Skip to content

GitLab

Commit 3b7017a5 authored by Chen Chen's avatar Chen Chen Committed by A. Unique TensorFlower
Browse files

Add documentation to explain the input_path in QADataConfig. 

PiperOrigin-RevId: 347443298
parent 957f32ad
Show whitespace changes
Inline Side-by-side
Showing with 2 additions and 0 deletions
official/nlp/data/question_answering_dataloader.py
View file @ 3b7017a5
...@@ -27,6 +27,8 @@ from official.nlp.data import data_loader_factory ...@@ -27,6 +27,8 @@ from official.nlp.data import data_loader_factory
@dataclasses.dataclass @dataclasses.dataclass
class QADataConfig(cfg.DataConfig): class QADataConfig(cfg.DataConfig):
"""Data config for question answering task (tasks/question_answering).""" """Data config for question answering task (tasks/question_answering)."""
# For training, `input_path` is expected to be a pre-processed TFRecord file,
# while for evaluation, it is expected to be a raw JSON file (b/173814590).
input_path: str = '' input_path: str = ''
global_batch_size: int = 48 global_batch_size: int = 48
is_training: bool = True is_training: bool = True
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment