model backend

imageDiseaseAPI.py

+import numpy as np
+from flask import Flask, jsonify, request, render_template
+from pyexpat import model
+from xmlrpc.client import boolean
+from flask import Flask
+import pickle
+from flask_cors import CORS, cross_origin
+from flask_restful import Api, Resource
+import keras as keras
+import cv2
+
+import main_image
+
+# model = hdf5.load(open('Poultry_Diseases_11.h5', 'rb'))
+
+input_dim_x = 200
+input_dim_y = 200
+weight_file_name = 'Poultry_Diseases_11.h5'
+
+from keras import models
+from keras.models import load_model
+
+model_DeepLab = load_model(weight_file_name)
+
+#app = Flask(__name__)
+
+path = 'D:/New folder (9)/'
+
+
+
+image_name_1 = 'cocci.0.jpg'
+image_name_2 = 'cocci.1.jpg'
+image_name_3 = 'cocci.2.jpg'
+
+image_name_4 = 'salmo.4.jpg'
+image_name_5 = 'salmo.5.jpg'
+image_names = [image_name_1, image_name_2, image_name_3, image_name_4, image_name_5]
+
+
+def normalize(arr):
+    diff = np.amax(arr) - np.amin(arr)
+    diff = 255 if diff == 0 else diff
+    arr = arr / np.absolute(diff)
+    return arr
+
+
+# Data Generator
+class DataGenerator:
+    # Constructor
+    def __init__(self, path, split_ratio, x, y):
+        self.x = x
+        self.y = y
+        self.image_path = path
+
+    # Custom Image Data Generator
+    def generate_data(self, image_name):
+        image_batch = []
+
+        image = cv2.imread(self.image_path + image_name, 1)
+
+        # resize image
+        dimensions = (self.x, self.y)
+        resized_image = cv2.resize(image, dimensions, interpolation=cv2.INTER_AREA)
+
+        image_batch.append(resized_image.astype("float32"))
+
+        image_batch = normalize(np.array(image_batch))
+        return image_batch
+
+#image1 = main_image.prediction().image1
+
+#def predict(image1):
+  #  image_data = DataGenerator(path, split_ratio=0.0, x=input_dim_x, y=input_dim_y)
+  #  import cv2
+
+  #  demo_data = image_data.generate_data(image1)
+   # print (demo_data)
+   # DeepLab = model_DeepLab.predict(demo_data, verbose = 1)
+
+ #   print(DeepLab)
+ #   return DeepLab
+
+
+def predict():
+    image_data = DataGenerator(path, split_ratio=0.0, x=input_dim_x, y=input_dim_y)
+    import cv2
+
+    demo_data = image_data.generate_data(image_name_5)
+
+    DeepLab = model_DeepLab.predict(demo_data, verbose = 1)
+
+    print(DeepLab)
+    return DeepLab
+
+#predict()