Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
@@ -1,10 +1,8 @@
|
|
1 |
-
# train_model.py
|
2 |
-
|
3 |
import torch
|
4 |
import torch.nn as nn
|
5 |
-
import
|
6 |
-
|
7 |
-
from
|
8 |
|
9 |
# Define the neural network model
|
10 |
class Net(nn.Module):
|
@@ -21,33 +19,35 @@ class Net(nn.Module):
|
|
21 |
x = self.fc3(x)
|
22 |
return torch.log_softmax(x, dim=1)
|
23 |
|
24 |
-
# Load
|
|
|
|
|
|
|
|
|
|
|
25 |
transform = transforms.Compose([
|
|
|
|
|
26 |
transforms.ToTensor(),
|
27 |
transforms.Normalize((0.5,), (0.5,))
|
28 |
])
|
29 |
|
30 |
-
|
31 |
-
|
32 |
-
|
33 |
-
|
34 |
-
|
35 |
-
|
36 |
-
|
37 |
-
|
38 |
-
#
|
39 |
-
|
40 |
-
|
41 |
-
|
42 |
-
|
43 |
-
|
44 |
-
|
45 |
-
|
46 |
-
|
47 |
-
|
48 |
-
|
49 |
-
train(model, train_loader, criterion, optimizer)
|
50 |
-
|
51 |
-
# Save the trained model
|
52 |
-
torch.save(model.state_dict(), 'mnist_model.pth')
|
53 |
-
print("Model saved as 'mnist_model.pth'")
|
|
|
|
|
|
|
1 |
import torch
|
2 |
import torch.nn as nn
|
3 |
+
import torchvision.transforms as transforms
|
4 |
+
import gradio as gr
|
5 |
+
from PIL import Image
|
6 |
|
7 |
# Define the neural network model
|
8 |
class Net(nn.Module):
|
|
|
19 |
x = self.fc3(x)
|
20 |
return torch.log_softmax(x, dim=1)
|
21 |
|
22 |
+
# Load the trained model
|
23 |
+
model = Net()
|
24 |
+
model.load_state_dict(torch.load('mnist_model.pth', map_location=torch.device('cpu')))
|
25 |
+
model.eval()
|
26 |
+
|
27 |
+
# Define the transform to preprocess the input image
|
28 |
transform = transforms.Compose([
|
29 |
+
transforms.Grayscale(num_output_channels=1),
|
30 |
+
transforms.Resize((28, 28)),
|
31 |
transforms.ToTensor(),
|
32 |
transforms.Normalize((0.5,), (0.5,))
|
33 |
])
|
34 |
|
35 |
+
# Define the prediction function
|
36 |
+
def predict(image):
|
37 |
+
image = transform(image).unsqueeze(0) # Add batch dimension
|
38 |
+
with torch.no_grad():
|
39 |
+
output = model(image)
|
40 |
+
prediction = torch.argmax(output, dim=1).item()
|
41 |
+
return prediction
|
42 |
+
|
43 |
+
# Create the Gradio interface
|
44 |
+
iface = gr.Interface(
|
45 |
+
fn=predict,
|
46 |
+
inputs=gr.inputs.Image(shape=(28, 28), image_mode='L', invert_colors=False),
|
47 |
+
outputs="label",
|
48 |
+
live=True
|
49 |
+
)
|
50 |
+
|
51 |
+
# Launch the Gradio interface
|
52 |
+
if __name__ == "__main__":
|
53 |
+
iface.launch()
|
|
|
|
|
|
|
|
|
|