Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

JOSS Feedback - Readme clarification #15

Merged
merged 2 commits into from
Mar 3, 2023
Merged

JOSS Feedback - Readme clarification #15

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
7589a18
docs: update readme
a9p Mar 3, 2023
3b9b44f
Minor edits to README
arsedler9 Mar 3, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 11 additions & 2 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,16 @@

## Introduction

This repository provides a set of solutions for running [AutoLFADS](https:/snel-repo/autolfads-tf2/tree/main/lfads-tf2) in different compute environments and workflows. Use the solution matrix as a rough guide for identifying an suitable workflow:
This repository provides a set of solutions for running [AutoLFADS](https:/snel-repo/autolfads-tf2/tree/main/lfads-tf2) in a wider variety of compute environments. This enables more users to take better advantage of the hardware available to them to perform computationally demanding hyperparameter sweeps.

![](paper/solutions.png)

We provide three options for different cluster configurations and encourage the user to select the one that best suits their needs:
- _Local Compute_: users directly leverage a container image that bundles all the AutoLFADS software dependencies and provides an entrypoint directly to the LFADS package. Interactivity with this workflow is provided via YAML model configuration files and command line arguments.
- _Unmanaged Compute (Ray)_: users configure a Ray cluster and interact with the workflow by updating YAML model configurations, updating hyperparameter sweep scripts, and then running experiment code.
- _Managed Compute (KubeFlow)_: users interact with a KubeFlow service by providing an experiment specification that includes model configuration and hyperparameter sweep specifications either as a YAML file or using a code-less UI-based workflow.

The solution matrix below provides a rough guide for identifying an suitable workflow:

| | Local Container | Ray | KubeFlow |
|---------------------------------------------------|-----------------|-----------|---------------|
Expand All @@ -14,7 +23,7 @@ This repository provides a set of solutions for running [AutoLFADS](https://gith
| Infrastructure | Local | Unmanaged | Managed/Cloud |
| Cost | $ | $ - $$ | $ - $$$ |


> Details describing the AutoLFADS solutions and evaluation against the Neural Latents Benchmark datasets can be found in our [paper](paper/paper.pdf).

## Installation & Usage

Expand Down