Convert fp16 `val` to fp32

Convert fp32 `val` to fp16

Args: tok: id of token, ex 4 idx_to_tag: dictionary {4: "B-PER", ...} Returns: tuple: "B", "PER"

Gets a collection of `InputExample`s for the dev set.

See base class.

Gets a collection of `InputExample`s for the train set.

Convert a sequence of head indexes into a tree object.

Copy model gradients to master gradients. Args: model_params: List of model parameters created by :func:`prep_param_lists`.

An iterator over the sentences in an individual CONLL formatted file.

We take `pre-tokenized` input here, because we don't have a tokenizer in this class.

Train the model

Convert a tree object to an (numpy) adjacency matrix.

Outputs random values from a truncated normal distribution. The generated values follow a normal distribution with specified mean and standard

https://stackoverflow.com/questions/3232943/update-value-of-a-nested-dictionary-of-varying-depth

Copy the ``.grad`` attribute from stored references to fp16 parameters to the ``.grad`` attribute of the fp32 master parameters that

Constructs a InputExample. Args: guid: Unique id for the example. text_a: string. The untokenized text of the first se

?

Initialize the weights.