 |
+
|
32 |
+
|
33 |
+
|
34 |
+
|
35 |
with gr.Blocks() as iface:
|
36 |
gr.Markdown("# MNIST + Gradio End to End")
|
37 |
gr.HTML("Shows end to end MNIST training with Gradio interface")
|
38 |
+
with gr.Tab("Linear Model"):
|
39 |
+
with gr.Row():
|
40 |
+
with gr.Column():
|
41 |
+
sp = gr.Sketchpad(shape=(28, 28))
|
42 |
+
with gr.Row():
|
43 |
+
with gr.Column():
|
44 |
+
pred_button = gr.Button("Predict")
|
45 |
+
with gr.Column():
|
46 |
+
clear_button = gr.Button("Clear")
|
47 |
+
with gr.Column():
|
48 |
+
label1 = gr.Label(label='1st Pred')
|
49 |
+
label2 = gr.Label(label='2nd Pred')
|
50 |
+
|
51 |
+
with gr.Tab("Convolution Model"):
|
52 |
+
with gr.Row():
|
53 |
+
with gr.Column():
|
54 |
+
sp_conv = gr.Sketchpad(shape=(28, 28))
|
55 |
+
with gr.Row():
|
56 |
+
with gr.Column():
|
57 |
+
pred_conv_button = gr.Button("Predict")
|
58 |
+
with gr.Column():
|
59 |
+
clear_button_conv = gr.Button("Clear")
|
60 |
+
with gr.Column():
|
61 |
+
label1_conv = gr.Label(label='1st Pred')
|
62 |
+
label2_conv = gr.Label(label='2nd Pred')
|
63 |
+
def clear():
|
64 |
+
return ['','',None,'','',None]
|
65 |
pred_button.click(predict, inputs=sp, outputs=[label1,label2])
|
66 |
+
pred_conv_button.click(predict_conv, inputs=sp_conv, outputs=[label1_conv,label2_conv])
|
67 |
+
clear_button.click( lambda: ['','',None], None, [label1,label2,sp,], queue=False)
|
68 |
+
clear_button_conv.click( lambda: ['','',None], None, [label1_conv,label2_conv, sp_conv], queue=False)
|
69 |
+
|
70 |
+
|
71 |
iface.launch()
|
mnist.pth
CHANGED
Binary files a/mnist.pth and b/mnist.pth differ
|
|