[MRG] Update image resize functinality (#326)

* optimize image resizing with is_resize_needed function

* update-1

* fix

autokeras/image/image_supervised.py

@@ -132,15 +132,21 @@ def loss(self):
     def fit(self, x, y, x_test=None, y_test=None, time_limit=None):
         x = np.array(x)
 
-        if len(x.shape) != 0 and len(x[0].shape) == 3:
-            if self.verbose:
-                print("Preprocessing the images.")
+        if self.verbose:
+            print("Preprocessing the images.")
+
+        if x is not None and (len(x.shape) == 4 or len(x.shape) == 1 and len(x[0].shape) == 3):
             self.resize_height, self.resize_width = compute_image_resize_params(x)
+
+        if self.resize_height is not None:
             x = resize_image_data(x, self.resize_height, self.resize_width)
-            if x_test is not None:
-                x_test = resize_image_data(x_test, self.resize_height, self.resize_width)
-            if self.verbose:
-                print("Preprocessing finished.")
+            print("x is ", x.shape)
+
+        if self.resize_height is not None:
+            x_test = resize_image_data(x_test, self.resize_height, self.resize_width)
+
+        if self.verbose:
+            print("Preprocessing finished.")
 
         y = np.array(y).flatten()
         validate_xy(x, y)
@@ -206,7 +212,8 @@ def inverse_transform_y(self, output):
 
     def evaluate(self, x_test, y_test):
         """Return the accuracy score between predict value and `y_test`."""
-        if len(x_test.shape) != 0 and len(x_test[0].shape) == 3:
+        print("x test is ", x_test.shape)
+        if self.resize_height is not None:
             x_test = resize_image_data(x_test, self.resize_height, self.resize_width)
         y_predict = self.predict(x_test)
         return self.metric().evaluate(y_test, y_predict)
@@ -225,10 +232,11 @@ def final_fit(self, x_train, y_train, x_test, y_test, trainer_args=None, retrain
         if trainer_args is None:
             trainer_args = {'max_no_improvement_num': 30}
 
-        if len(x_train.shape) != 0 and len(x_train[0].shape) == 3:
+        if self.resize_height is not None:
             x_train = resize_image_data(x_train, self.resize_height, self.resize_width)
-            if x_test is not None:
-                x_test = resize_image_data(x_test, self.resize_height, self.resize_width)
+
+        if self.resize_height is not None:
+            x_test = resize_image_data(x_test, self.resize_height, self.resize_width)
 
         y_train = self.transform_y(y_train)
         y_test = self.transform_y(y_test)
@@ -392,7 +400,7 @@ def inverse_transform_y(self, output):
 
     def evaluate(self, x_test, y_test):
         """Return the accuracy score between predict value and `y_test`."""
-        if len(x_test.shape) != 0 and len(x_test.shape) == 3:
+        if self.resize_height is not None:
             x_test = resize_image_data(x_test, self.resize_height, self.resize_width)
         y_predict = self.predict(x_test)
         return self.metric().evaluate(y_test, y_predict)

autokeras/utils.py

@@ -201,6 +201,9 @@ def compute_image_resize_params(data):
         median height: Median height of all images in the data.
         median width: Median width of all images in the data.
     """
+    if len(data.shape) == 1 and len(data[0].shape) != 3:
+        return None, None
+
     median_height, median_width = numpy.median(numpy.array(list(map(lambda x: x.shape, data))), axis=0)[:2]
 
     if median_height * median_width > Constant.MAX_IMAGE_SIZE:
@@ -211,7 +214,7 @@ def compute_image_resize_params(data):
     return int(median_height), int(median_width)
 
 
-def resize_image_data(data, height, weight):
+def resize_image_data(data, height, width):
     """Resize images to provided height and width.
 
     Resize all images in data to size h x w x c, where h is the height, w is the width and c is the number of channels.
@@ -225,13 +228,18 @@ def resize_image_data(data, height, weight):
     Returns:
         data: Resize data.
     """
+    if data is None:
+        return data
+
+    if len(data.shape) == 4 and data[0].shape[0] == height and data[0].shape[1] == width:
+        return data
 
     output_data = []
     for im in data:
         if len(im.shape) != 3:
             return data
         output_data.append(resize(image=im,
-                                  output_shape=(height, weight, im.shape[-1]),
+                                  output_shape=(height, width, im.shape[-1]),
                                   mode='edge',
                                   preserve_range=True))