Commit c6f8ed57 authored by Nane Kratzke's avatar Nane Kratzke
Browse files

Ignore checkpoints

parent 94ef240d
%% Cell type:markdown id:4d709f6a-b9a1-4824-8fb8-cd64e8956c0a tags:
# Basic classification using the Zalando Dataset: Classify images of clothing
This guide trains a neural network model to classify images of clothing, like sneakers and shirts. It's okay if you don't understand all the details; this is a fast-paced overview of a complete TensorFlow program with the details explained as you go.
This guide uses [tf.keras](, a high-level API to build and train models in TensorFlow.
**To follow this tutorial simply step interactively through the Python code cells using the ► button**.
**To follow this tutorial simply step interactively through the notebook cells using the ► button**.
%% Cell type:code id:e66f6149-1a05-447a-9098-526ee537344c tags:
``` python
# Preliminary checks to import Tensorflow
import os, sys
if 'tensorflow' not in os.environ.get('JUPYTER_IMAGE').lower():
print("Error: You need a tensorflow image to run this notebook!", file=sys.stderr)
# TensorFlow and tf.keras
import tensorflow as tf
# Helper libraries
import numpy as np
import matplotlib.pyplot as plt
print(f"Great, your Tensorflow Version is {tf.__version__}")
%% Output
Great, your Tensorflow Version is 2.5.0
%% Cell type:markdown id:240e43d6-5e5b-451f-9f8f-e9383270479c tags:
## Import the Fashion MNIST dataset
This guide uses the Fashion [MNIST dataset]( which contains 70,000 grayscale images in 10 categories. The images show individual articles of clothing at low resolution (28 by 28 pixels), as seen here:
![Fashion Dataset](
Fashion MNIST is intended as a drop-in replacement for the classic [MNIST]( dataset—often used as the "Hello, World" of machine learning programs for computer vision. The MNIST dataset contains images of handwritten digits (0, 1, 2, etc.) in a format identical to that of the articles of clothing you'll use here.
This guide uses Fashion MNIST for variety, and because it's a slightly more challenging problem than regular MNIST. Both datasets are relatively small and are used to verify that an algorithm works as expected. They're good starting points to test and debug code.
Here, 60,000 images are used to train the network and 10,000 images to evaluate how accurately the network learned to classify images. You can access the Fashion MNIST directly from TensorFlow. **Import and load the Fashion MNIST data directly from TensorFlow:**
%% Cell type:code id:0c8dd540-4689-4626-8611-ba85fdfb0af4 tags:
``` python
fashion_mnist = tf.keras.datasets.fashion_mnist
(train_images, train_labels), (test_images, test_labels) = fashion_mnist.load_data()
%% Cell type:markdown id:816889a8-c680-48d6-800d-08d349d3f7b6 tags:
Loading the dataset returns four NumPy arrays:
- The `train_images` and `train_labels` arrays are the training set—the data the model uses to learn.
- The model is tested against the test set, the `test_images`, and `test_labels` arrays.
- The images are 28x28 NumPy arrays, with pixel values ranging from 0 to 255. The labels are an array of integers, ranging from 0 to 9.
