tf.raw_ops.InitializeTableFromTextFile

Initializes a table from a text file.

tf.raw_ops.InitializeTableFromTextFile(
    table_handle, filename, key_index, value_index, vocab_size=-1,
    delimiter='\t', offset=0, name=None
)

It inserts one key-value pair into the table for each line of the file. The key and value is extracted from the whole line content, elements from the split line based on delimiter or the line number (starting from zero). Where to extract the key and value from a line is specified by key_index and value_index.

  • A value of -1 means use the line number(starting from zero), expects int64.
  • A value of -2 means use the whole line content, expects string.
  • A value >= 0 means use the index (starting at zero) of the split line based on delimiter.

table_handle A Tensor of type mutable string. Handle to a table which will be initialized.
filename A Tensor of type string. Filename of a vocabulary text file.
key_index An int that is >= -2. Column index in a line to get the table key values from.
value_index An int that is >= -2. Column index that represents information of a line to get the table value values from.
vocab_size An optional int that is >= -1. Defaults to -1. Number of elements of the file, use -1 if unknown.
delimiter An optional string. Defaults to "\t". Delimiter to separate fields in a line.
offset An optional int. Defaults to 0.
name A name for the operation (optional).

The created Operation.