upload source new

Author: chen0040

keras_image_classifier_web/cifar10_classifier.py

@@ -0,0 +1,43 @@
+from keras.models import model_from_json
+from PIL import Image
+import numpy as np
+import os
+
+class Cifar10Classifier:
+    cifar10_model = None
+
+    def __init__(self):
+        # load and configure the cifar19 classifier model
+        self.cifar10_model = model_from_json(
+            open(os.path.join('../keras_image_classifier/models', 'cnn_cifar10_architecture.json')).read())
+        self.cifar10_model.load_weights(os.path.join('../keras_image_classifier/models', 'cnn_cifar10_weights.h5'))
+        self.cifar10_model.compile(optimizer='rmsprop', loss='categorical_crossentropy', metrics=['accuracy'])
+
+    def predict(self, filename):
+        img = Image.open(filename)
+        img = img.resize((32, 32), Image.ANTIALIAS)
+
+        input = np.asarray(img)
+        input = input.astype('float32') / 255
+        input = np.expand_dims(input, axis=0)
+
+        print(input.shape)
+
+        predicted_class = self.cifar10_model.predict_classes(input)[0]
+
+        labels = [
+            "airplane",
+            "automobile",
+            "bird",
+            "cat",
+            "deer",
+            "dog",
+            "frog",
+            "horse",
+            "ship",
+            "truck"
+        ]
+        return predicted_class, labels[predicted_class]
+
+    def run_test(self):
+        print(self.predict('../keras_image_classifier/bi_classifier_data/training/cat/cat.2.jpg'))

keras_image_classifier_web/cnn_bi_classifier.py

@@ -0,0 +1,34 @@
+from keras.models import model_from_json
+import os
+from PIL import Image
+import numpy as np
+
+
+class BiClassifier(object):
+    bi_model = None
+
+    def __init__(self):
+        # load and configure the binary classifier model for "cats vs dogs"
+        self.bi_model = model_from_json(
+            open(os.path.join('../keras_image_classifier/models', 'cnn_bi_classifier_architecture.json')).read())
+        self.bi_model.load_weights(os.path.join('../keras_image_classifier/models', 'cnn_bi_classifier_weights.h5'))
+        self.bi_model.compile(optimizer='rmsprop', loss='binary_crossentropy', metrics=['accuracy'])
+
+    def run_test(self):
+        print(self.predict('../keras_image_classifier/bi_classifier_data/training/cat/cat.2.jpg'))
+
+    def predict(self, filename):
+        img = Image.open(filename)
+        img = img.resize((150, 150), Image.ANTIALIAS)
+
+        input = np.asarray(img)
+        input = input.astype('float32') / 255
+        input = np.expand_dims(input, axis=0)
+
+        print(input.shape)
+
+        output = self.bi_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

keras_image_classifier_web/flaskr.py

@@ -1,7 +1,7 @@
 import os
-from keras.models import model_from_json
-from PIL import Image
-import numpy as np
+from keras_image_classifier_web.cnn_bi_classifier import BiClassifier
+from keras_image_classifier_web.cifar10_classifier import Cifar10Classifier
+from keras_image_classifier_web.vgg16_classifier import VGG16Classifier
 
 from flask import Flask, request, session, g, redirect, url_for, abort, \
     render_template, flash
@@ -18,68 +18,14 @@
 app.config.from_envvar('FLASKR_SETTINGS', silent=True)
 app.config['MAX_CONTENT_LENGTH'] = 16 * 1024 * 1024
 
-# load and configure the binary classifier model for "cats vs dogs"
-bi_model = model_from_json(
-    open(os.path.join('../keras_image_classifier/models', 'cnn_bi_classifier_architecture.json')).read())
-bi_model.load_weights(os.path.join('../keras_image_classifier/models', 'cnn_bi_classifier_weights.h5'))
+bi_classifier = BiClassifier()
+bi_classifier.run_test()
 