These correspond to the class of clothing the image represents:
| Label | Class |
| ----- | ----- |
| 0 | T-shirt/top |
| 1 | Trouser |
| 2 | Pullover |
| 3 | Dress |
| 4 | Coat |
| 5 | Sandal |
| 6 | Shirt |
| 7 | Sneaker |
| 8 | Bag |
| 9 | Ankle boot |
Each image is mapped to a single label. Since the class names are not included with the dataset, store them here to use later when plotting the images:
%% Cell type:code id:00a6ed38-ea17-4d3d-a745-7e1fa1507ea6 tags:
``` python
class_names = ['T-shirt/top', 'Trouser', 'Pullover', 'Dress', 'Coat', 'Sandal', 'Shirt', 'Sneaker', 'Bag', 'Ankle boot']
%% Cell type:markdown id:867fb3cd-5e7f-4200-8a51-2488865af797 tags:
## Explore the data
Let's explore the format of the dataset before training the model. The following shows there are 60,000 images in the training set, with each image represented as 28 x 28 pixels:
%% Cell type:code id:de6edb3a-88f6-4e1a-907e-d794cefea2f8 tags:
``` python
%% Cell type:markdown id:b628c5e0-f81c-41fa-811c-a5d7e827d8f4 tags:
Likewise, there are 60,000 labels in the training set:
%% Cell type:code id:0edbb781-102d-4431-9f17-c450f4666a45 tags:
``` python
%% Cell type:markdown id:e6e60a8e-814d-41f8-a643-e032aacf413a tags:
Each label is an integer between 0 and 9:
%% Cell type:code id:883d6001-7848-4a0f-a29c-a564bebb7a79 tags:
``` python
%% Cell type:markdown id:dd47595b-f4d3-406b-9f7c-de4f01c4d5cc tags:
And the test set contains 10,000 images labels:
%% Cell type:code id:a1eeb5ac-208e-4998-a986-8f23dc9521e1 tags:
``` python
%% Cell type:markdown id:eb9d279b-92d2-4e5c-b624-79ece636828a tags:
## Preprocess the data
The data must be preprocessed before training the network. If you inspect the first image in the training set, you will see that the pixel values fall in the range of 0 to 255:
%% Cell type:code id:76f43538-99de-4677-81a5-730c0cf6b52c tags:
``` python
%% Cell type:markdown id:31c47da5-f74a-4316-a567-dbd95573f382 tags:
Scale these values to a range of 0 to 1 before feeding them to the neural network model. To do so, divide the values by 255. It's important that the training set and the testing set be preprocessed in the same way:
%% Cell type:code id:0aa57ac1-5651-4212-bc1a-c1c489ccdfd9 tags:
``` python
train_images = train_images / 255.0
test_images = test_images / 255.0
%% Cell type:markdown id:f081f6ef-908f-4877-89f6-7513fc0ec2a9 tags:
To verify that the data is in the correct format and that you're ready to build and train the network, let's display the first 25 images from the training set and display the class name below each image.
%% Cell type:code id:3b9370ea-bbd6-4534-8157-4b865a39c3b0 tags:
``` python
for i in range(25):
%% Cell type:markdown id:3be87ee6-a76c-478f-b1b6-b112fe6516d0 tags:
## Build the model
Building the neural network requires configuring the layers of the model, then compiling the model.
### Set up the layers
The basic building block of a neural network is the [layer]( Layers extract representations from the data fed into them. Hopefully, these representations are meaningful for the problem at hand.
Most of deep learning consists of chaining together simple layers. Most layers, such as [tf.keras.layers.Dense](, have parameters that are learned during training.
%% Cell type:code id:209b0486-3c73-429d-a827-027c243e4157 tags:
``` python
model = tf.keras.Sequential([
tf.keras.layers.Flatten(input_shape=(28, 28)),
tf.keras.layers.Dense(128, activation='relu'),
%% Cell type:markdown id:0cecd604-5b53-4702-87d1-2b26dedae19c tags:
The first layer in this network, [tf.keras.layers.Flatten](, transforms the format of the images from a two-dimensional array (of 28 by 28 pixels) to a one-dimensional array (of 28 * 28 = 784 pixels). Think of this layer as unstacking rows of pixels in the image and lining them up. This layer has no parameters to learn; it only reformats the data.
After the pixels are flattened, the network consists of a sequence of [two tf.keras.layers.Dense]( layers. These are densely connected, or fully connected, neural layers. The first Dense layer has 128 nodes (or neurons). The second (and last) layer returns a logits array with length of 10. Each node contains a score that indicates the current image belongs to one of the 10 classes.
### Compile the model
Before the model is ready for training, it needs a few more settings. These are added during the model's compile step:
- [Loss function]( — This measures how accurate the model is during training. You want to minimize this function to "steer" the model in the right direction.
