File size: 2,443 Bytes
ac456a1 e99930d ac456a1 e99930d ac456a1 e99930d 7d86658 e99930d 7d86658 e99930d 7d86658 e99930d 7d86658 e99930d 7d86658 84c6ebc e99930d 7d86658 e99930d |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 |
import gradio as gr
import torch
import torch.nn as nn
import torch.nn.functional as F
import torchvision.transforms as transforms
from PIL import Image
import numpy as np
# Define the same model architecture
class ConvNet(nn.Module):
def __init__(self):
super().__init__()
self.conv1 = nn.Conv2d(3, 32, 3)
self.pool = nn.MaxPool2d(2, 2)
self.conv2 = nn.Conv2d(32, 64, 3)
self.conv3 = nn.Conv2d(64, 64, 3)
self.fc1 = nn.Linear(64 * 4 * 4, 64)
self.fc2 = nn.Linear(64, 10)
def forward(self, x):
x = F.relu(self.conv1(x))
x = self.pool(x)
x = F.relu(self.conv2(x))
x = self.pool(x)
x = F.relu(self.conv3(x))
x = torch.flatten(x, 1)
x = F.relu(self.fc1(x))
x = self.fc2(x)
return x
# Initialize model and load weights
model = ConvNet()
model.load_state_dict(torch.load('cnn.pth', map_location=torch.device('cpu')))
model.eval()
# Define classes
classes = ('plane', 'car', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse', 'ship', 'truck')
# Define preprocessing
transform = transforms.Compose([
transforms.Resize((32, 32)),
transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))
])
def predict(img):
if img is None:
return None
# Convert to PIL Image if needed
if not isinstance(img, Image.Image):
img = Image.fromarray(img)
# Preprocess the image
img = transform(img).unsqueeze(0)
# Get predictions
with torch.no_grad():
outputs = model(img)
probabilities = F.softmax(outputs, dim=1)[0]
predictions = {
classes[i]: float(probabilities[i]) * 100 # Convert to percentage
for i in range(len(classes))
}
# Sort predictions by probability
sorted_predictions = dict(sorted(predictions.items(), key=lambda x: x[1], reverse=True))
return sorted_predictions
iface = gr.Interface(
fn=predict,
inputs=gr.Image(type="pil"),
outputs=gr.Label(num_top_classes=10), # Show all 10 classes
examples=[["example1.jpeg"], ["example2.jpeg"]], # Optional: Add example images
title="CIFAR-10 Image Classifier",
description="Upload an image to classify it into one of these categories: plane, car, bird, cat, deer, dog, frog, horse, ship, or truck. Results show prediction confidence for all classes as percentages."
)
iface.launch() |