-bi_model.compile(optimizer='rmsprop', loss='binary_crossentropy', metrics=['accuracy'])
-
-# load and configure the cifar19 classifier model
-cifar10_model = model_from_json(
-    open(os.path.join('../keras_image_classifier/models', 'cnn_cifar10_architecture.json')).read())
-cifar10_model.load_weights(os.path.join('../keras_image_classifier/models', 'cnn_cifar10_weights.h5'))
-
-cifar10_model.compile(optimizer='rmsprop', loss='categorical_crossentropy', metrics=['accuracy'])
-
-
-def predict_binary(filename):
-    img = Image.open(filename)
-    img = img.resize((150, 150), Image.ANTIALIAS)
-
-    input = np.asarray(img)
-    input = input.astype('float32') / 255
-    input = np.expand_dims(input, axis=0)
-
-    print(input.shape)
-
-    output = bi_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
-
-
-def predict_cifar10(filename):
-    img = Image.open(filename)
-    img = img.resize((32, 32), Image.ANTIALIAS)
-
-    input = np.asarray(img)
-    input = input.astype('float32') / 255
-    input = np.expand_dims(input, axis=0)
-
-    print(input.shape)
-
-    predicted_class = cifar10_model.predict_classes(input)[0]
-
-    labels = [
-        "airplane",
-        "automobile",
-        "bird",
-        "cat",
-        "deer",
-        "dog",
-        "frog",
-        "horse",
-        "ship",
-        "truck"
-    ]
-    return predicted_class, labels[predicted_class]
-
-
-predict_binary('../keras_image_classifier/bi_classifier_data/training/cat/cat.2.jpg')
-predict_cifar10('../keras_image_classifier/bi_classifier_data/training/cat/cat.2.jpg')
+cifar10_classifier = Cifar10Classifier()
+cifar10_classifier.run_test()
 
+vgg16_classifier = VGG16Classifier()
+vgg16_classifier.run_test()
 
 
 @app.route('/')
@@ -92,6 +38,24 @@ def allowed_file(filename):
            filename.rsplit('.', 1)[1].lower() in ALLOWED_EXTENSIONS
 
 
+def store_uploaded_image(action):
+    # check if the post request has the file part
+    if 'file' not in request.files:
+        flash('No file part')
+        return redirect(request.url)
+    file = request.files['file']
+    # if user does not select file, browser also
+    # submit a empty part without filename
+    if file.filename == '':
+        flash('No selected file')
+        return redirect(request.url)
+    if file and allowed_file(file.filename):
+        filename = secure_filename(file.filename)
+        file.save(os.path.join(app.config['UPLOAD_FOLDER'], filename))
+        return redirect(url_for(action,
+                                filename=filename))
+
+
 @app.route('/about', methods=['GET'])
 def about():
     return 'about us'
@@ -100,61 +64,48 @@ def about():
 @app.route('/cats_vs_dogs', methods=['GET', 'POST'])
 def cats_vs_dogs():
     if request.method == 'POST':
-        # check if the post request has the file part
-        if 'file' not in request.files:
-            flash('No file part')
-            return redirect(request.url)
-        file = request.files['file']
-        # if user does not select file, browser also
-        # submit a empty part without filename
-        if file.filename == '':
-            flash('No selected file')
-            return redirect(request.url)
-        if file and allowed_file(file.filename):
-            filename = secure_filename(file.filename)
-            file.save(os.path.join(app.config['UPLOAD_FOLDER'], filename))
-            return redirect(url_for('cats_vs_dogs_result',
-                                    filename=filename))
+        return store_uploaded_image('cats_vs_dogs_result')
     return render_template('cats_vs_dogs.html')
 
 
 @app.route('/cifar10', methods=['GET', 'POST'])
 def cifar10():
     if request.method == 'POST':
-        # check if the post request has the file part
-        if 'file' not in request.files:
-            flash('No file part')
-            return redirect(request.url)
-        file = request.files['file']
-        # if user does not select file, browser also
-        # submit a empty part without filename
-        if file.filename == '':
-            flash('No selected file')
-            return redirect(request.url)
-        if file and allowed_file(file.filename):
-            filename = secure_filename(file.filename)
-            file.save(os.path.join(app.config['UPLOAD_FOLDER'], filename))
-            return redirect(url_for('cifar10_result',
-                                    filename=filename))
+        return store_uploaded_image('cifar10_result')
     return render_template('cifar10.html')
 
