In this document we will demonstrate the basic usage of the
feature_spec interface in
feature_spec interface is a user friendly interface to
feature_columns. It allows us to specify column transformations and representations when working with structured data.
We will use the
hearts dataset and it can be loaded with
library(tfdatasets) library(dplyr) data(hearts)
We want to train a model to predict the
target variable using Keras but, before that we need to prepare the data. We need to transform the categorical variables into some form of dense variable, we usually want to normalize all numeric columns too.
The feature spec interface works with
data.frames or TensorFlow datasets objects.
<- sample.int(nrow(hearts), size = 0.75*nrow(hearts)) ids_train <- hearts[ids_train,] hearts_train <- hearts[-ids_train,]hearts_test
Now let’s start creating our feature specification:
<- feature_spec(hearts_train, target ~ .)spec
The first thing we need to do after creating the feature_spec is decide on the variables’ types.
We can do this by adding steps to the
<- spec %>% spec step_numeric_column( all_numeric(), -cp, -restecg, -exang, -sex, -fbs, normalizer_fn = scaler_standard() %>% ) step_categorical_column_with_vocabulary_list(thal)
The following steps can be used to define the variable type:
step_numeric_columnto define numeric variables
step_categorical_with_vocabulary_listfor categorical variables with a fixed vocabulary
step_categorical_column_with_hash_bucketfor categorical variables using the hash trick
step_categorical_column_with_identityto store categorical variables as integers
step_categorical_column_with_vocabulary_filewhen you have the possible vocabulary in a file
step_categorical_column_with_vocabulary_list you can also provide a
vocabulary argument with the fixed vocabulary. The recipe will find all the unique values in the dataset and use it as the vocabulary.
You can also specify a
normalizer_fn to the
step_numeric_column. In this case the variable will be transformed by the feature column. Note that the transformation will occur in the TensorFlow Graph, so it must use only TensorFlow ops. Like in the example we offer pre-made normalizers - and they will compute the normalizing function during the recipe preparation.
You can also use selectors like:
matches()etc. (from tidyselect)
all_numeric()to select all numeric variables
all_nominal()to select all strings
has_type("float32")to select based on TensorFlow variable type.
Now we can print the recipe:
After specifying the types of the columns you can add transformation steps. For example you may want to bucketize a numeric column:
<- spec %>% spec step_bucketized_column(age, boundaries = c(18, 25, 30, 35, 40, 45, 50, 55, 60, 65))
You can also specify the kind of numeric representation that you want to use for your categorical variables.
<- spec %>% spec step_indicator_column(thal) %>% step_embedding_column(thal, dimension = 2)
Another common transformation is to add interactions between variables using crossed columns.
<- spec %>% spec step_crossed_column(thal_and_age = c(thal, bucketized_age), hash_bucket_size = 1000) %>% step_indicator_column(thal_and_age)
Note that the
crossed_column is a categorical column, so we need to also specify what kind of numeric tranformation we want to use. Also note that we can name the transformed variables - each step uses a default naming for columns, eg.
bucketized_age is the default name when you use
step_bucketized_column with column called
With the above code we have created our recipe. Note we can also define the recipe by chaining a sequence of methods:
<- feature_spec(hearts_train, target ~ .) %>% spec step_numeric_column( all_numeric(), -cp, -restecg, -exang, -sex, -fbs, normalizer_fn = scaler_standard() %>% ) step_categorical_column_with_vocabulary_list(thal) %>% step_bucketized_column(age, boundaries = c(18, 25, 30, 35, 40, 45, 50, 55, 60, 65)) %>% step_indicator_column(thal) %>% step_embedding_column(thal, dimension = 2) %>% step_crossed_column(c(thal, bucketized_age), hash_bucket_size = 10) %>% step_indicator_column(crossed_thal_bucketized_age)
After defining the recipe we need to
fit it. It’s when fitting that we compute the vocabulary list for categorical variables or find the mean and standard deviation for the normalizing functions. Fitting involves evaluating the full dataset, so if you have provided the vocabulary list and your columns are already normalized you can skip the fitting step (TODO).
In our case, we will fit the feature spec, since we didn’t specify the vocabulary list for the categorical variables.
After preparing we can see the list of dense features that were defined:
Now we are ready to define our model in Keras. We will use a specialized
layer_dense_features that knows what to do with the feature columns specification.
We also use a new
layer_input_from_dataset that is useful to create a Keras input object copying the structure from a
data.frame or TensorFlow dataset.
library(keras) <- layer_input_from_dataset(hearts_train %>% select(-target)) input <- input %>% output layer_dense_features(dense_features(spec_prep)) %>% layer_dense(units = 32, activation = "relu") %>% layer_dense(units = 1, activation = "sigmoid") <- keras_model(input, output) model %>% compile( model loss = loss_binary_crossentropy, optimizer = "adam", metrics = "binary_accuracy" )
We can finally train the model on the dataset:
<- model %>% history fit( x = hearts_train %>% select(-target), y = hearts_train$target, epochs = 15, validation_split = 0.2 ) plot(history)
Finally we can make predictions in the test set and calculate performance metrics like the AUC of the ROC curve:
$pred <- predict(model, hearts_test %>% select(-target)) hearts_test::auc(hearts_test$target, hearts_test$pred)Metrics