Save the date! Google I/O returns May 18-20 Register now

tf.contrib.framework.load_linear_multiclass_bias_initializer

View source on GitHub

Loads pre-trained multi-class biases for linear models from checkpoint.

Wrapper around load_and_remap_matrix_initializer() specialized for loading multi-class bias and remapping according to the provided vocab files. See docs for load_and_remap_matrix_initializer() for more details. In this case, the provided row_vocab is the class vocabulary, and the expected shape is [new_class_vocab_size, 1].

ckpt_path Path to the TensorFlow checkpoint (version 2, TensorBundle) from which the old matrix Tensor will be loaded.
bias_tensor_name Tensor name to load from in the checkpoints.
new_class_vocab_size Number of entries in the new class vocab.
old_class_vocab_file A scalar Tensor of type string containing the path to the old class vocabulary file.
new_class_vocab_file A scalar Tensor of type string containing the path to the new class vocabulary file.
num_class_oov_buckets int specifying the number of out-of-vocabulary buckets to use for the classes. Must be >= 0.
initializer Initializer function that accepts a 1-D tensor as the arg to specify the shape of the returned tensor. If None, defaults to using zeros_initializer().
max_rows_in_memory int specifying the maximum number of rows to load from the checkpoint at once. If less than or equal to 0, the entire matrix will be loaded into memory. Setting this arg trades increased disk reads for lower memory usage.

A variable initializer function.