SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2203.13834 · CVPR · 2022

A Stitch in Time Saves Nine: A Train-Time Regularizing Loss for Improved Neural Network Calibration

Ramya Hebbalaguppe, Chetan Arora, Jatin Prakash, Neelabh Madan

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 5 functions out of this paper's own repositories and ran 1 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
mdca-loss/mdca-calibration canonical 1 of 5
FunctionStatusWhere it lives
conv3x3 Ran mdca-loss/mdca-calibration/models/resnet.py
code served (permissive licence) · get_code("fac5364e2f53c6db")
conv3x3 Not yet run mdca-loss/mdca-calibration/models/resnet_imagenet.py
code served (permissive licence) · get_code("2ace1c98fa5f2cd5")
get_datasets Not yet run mdca-loss/mdca-calibration/datasets/imagenet.py
code served (permissive licence) · get_code("f077346ebc45ed1b")
get_train_valid_test_loader Not yet run mdca-loss/mdca-calibration/datasets/imagenet.py
code served (permissive licence) · get_code("c4fa9bed9d3b3fe8")
resnet18_pacs Not yet run mdca-loss/mdca-calibration/models/resnet_pacs.py
code served (permissive licence) · get_code("0ba4e80f87b2a344")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

https://github.com/mdca-loss Airplane 92 % Bird 7.2 % Frog 0.0 % (a) GT: Bird Truck 69 % Automobile 29 % Airplane 0.8 % (b) GT: Truck Airplane 52 % Bird 47 % Frog 0.0 % Truck 99 % Automobile 0.0 % Airplane 0.0 % Person 99 % Dog 0.1 % Guitar 0.0 % (c) GT: Person Guitar 95 % Person 2.0 % Dog 1.2 % (d) GT: Person Person 99 % Dog 0.2 % Guitar 0.0 % Guitar 45 % Person 30 % Dog 8.4 %

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2203.13834")
get_code_for_paper("2203.13834")
have("2203.13834")

Connect an agent — have() is free.