akshaikrishna
commited on
Commit
·
3df2514
1
Parent(s):
e6119fe
Add application file
Browse files
app.py
ADDED
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import gradio as gr
|
2 |
+
import torch
|
3 |
+
from torchvision import models
|
4 |
+
from PIL import Image
|
5 |
+
from torch import nn
|
6 |
+
|
7 |
+
|
8 |
+
num_classes = 3
|
9 |
+
device = "cuda" if torch.cuda.is_available() else "cpu"
|
10 |
+
class_names = ["pizza", "steak", "sushi"]
|
11 |
+
class_dict = {"pizza": 0, "steak": 1, "sushi": 2}
|
12 |
+
|
13 |
+
|
14 |
+
def classify_image(image_path):
|
15 |
+
# Load the pre-trained model
|
16 |
+
model = models.mobilenet_v3_large(
|
17 |
+
weights=models.MobileNet_V3_Large_Weights.IMAGENET1K_V1
|
18 |
+
)
|
19 |
+
model.classifier[-1] = nn.Linear(model.classifier[-1].in_features, num_classes)
|
20 |
+
model.load_state_dict(
|
21 |
+
torch.load(
|
22 |
+
"MobileNetV3-Food-Classification.pth",
|
23 |
+
weights_only=True,
|
24 |
+
map_location=device,
|
25 |
+
)
|
26 |
+
)
|
27 |
+
model.to(device)
|
28 |
+
model.eval()
|
29 |
+
|
30 |
+
# Get the proper transforms directly from the weights
|
31 |
+
weights = models.MobileNet_V3_Large_Weights.IMAGENET1K_V1
|
32 |
+
preprocess = weights.transforms()
|
33 |
+
|
34 |
+
# Load and transform the image
|
35 |
+
image = Image.open(image_path)
|
36 |
+
input_tensor = preprocess(image)
|
37 |
+
|
38 |
+
# Add batch dimension
|
39 |
+
input_batch = input_tensor.unsqueeze(0)
|
40 |
+
|
41 |
+
# Move to GPU if available
|
42 |
+
input_batch = input_batch.to(device)
|
43 |
+
|
44 |
+
# Perform inference
|
45 |
+
with torch.no_grad():
|
46 |
+
output = model(input_batch)
|
47 |
+
|
48 |
+
# Get predictions
|
49 |
+
probabilities = torch.nn.functional.softmax(output[0], dim=0)
|
50 |
+
|
51 |
+
# Get the top prediction
|
52 |
+
top_prob, top_catid = torch.topk(probabilities, 1)
|
53 |
+
top_category = class_names[top_catid.item()] # type: ignore
|
54 |
+
top_probability = top_prob.item()
|
55 |
+
|
56 |
+
# Format the output string nicely
|
57 |
+
return f"Prediction: {top_category.title()} ({top_probability:.1%} confident)"
|
58 |
+
|
59 |
+
|
60 |
+
# Update the interface with label and better output type
|
61 |
+
demo = gr.Interface(
|
62 |
+
fn=classify_image,
|
63 |
+
inputs=gr.Image(
|
64 |
+
type="filepath", label="Upload a food image (pizza, steak, or sushi)"
|
65 |
+
),
|
66 |
+
outputs=gr.Text(label="Classification Result"),
|
67 |
+
title="Food Classifier",
|
68 |
+
description="This model classifies images of pizza, steak, and sushi.",
|
69 |
+
)
|
70 |
+
demo.launch()
|