 
+@app.route('/vgg16', methods=['GET', 'POST'])
+def vgg16():
+    if request.method == 'POST':
+        return store_uploaded_image('vgg16_result')
+    return render_template('vgg16.html')
+
+
 @app.route('/cats_vs_dogs_result/<filename>')
 def cats_vs_dogs_result(filename):
     filepath = os.path.join(app.config['UPLOAD_FOLDER'], filename)
-    probability_of_dog, predicted_label = predict_binary(filepath)
+    probability_of_dog, predicted_label = bi_classifier.predict(filepath)
     return render_template('cats_vs_dogs_result.html', filename=filename,
                            probability_of_dog=probability_of_dog, predicted_label=predicted_label)
 
 
 @app.route('/cifar10_result/<filename>')
 def cifar10_result(filename):
     filepath = os.path.join(app.config['UPLOAD_FOLDER'], filename)
-    predicted_class, predicted_label = predict_cifar10(filepath)
+    predicted_class, predicted_label = cifar10_classifier.predict(filepath)
     return render_template('cifar10_result.html', filename=filename,
                            predicted_class=predicted_class, predicted_label=predicted_label)
 
 
+@app.route('/vgg16_result/<filename>')
+def vgg16_result(filename):
+    filepath = os.path.join(app.config['UPLOAD_FOLDER'], filename)
+    top3 = vgg16_classifier.predict(filepath)
+    return render_template('vgg16_result.html', filename=filename,
+                           top3=top3)
+
+
 @app.route('/images/<filename>')
 def get_image(filename):
     return send_from_directory(app.config['UPLOAD_FOLDER'],

keras_image_classifier_web/templates/classifiers.html

@@ -3,5 +3,6 @@
   <ul class=entries>
     <li><a href="{{ url_for('cats_vs_dogs') }}">Cats vs Dogs</a></li>
     <li><a href="{{ url_for('cifar10') }}">CIFAR-10</a></li>
+    <li><a href="{{ url_for('vgg16') }}">VGG-16</a></li>
   </ul>
 {% endblock %}

keras_image_classifier_web/templates/vgg16.html

@@ -0,0 +1,8 @@
+{% extends "layout.html" %}
+{% block body %}
+    <p>Upload your picture and we will try to what it is using VGG16 classifier</p>
+    <form method=post enctype=multipart/form-data>
+      <p><input type=file name=file>
+         <input type=submit value=Upload>
+    </form>
+{% endblock %}

keras_image_classifier_web/templates/vgg16_result.html

@@ -0,0 +1,14 @@
+{% extends "layout.html" %}
+{% block body %}
+    <img src="/images/{{ filename }}" />
+    <br />
+    <hr />
+    Below is my top 3 guesses of what is inside your picture:
+    <ol>
+    {% for pred in top3 %}
+        <li>{{ pred[1] }} ({{ pred[2] }})</li>
+    {% endfor %}
+    </ol>
+    <hr />
+    <a href="{{ url_for('vgg16') }}">Try another picture</a>
+{% endblock %}

keras_image_classifier_web/vgg16_classifier.py

@@ -0,0 +1,26 @@
+from keras.models import Model
+from keras.applications.vgg16 import VGG16, preprocess_input, decode_predictions
+from keras.optimizers import SGD
+from PIL import Image
+from keras.preprocessing.image import img_to_array
+import numpy as np
+
+
+class VGG16Classifier:
+    model = None
+
+    def __init__(self):
+        self.model = VGG16(include_top=True, weights='imagenet')
+        self.model.compile(optimizer=SGD(), loss='categorical_crossentropy', metrics=['accuracy'])
+
+    def predict(self, filename):
+        img = Image.open(filename)
+        img = img.resize((224, 224), Image.ANTIALIAS)
+        input = img_to_array(img)
+        input = np.expand_dims(input, axis=0)
+        input = preprocess_input(input)
+        output = decode_predictions(self.model.predict(input), top=3)
+        return output[0]
+
+    def run_test(self):
+        print(self.predict('../keras_image_classifier/bi_classifier_data/training/cat/cat.3.jpg'))