nyu-mll / jiant-v1-legacy

The jiant toolkit for general-purpose text understanding models
MIT License
21 stars 9 forks source link

[CLOSED] Fix number of training examples in target training #1071

Closed jeswan closed 4 years ago

jeswan commented 4 years ago

Issue by HaokunLiu Thursday Apr 16, 2020 at 04:40 GMT Originally opened as https://github.com/nyu-mll/jiant/pull/1071


Now target_train_data_fraction will correctly affect the number of training examples in trainer, and logging and scheduler that depend on it.

We will need this merged before running the 5k probing tasks in taskmaster.


HaokunLiu included the following code: https://github.com/nyu-mll/jiant/pull/1071/commits