90-keras-basic-classification.Rmd

---
title: "Keras Example"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)

```
```{bash}
pip3 install virtualenv
```

```{r}
library(keras)
install_keras()
```


```{r}
fashion_mnist <- dataset_fashion_mnist()

c(train_images, train_labels) %<-% fashion_mnist$train
c(test_images, test_labels) %<-% fashion_mnist$test
```

```{r}
class_names = c('T-shirt/top',
                'Trouser',
                'Pullover',
                'Dress',
                'Coat', 
                'Sandal',
                'Shirt',
                'Sneaker',
                'Bag',
                'Ankle boot')
```
```{r}
dim(train_images)
```
```{r}
dim(train_labels)
```
```{r}
train_labels[1:20]
```
```{r}
dim(test_images)
```
```{r}
dim(test_labels)
```

## Preprocess the data
```{r}
library(tidyr)
library(ggplot2)

image_1 <- as.data.frame(train_images[1, , ])
colnames(image_1) <- seq_len(ncol(image_1))
image_1$y <- seq_len(nrow(image_1))
image_1 <- gather(image_1, "x", "value", -y)
image_1$x <- as.integer(image_1$x)

ggplot(image_1, aes(x = x, y = y, fill = value)) +
  geom_tile() +
  scale_fill_gradient(low = "white", high = "black", na.value = NA) +
  scale_y_reverse() +
  theme_minimal() +
  theme(panel.grid = element_blank())   +
  theme(aspect.ratio = 1) +
  xlab("") +
  ylab("")
```
```{r}
train_images <- train_images / 255
test_images <- test_images / 255
```
```{r}
par(mfcol=c(5,5))
par(mar=c(0, 0, 1.5, 0), xaxs='i', yaxs='i')
for (i in 1:25) { 
  img <- train_images[i, , ]
  img <- t(apply(img, 2, rev)) 
  image(1:28, 1:28, img, col = gray((0:255)/255), xaxt = 'n', yaxt = 'n',
        main = paste(class_names[train_labels[i] + 1]))
}
```

## Train the model

### Setup the layers
```{r}
model <- keras_model_sequential()
model %>%
  layer_flatten(input_shape = c(28, 28)) %>%
  layer_dense(units = 128, activation = 'relu') %>%
  layer_dense(units = 10, activation = 'softmax')
```
### Compile the model
```{r}
model %>% compile(
  optimizer = 'adam', 
  loss = 'sparse_categorical_crossentropy',
  metrics = c('accuracy')
)
```
### Train the model
```{r}
model %>% fit(train_images, train_labels, epochs = 5)
```

```{r}
score <- model %>% evaluate(test_images, test_labels)

cat('Test loss:', score$loss, "\n")
cat('Test accuracy:', score$acc, "\n")
```
### Make predictions
```{r}
predictions <- model %>% predict(test_images)
```

Here, the model has predicted the label for each image in the testing set. Let’s take a look at the first prediction:

```{r}
predictions[1, ]
```
[1] 1.247006e-06 8.371295e-08 1.987903e-07 1.078801e-06 2.053094e-07 1.419330e-02 9.533844e-06
[8] 5.745503e-02 2.694531e-05 9.283124e-01

A prediction is an array of 10 numbers. These describe the “confidence” of the model that the image corresponds to each of the 10 different articles of clothing. We can see which label has the highest confidence value:

```{r}
which.max(predictions[1, ])
```
[1] 10

Alternatively, we can also directly get the class prediction:

```{r}
class_pred <- model %>% predict_classes(test_images)
class_pred[1:20]
```
[1] 9 2 1 1 6 1 4 6 5 7 4 5 5 3 4 1 2 2 8 0

As the labels are 0-based, this actually means a predicted label of 9 (to be found in class_names[9]). So the model is most confident that this image is an ankle boot. And we can check the test label to see this is correct:

```{r}
test_labels[1]
```
[1] 9

Let’s plot several images with their predictions. Correct prediction labels are green and incorrect prediction labels are red.

```{r}
par(mfcol=c(5,5))
par(mar=c(0, 0, 1.5, 0), xaxs='i', yaxs='i')
for (i in 1:25) { 
  img <- test_images[i, , ]
  img <- t(apply(img, 2, rev)) 
  # subtract 1 as labels go from 0 to 9
  predicted_label <- which.max(predictions[i, ]) - 1
  true_label <- test_labels[i]
  if (predicted_label == true_label) {
    color <- '#008800' 
  } else {
    color <- '#bb0000'
  }
  image(1:28, 1:28, img, col = gray((0:255)/255), xaxt = 'n', yaxt = 'n',
        main = paste0(class_names[predicted_label + 1], " (",
                      class_names[true_label + 1], ")"),
        col.main = color)
}
```

Finally, use the trained model to make a prediction about a single image.

```{r}
# Grab an image from the test dataset
# take care to keep the batch dimension, as this is expected by the model
img <- test_images[1, , , drop = FALSE]
dim(img)
```
[1]  1 28 28

Now predict the image:

```{r}
predictions <- model %>% predict(img)
predictions
```
[,1]         [,2]         [,3]       [,4]         [,5]       [,6]         [,7]
[1,] 1.247006e-06 8.371312e-08 1.987903e-07 1.0788e-06 2.053096e-07 0.01419331 9.533853e-06
           [,8]         [,9]     [,10]
[1,] 0.05745504 2.694531e-05 0.9283124

predict returns a list of lists, one for each image in the batch of data. Grab the predictions for our (only) image in the batch:

```{r}
# subtract 1 as labels are 0-based
prediction <- predictions[1, ] - 1
which.max(prediction)
```
[1] 10

Or, directly getting the class prediction again:

```{r}
class_pred <- model %>% predict_classes(img)
class_pred
```
[1] 9

And, as before, the model predicts a label of 9.