add samples for experiment

gradio/inputs.py

@@ -134,6 +134,7 @@ class Sketchpad(AbstractInput):
         if self.sample_inputs is not None:
             for input in self.sample_inputs:
                 encoded_images.append(preprocessing_utils.encode_array_to_base64(input))
+        print(encoded_images)
         return encoded_images
 
 

gradio/preprocessing_utils.py

@@ -24,7 +24,8 @@ def encode_array_to_base64(image_array):
         bytes_data = output_bytes.getvalue()
 
     base64_str = str(base64.b64encode(bytes_data), 'utf-8')
-    return base64_str
+
+    return "data:image/png;base64," + base64_str
 
 
 def resize_and_crop(img, size, crop_type='top'):

gradio/test.py

@@ -0,0 +1,31 @@
+import tensorflow as tf
+import gradio
+
+
+#CREATE INTERFACE USING 'GOOD' MNIST MODEL
+(x_train, y_train),(x_test, y_test) = tf.keras.datasets.mnist.load_data()
+model = tf.keras.models.load_model('MNIST_9344.h5')
+input = gradio.inputs.Sketchpad(sample_inputs=x_train[:10])
+iface = gradio.Interface(inputs=input, outputs="label", model=model, model_type='keras')
+iface.launch(inline=False, share=False, inbrowser=True);
+
+
+#CREATE INTERFACE BY TRAINING MSNIST MODEL
+
+# (x_train, y_train),(x_test, y_test) = tf.keras.datasets.mnist.load_data()
+# x_train, x_test = x_train / 255.0, x_test / 255.0
+#
+# model = tf.keras.models.Sequential([
+#   tf.keras.layers.Flatten(),
+#   tf.keras.layers.Dense(512, activation=tf.nn.relu),
+#   tf.keras.layers.Dropout(0.2),
+#   tf.keras.layers.Dense(10, activation=tf.nn.softmax)
+# ])
+#
+# model.compile(optimizer='adam',
+#               loss='sparse_categorical_crossentropy',
+#               metrics=['accuracy'])
+# model.fit(x_train, y_train, epochs=1)
+# inp = gradio.inputs.Sketchpad(sample_inputs=x_train[:10])
+# iface = gradio.Interface(inputs=inp, outputs="label", model=model, model_type='keras')
+# iface.launch(inline=False, share=False, inbrowser=True);