Copyright 2019 The TensorFlow Authors.
#@title Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# https://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.#Improving Computer Vision Accuracy using Convolutions
In the previous lessons you saw how to do fashion recognition using a Deep Neural Network (DNN) containing three layers -- the input layer (in the shape of the data), the output layer (in the shape of the desired output) and a hidden layer. You experimented with the impact of different sizes of hidden layer, number of training epochs etc on the final accuracy.
For convenience, here's the entire code again. Run it and take a note of the test accuracy that is printed out at the end.
import tensorflow as tf
mnist = tf.keras.datasets.fashion_mnist
(training_images, training_labels), (test_images, test_labels) = mnist.load_data()
training_images=training_images / 255.0
test_images=test_images / 255.0
model = tf.keras.models.Sequential([
tf.keras.layers.Flatten(),
tf.keras.layers.Dense(128, activation=tf.nn.relu),
tf.keras.layers.Dense(10, activation=tf.nn.softmax)
])
model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])
model.fit(training_images, training_labels, epochs=5)
test_loss = model.evaluate(test_images, test_labels)Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/train-labels-idx1-ubyte.gz
32768/29515 [=================================] - 0s 0us/step
Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/train-images-idx3-ubyte.gz
26427392/26421880 [==============================] - 0s 0us/step
Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/t10k-labels-idx1-ubyte.gz
8192/5148 [===============================================] - 0s 0us/step
Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/t10k-images-idx3-ubyte.gz
4423680/4422102 [==============================] - 0s 0us/step
Epoch 1/5
1875/1875 [==============================] - 3s 2ms/step - loss: 0.4967 - accuracy: 0.8252
Epoch 2/5
1875/1875 [==============================] - 3s 2ms/step - loss: 0.3703 - accuracy: 0.8666
Epoch 3/5
1875/1875 [==============================] - 3s 2ms/step - loss: 0.3365 - accuracy: 0.8768
Epoch 4/5
1875/1875 [==============================] - 3s 2ms/step - loss: 0.3123 - accuracy: 0.8857
Epoch 5/5
1875/1875 [==============================] - 3s 2ms/step - loss: 0.2926 - accuracy: 0.8921
313/313 [==============================] - 1s 2ms/step - loss: 0.3789 - accuracy: 0.8673
Your accuracy is probably about 89% on training and 87% on validation...not bad...But how do you make that even better? One way is to use something called Convolutions. I'm not going to details on Convolutions here, but the ultimate concept is that they narrow down the content of the image to focus on specific, distinct, details.
If you've ever done image processing using a filter (like this: https://en.wikipedia.org/wiki/Kernel_(image_processing)) then convolutions will look very familiar.
In short, you take an array (usually 3x3 or 5x5) and pass it over the image. By changing the underlying pixels based on the formula within that matrix, you can do things like edge detection. So, for example, if you look at the above link, you'll see a 3x3 that is defined for edge detection where the middle cell is 8, and all of its neighbors are -1. In this case, for each pixel, you would multiply its value by 8, then subtract the value of each neighbor. Do this for every pixel, and you'll end up with a new image that has the edges enhanced.
This is perfect for computer vision, because often it's features that can get highlighted like this that distinguish one item for another, and the amount of information needed is then much less...because you'll just train on the highlighted features.
That's the concept of Convolutional Neural Networks. Add some layers to do convolution before you have the dense layers, and then the information going to the dense layers is more focussed, and possibly more accurate.
Run the below code -- this is the same neural network as earlier, but this time with Convolutional layers added first. It will take longer, but look at the impact on the accuracy: