Skip to content
This repository has been archived by the owner on Nov 28, 2023. It is now read-only.

Added class weights to the CrossEntropy Loss #120

Merged
merged 5 commits into from
Feb 12, 2020
Merged

Added class weights to the CrossEntropy Loss #120

merged 5 commits into from
Feb 12, 2020

Commits on Feb 11, 2020

  1. Added class weights to the CrossEntropy Loss

    @sonjageorgievska
    sonjageorgievska committed Feb 11, 2020
    Configuration menu
    Copy the full SHA
    36a7417 View commit details
    Browse the repository at this point in the history

Commits on Feb 12, 2020

  1. update class weight

    @CunliangGeng
    CunliangGeng committed Feb 12, 2020
    Configuration menu
    Copy the full SHA
    7633882 View commit details
    Browse the repository at this point in the history

  2. udpate batch size in test examples 

    the batch size must be smaller than total number of samples when using batchnorm, otherwise `drop_last` will remove all sampels and give an empty batch.
    @CunliangGeng
    CunliangGeng committed Feb 12, 2020
    Configuration menu
    Copy the full SHA
    ca1b347 View commit details
    Browse the repository at this point in the history

  3. set torch version < 1.4.0

    @CunliangGeng
    CunliangGeng committed Feb 12, 2020
    Configuration menu
    Copy the full SHA
    554a01a View commit details
    Browse the repository at this point in the history

  4. set distr to bionic

    @CunliangGeng
    CunliangGeng committed Feb 12, 2020
    Configuration menu
    Copy the full SHA
    d021ad0 View commit details
    Browse the repository at this point in the history