- [Optimizer]( — This is how the model is updated based on the data it sees and its loss function.
- [Metrics]( — Used to monitor the training and testing steps. The following example uses accuracy, the fraction of the images that are correctly classified.
%% Cell type:code id:3736e82e-1de8-477b-8c44-4c4bf9100a7b tags:
``` python
%% Cell type:markdown id:fca063a0-325c-4857-a988-ba48fc7128c7 tags:
## Train the model
Training the neural network model requires the following steps:
1. Feed the training data to the model. In this example, the training data is in the `train_images` and `train_labels` arrays.
2. The model learns to associate images and labels.
3. You ask the model to make predictions about a test set—in this example, the `test_images` array.
4. Verify that the predictions match the labels from the `test_labels` array.
### Feed the model
To start training, call the []( method—so called because it "fits" the model to the training data:
%% Cell type:code id:d3c6f740-c1ea-43de-912a-bed799cd58a0 tags:
``` python, train_labels, epochs=10)
%% Cell type:markdown id:ccead956-9389-4efd-900e-0f736597c8f1 tags:
As the model trains, the loss and accuracy metrics are displayed. This model reaches an accuracy of about 0.91 (or 91%) on the training data.
### Evaluate accuracy
Next, compare how the model performs on the test dataset:
%% Cell type:code id:889c817e-3690-4fba-b155-1575146b0ab9 tags:
``` python
test_loss, test_acc = model.evaluate(test_images, test_labels, verbose=2)
print('\nTest accuracy:', test_acc)
%% Cell type:markdown id:b8fc59c9-c04f-4a32-be26-327fafa7eec5 tags:
It turns out that the accuracy on the test dataset is a little less than the accuracy on the training dataset. This gap between training accuracy and test accuracy represents overfitting. Overfitting happens when a machine learning model performs worse on new, previously unseen inputs than it does on the training data. An overfitted model "memorizes" the noise and details in the training dataset to a point where it negatively impacts the performance of the model on the new data. For more information, see the following:
- [Demonstrate overfitting](
- [Strategies to prevent overfitting](
### Make predictions
With the model trained, you can use it to make predictions about some images. The model's linear outputs, [logits]( Attach a softmax layer to convert the logits to probabilities, which are easier to interpret.
%% Cell type:code id:6ea82cf4-a25e-4a0c-a634-471bc2816982 tags:
``` python
probability_model = tf.keras.Sequential([model, tf.keras.layers.Softmax()])
predictions = probability_model.predict(test_images)
%% Cell type:markdown id:cddb5534-9075-4aff-9f5f-1257e69d6a6d tags:
Here, the model has predicted the label for each image in the testing set. Let's take a look at the first prediction:
%% Cell type:code id:133f56fb-d423-4177-8a68-d2c0c31ed798 tags:
``` python
%% Cell type:markdown id:f269ae56-58ff-4238-8b3b-e314b61bc163 tags:
A prediction is an array of 10 numbers. They represent the model's "confidence" that the image corresponds to each of the 10 different articles of clothing. You can see which label has the highest confidence value:
%% Cell type:code id:36082941-693e-4f61-9112-c51019dcbee8 tags:
``` python
%% Cell type:markdown id:d8bd872e-b82c-474e-b397-3d78e6621acc tags:
So, the model is most confident that this image is an ankle boot, or `class_names[9]`. Examining the test label shows that this classification is correct:
%% Cell type:code id:07e27687-b06b-42fe-ab0b-d9430f6d79af tags:
``` python
%% Cell type:markdown id:e2ba5393-69c1-4970-93a9-9cebd7052a97 tags:
Graph this to look at the full set of 10 class predictions.
