Returns a dense tensor as input layer based on given feature_columns
.
At the first layer of the model, this column oriented data should be converted
to a single tensor.
input_layer(features, feature_columns, weight_collections = NULL,
trainable = TRUE)
A mapping from key to tensors. Feature columns look up via
these keys. For example column_numeric('price')
will look at 'price' key
in this dict. Values can be a sparse tensor or tensor depends on
corresponding feature column.
An iterable containing the FeatureColumns to use as
inputs to your model. All items should be instances of classes derived from
a dense column such as column_numeric()
, column_embedding()
,
column_bucketized()
, column_indicator()
. If you have categorical features,
you can wrap them with an column_embedding()
or column_indicator()
.
A list of collection names to which the Variable
will be added. Note that, variables will also be added to collections
graph_keys()$GLOBAL_VARIABLES
and graph_keys()$MODEL_VARIABLES
.
If TRUE
also add the variable to the graph collection
graph_keys()$TRAINABLE_VARIABLES
(see tf$Variable
).
A tensor which represents input layer of a model. Its shape is
(batch_size, first_layer_dimension) and its dtype is float32
.
first_layer_dimension is determined based on given feature_columns
.
ValueError: if an item in feature_columns
is not a dense column.
Other feature column constructors: column_bucketized
,
column_categorical_weighted
,
column_categorical_with_hash_bucket
,
column_categorical_with_identity
,
column_categorical_with_vocabulary_file
,
column_categorical_with_vocabulary_list
,
column_crossed
,
column_embedding
,
column_numeric