binary classifier for cat and dog working

Author: chen0040

keras_image_classifier_web/flaskr.py

@@ -1,14 +1,14 @@
 import os
 from keras.models import model_from_json
-import sqlite3
-import uuid
+from PIL import Image
+import numpy as np
 
 from flask import Flask, request, session, g, redirect, url_for, abort, \
     render_template, flash
 from flask import send_from_directory
 from werkzeug.utils import secure_filename
 
-UPLOAD_FOLDER = '../uploads'
+UPLOAD_FOLDER = 'uploads'
 ALLOWED_EXTENSIONS = set(['txt', 'pdf', 'png', 'jpg', 'jpeg', 'gif'])
 
 app = Flask(__name__)  # create the application instance :)
@@ -18,12 +18,32 @@
 app.config.from_envvar('FLASKR_SETTINGS', silent=True)
 app.config['MAX_CONTENT_LENGTH'] = 16 * 1024 * 1024
 
-model = model_from_json(open(os.path.join('models', '/cnn_bi_classifier_architecture.json')).read())
-model.load_weights(os.path.join('models', '/cnn_bi_classifier_weights.h5'))
+model = model_from_json(open(os.path.join('../keras_image_classifier/models', 'cnn_bi_classifier_architecture.json')).read())
+model.load_weights(os.path.join('../keras_image_classifier/models', 'cnn_bi_classifier_weights.h5'))
 
 model.compile(optimizer='rmsprop', loss='binary_crossentropy', metrics=['accuracy'])
 
 
+def predict(filename):
+    img = Image.open(filename)
+    img = img.resize((150, 150), Image.ANTIALIAS)
+
+    input = np.asarray(img)
+    input = np.expand_dims(input, axis=0)
+
+    print(input.shape)
+
+    output = model.predict(input)
+
+    probability_of_a_dog = output[0][0]
+    predicted_label = ("Dog" if probability_of_a_dog > 0.5 else "Cat")
+    return probability_of_a_dog, predicted_label
+
+prob, label = predict('../keras_image_classifier/bi_classifier_data/training/cat/cat.2.jpg')
+print(prob)
+print(label)
+
+
 @app.route('/')
 def classifiers():
     return render_template('classifiers.html')
@@ -63,7 +83,9 @@ def cats_vs_dogs():
 @app.route('/cats_vs_dogs_result/<filename>')
 def cats_vs_dogs_result(filename):
     filepath = os.path.join(app.config['UPLOAD_FOLDER'], filename)
-    return render_template('cats_vs_dogs_result.html', filename=filename)
+    probability_of_dog, predicted_label = predict(filepath)
+    return render_template('cats_vs_dogs_result.html', filename=filename,
+                           probability_of_dog=probability_of_dog, predicted_label=predicted_label)
 
 
 @app.route('/images/<filename>')

keras_image_classifier_web/models/cnn_bi_classifier_architecture.json

@@ -1,5 +1,6 @@
 {% extends "layout.html" %}
 {% block body %}
+    <p>Upload your Cat or Dog picture and we will try to guess it is a Cat or a Dog</p>
     <form method=post enctype=multipart/form-data>
       <p><input type=file name=file>
          <input type=submit value=Upload>

keras_image_classifier_web/models/cnn_bi_classifier_weights.h5

@@ -1,7 +1,10 @@
 {% extends "layout.html" %}
 {% block body %}
-    <form method=post enctype=multipart/form-data>
-      <p><input type=file name=file>
-         <input type=submit value=Upload>
-    </form>
+    <img src="/images/{{ filename }}" />
+    <br />
+    <hr />
+    I guess you have uploaded a {{ predicted_label }} picture, as the probability of this being a dog picture is
+    {{ probability_of_dog }}
+    <hr />
+    <a href="{{ url_for('cats_vs_dogs') }}">Try another picture</a>
 {% endblock %}