File size: 1,166 Bytes
f85da4b
 
 
 
 
 
 
 
 
0277979
 
 
 
 
f85da4b
 
 
 
 
 
 
 
 
 
 
0277979
 
f85da4b
 
 
 
0277979
 
f85da4b
 
 
 
 
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
# Workaround to install the lib without "setup.py"
import sys
from git import Repo
Repo.clone_from("https://github.com/dimitreOliveira/hub.git", "./hub")
sys.path.append("/hub")

import gradio as gr
from hub.tensorflow_hub.hf_utils import pull_from_hub

# import requests
# # Download human-readable labels for ImageNet.
# response = requests.get("https://storage.googleapis.com/download.tensorflow.org/data/ImageNetLabels.txt")
# labels = [x for x in response.text.split("\n") if x != ""]

model = pull_from_hub(repo_id="Dimitre/mobilenet_v3_small")

def preprocess(image):
    print(image)
    print("***********")
    image = image.reshape((-1, 224, 224, 3))
    print(image)
    print("***********")
    print(image / 255.)
    return image / 255.

# def postprocess(prediction):
#     return {labels[i]: prediction[i] for i in range(len(labels))}

def predict_fn(image):
    image = preprocess(image)
    prediction = model([image])
    # scores = postprocess(prediction)
    return prediction

iface = gr.Interface(fn=predict_fn, 
                     inputs=gr.Image(shape=(224, 224)), 
                     outputs=gr.Label(num_top_classes=5))
iface.launch()