Kiran Karra

how to use all_gather in training loop?

11

I have defined my train_step in the exact same way as in the [cifar10 example](https://github.com/pytorch/ignite/blob/master/examples/contrib/cifar10/main.py#L319). Is it possible to gather all of the predictions before computing the loss? I haven't...

Feature/unsup multichan waveform dataset

5

Added Dataset which supports multichannel audio-samples. Updated collator to drop pad-tracks.

Help ingesting events into ElasticSearch

1

Hello, I am trying to setup the event exporter to send events to ElasticSearch, but am getting the following error, which I can observe by looking at the kubernetes-event-exporter logs...

Wondering if anyone has built (or has plans to) build support for nn.MaxPool2d?

Separating repeated processing from classifier models

3

In between different runs of the ATM, the outputs of all the steps of the pipeline are "static," except for the input and output to the classifier that is chosen...

Lesson Feedback%3A Verifying Storage on Filecoin - Lesson 2 (Preparing and transferring data)

1

Link to CAR file needs to be updated from old to: https://ipld.io/specs/transport/car/

Fip0081 gradual activation

10

This pull request updates FIP0081 to include a ramped activation. Pseudocode is provided for how a ramped activation could be realized. The pseudocode is written from the perspective of how...

Moved constants to separate file

2

- Added comment for updating 2.8888 to 2.76 - Updated baseline b0 multiply value to be clearer