Spaces:
Sleeping
Sleeping
Upload app.py
Browse files
app.py
ADDED
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
|
2 |
+
import gradio as gr
|
3 |
+
import os
|
4 |
+
import torch
|
5 |
+
|
6 |
+
from model import create_effnetb2_model
|
7 |
+
from timeit import default_timer as timer
|
8 |
+
from typing import Tuple,Dict
|
9 |
+
|
10 |
+
with open('class_names.txt','r') as f:
|
11 |
+
class_names = [food.strip() for food in f.readlines()]
|
12 |
+
|
13 |
+
effnetb2 , effnetb2_transforms = create_effnetb2_model(num_classes=101)
|
14 |
+
|
15 |
+
effnetb2.load_state_dict(torch.load(f='09_pretrained_effnetb2_feature_extractor_food101_20_percent.pth',
|
16 |
+
map_location= torch.device('cpu')))
|
17 |
+
|
18 |
+
def predict(img) -> Tuple[Dict,float]:
|
19 |
+
|
20 |
+
start_time = timer()
|
21 |
+
|
22 |
+
img = effnetb2_transforms(img).unsqueeze(0)
|
23 |
+
|
24 |
+
effnetb2.eval()
|
25 |
+
with torch.inference_mode():
|
26 |
+
pred_prob = torch.softmax(effnetb2(img),dim=1)
|
27 |
+
pred_labels_and_probs = {class_names[i]:float(pred_prob[0][i])for i in range(len(class_names))}
|
28 |
+
end_time = timer()
|
29 |
+
pred_time = round(end_time - start_time,4)
|
30 |
+
return pred_labels_and_probs,pred_time
|
31 |
+
|
32 |
+
|
33 |
+
|
34 |
+
|
35 |
+
# Create title, description and article strings
|
36 |
+
title = "FoodVision Big ππ"
|
37 |
+
description = "An EfficientNetB2 feature extractor computer vision model to classify images of food into [101 different classes](https://github.com/mrdbourke/pytorch-deep-learning/blob/main/extras/food101_class_names.txt)."
|
38 |
+
example_list = [['examples'+example] for example in os.listdir('examples') ]
|
39 |
+
|
40 |
+
# Create the Gradio demo
|
41 |
+
demo = gr.Interface(fn=predict, # mapping function from input to output
|
42 |
+
inputs=gr.Image(type="pil"), # what are the inputs?
|
43 |
+
outputs=[gr.Label(num_top_classes=3, label="Predictions"), # what are the outputs?
|
44 |
+
gr.Number(label="Prediction time (s)")], # our fn has two outputs, therefore we have two outputs
|
45 |
+
examples=example_list,
|
46 |
+
title=title,
|
47 |
+
description=description,
|
48 |
+
)
|
49 |
+
|
50 |
+
# Launch the demo!
|
51 |
+
demo.launch() # generate a publically shareable URL?
|