%% Cell type:code id:00b20016-d87b-4e63-884b-dcc8c2d3176f tags:
``` python
def plot_image(i, predictions_array, true_label, img):
true_label, img = true_label[i], img[i]
predicted_label = np.argmax(predictions_array)
if predicted_label == true_label:
color = 'blue'
color = 'red'
plt.xlabel("{} {:2.0f}% ({})".format(class_names[predicted_label],
def plot_value_array(i, predictions_array, true_label):
true_label = true_label[i]
thisplot =, predictions_array, color="#777777")
plt.ylim([0, 1])
predicted_label = np.argmax(predictions_array)
%% Cell type:markdown id:3b4a3504-1298-4ff0-9a91-763dc4997bb0 tags:
### Verify predictions
With the model trained, you can use it to make predictions about some images.
Let's look at the 0th image, predictions, and prediction array. Correct prediction labels are blue and incorrect prediction labels are red. The number gives the percentage (out of 100) for the predicted label.
%% Cell type:code id:8ad7c1ab-7b7b-4674-9909-433721060870 tags:
``` python
i = 0
plot_image(i, predictions[i], test_labels, test_images)
plot_value_array(i, predictions[i], test_labels)
%% Cell type:code id:e94e37f6-15b8-4426-9221-78784ca119d8 tags:
``` python
i = 12
plot_image(i, predictions[i], test_labels, test_images)
plot_value_array(i, predictions[i], test_labels)
%% Cell type:markdown id:95d4a145-1f2e-4ec7-8b1f-b5db1124211e tags:
Let's plot several images with their predictions. Note that the model can be wrong even when very confident.
%% Cell type:code id:143c82ab-2d72-43ec-91ba-a6cd96e6700d tags:
``` python
# Plot the first X test images, their predicted labels, and the true labels.
# Color correct predictions in blue and incorrect predictions in red.
num_rows = 5
num_cols = 3
num_images = num_rows*num_cols
plt.figure(figsize=(2*2*num_cols, 2*num_rows))
for i in range(num_images):
plt.subplot(num_rows, 2*num_cols, 2*i+1)
plot_image(i, predictions[i], test_labels, test_images)
plt.subplot(num_rows, 2*num_cols, 2*i+2)
plot_value_array(i, predictions[i], test_labels)
%% Cell type:markdown id:1dda2d9e-e389-417b-98d8-5ebf3cfe5ae0 tags:
## Use the trained model
Finally, use the trained model to make a prediction about a single image.
%% Cell type:code id:dd526bca-2702-4db6-bff9-6a516cd26253 tags:
``` python
# Grab an image from the test dataset.
img = test_images[7]
%% Cell type:markdown id:f71ad288-5142-4585-809e-0306023ea40a tags:
[tf.keras]( models are optimized to make predictions on a batch, or collection, of examples at once. Accordingly, even though you're using a single image, you need to add it to a list:
%% Cell type:code id:4000f224-98a2-4187-bf59-1bda7bc79ec0 tags:
``` python
# Add the image to a batch where it's the only member.
img = (np.expand_dims(img,0))
%% Cell type:markdown id:9e44a542-866c-420d-b7a0-538359158e9f tags:
Now predict the correct label for this image:
%% Cell type:code id:548678d8-8e2b-4fd5-8038-0f1d2fc7337c tags:
``` python
predictions_single = probability_model.predict(img)
%% Cell type:code id:5c597d1b-c328-4b0c-8988-3f727fa3caa7 tags:
``` python
plot_value_array(1, predictions_single[0], test_labels)
_ = plt.xticks(range(10), class_names, rotation=45)
%% Cell type:markdown id:885e997f-f073-413d-8e8f-756a646a276e tags:
[tf.keras.Model.predict]( returns a list of lists—one list for each image in the batch of data. Grab the predictions for our (only) image in the batch:
%% Cell type:code id:cd887d2d-cd3f-4caf-a8c2-800e6dada870 tags:
``` python
%% Cell type:markdown id:f9a55cdf-7518-4332-bfb1-5fc9c785f4ad tags:
And the model predicts a label as expected.
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment