Update app.py
Browse files
app.py
CHANGED
@@ -1,11 +1,72 @@
|
|
1 |
from flask import Flask,render_template
|
2 |
from flask_socketio import SocketIO,emit
|
3 |
import base64
|
|
|
|
|
|
|
4 |
import numpy as np
|
5 |
import cv2
|
6 |
-
import numpy as np
|
7 |
|
8 |
|
9 |
app = Flask(__name__)
|
10 |
app.config['SECRET_KEY'] = 'secret!'
|
11 |
socket = SocketIO(app,async_mode="eventlet")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
from flask import Flask,render_template
|
2 |
from flask_socketio import SocketIO,emit
|
3 |
import base64
|
4 |
+
|
5 |
+
from keras.models import load_model
|
6 |
+
from PIL import Image
|
7 |
import numpy as np
|
8 |
import cv2
|
|
|
9 |
|
10 |
|
11 |
app = Flask(__name__)
|
12 |
app.config['SECRET_KEY'] = 'secret!'
|
13 |
socket = SocketIO(app,async_mode="eventlet")
|
14 |
+
|
15 |
+
|
16 |
+
#the following are to do with this interactive notebook code
|
17 |
+
from matplotlib import pyplot as plt # this lets you draw inline pictures in the notebooks
|
18 |
+
import pylab # this allows you to control figure size
|
19 |
+
pylab.rcParams['figure.figsize'] = (10.0, 8.0) # this controls figure size in the notebook
|
20 |
+
|
21 |
+
###loading model###
|
22 |
+
age_model = load_model('Copy of age_model_pretrained.h5')
|
23 |
+
gender_model = load_model('Copy of gender_model_pretrained.h5')
|
24 |
+
emotion_model = load_model('emotion_model_pretrained.h5')
|
25 |
+
|
26 |
+
# Labels on Age, Gender and Emotion to be predicted
|
27 |
+
age_ranges = ['1-2', '3-9', '10-20', '21-27', '28-45', '46-65', '66-116']
|
28 |
+
gender_ranges = ['male', 'female']
|
29 |
+
emotion_ranges= ['positive','negative','neutral']
|
30 |
+
|
31 |
+
def base64_to_image(base64_string):
|
32 |
+
# Extract the base64 encoded binary data from the input string
|
33 |
+
base64_data = base64_string.split(",")[1]
|
34 |
+
# Decode the base64 data to bytes
|
35 |
+
image_bytes = base64.b64decode(base64_data)
|
36 |
+
# Convert the bytes to numpy array
|
37 |
+
image_array = np.frombuffer(image_bytes, dtype=np.uint8)
|
38 |
+
# Decode the numpy array as an image using OpenCV
|
39 |
+
image = cv2.imdecode(image_array, cv2.IMREAD_COLOR)
|
40 |
+
return image
|
41 |
+
|
42 |
+
@socket.on("connect")
|
43 |
+
def test_connect():
|
44 |
+
print("Connected")
|
45 |
+
emit("my response", {"data": "Connected"})
|
46 |
+
|
47 |
+
@socket.on("image")
|
48 |
+
def receive_image(image):
|
49 |
+
# Decode the base64-encoded image data
|
50 |
+
image = base64_to_image(image)
|
51 |
+
image = cv2.resize(image, (224, 224), interpolation=cv2.INTER_AREA)
|
52 |
+
# emit("processed_image", image)
|
53 |
+
# Make the image a numpy array and reshape it to the models input shape.
|
54 |
+
image = np.asarray(image, dtype=np.float32).reshape(1, 224, 224, 3)
|
55 |
+
image = (image / 127.5) - 1
|
56 |
+
# Predicts the model
|
57 |
+
# prediction1 = age_model.predict(image)
|
58 |
+
prediction2 = gender_model.predict(image)
|
59 |
+
# prediction3 = emotion_model.predict(image)
|
60 |
+
|
61 |
+
index = np.argmax(prediction2)
|
62 |
+
gender_ranges = gender_ranges[index]
|
63 |
+
age = prediction1[0][index]
|
64 |
+
emit("result",{"gender":str(gender_ranges)})
|
65 |
+
|
66 |
+
@app.route("/")
|
67 |
+
def home():
|
68 |
+
return render_template("index.html")
|
69 |
+
|
70 |
+
if __name__ == '__main__':
|
71 |
+
# app.run(debug=True)
|
72 |
+
socket.run(app,host="0.0.0.0", port=7860)
|