openfl

openfl copied to clipboard

#456 Issue linked. Updated the document.

sun1lach

The pytorch kvasir unet CI test has been triggering a CI timeout sporadically in the middle of training: https://github.com/intel/openfl/runs/7525106079?check_suite_focus=true https://github.com/intel/openfl/runs/7524916214?check_suite_focus=true https://github.com/intel/openfl/runs/7510403486?check_suite_focus=true 

psfoley

Fixes #454 and #455 Collaborator now retries the call to the aggregator if federation connection is lost while sending results from collaborator -> aggregator

mansishr

walteriviera

A user has reported that if the network connection is lost during the GetAggregatedTensor RPC call from a collaborator to the aggregator, the collaborator cannot successfully reconnect after the outage....

psfoley

A user has reported that if the network connection is lost during the SendLocalTaskResults RPC call from a collaborator to the aggregator, the collaborator cannot successfully reconnect after the outage....

psfoley

Adding Google Colab notebook to the MedMNIST 2D example.

mansishr

Can’t use custom loss or metrics in Director-Based workflow

4

**Describe the bug** Can’t use custom loss or metrics to train a model with `tf.keras.Model` APIs in Director-Based workflow. Configurations: - OpenFL workflow: Director-Based - Model to train: 3D U-Net...

sky712345678

Corrupt shard utils and experiments

1

This PR introduces Corrupting Shard Descriptor wrapper. Decorating a Shard Descriptor (SD) class with this function will result in the following: 1. `corrupt` parameter is added to the SD init...

igor-davidyuk

Director / Envoy option to prompt user for experiment review before execution

2

The interactive API is currently intended for environments where the user wishes to do frequent experimentation without needing to redistribute the workload manually. To make the interactive API safer for...

psfoley