- Berkeley
- https://fir.sh
- @bfirsh
Highlights
Block or Report
Block or report bfirsh
Report abuse
Contact GitHub support about this user’s behavior. Learn more about reporting abuse.
Report abusePinned
-
-
arxiv-vanity/engrafo Public
Convert LaTeX documents into beautiful responsive web pages using LaTeXML.
-
A guide to help you write better command-line programs, taking traditional UNIX principles and updating them for the modern day.
-
-
1,613 contributions in the last year
Less
More
Activity overview
Contributed to
replicate/cog,
replicate/cog-examples,
arxiv-vanity/arxiv-vanity
and 5 other
repositories
Contribution activity
March 2022
Created 11 commits in 1 repository
Created 1 repository
- bfirsh/Wave-U-Net-Pytorch Python
Created a pull request in replicate/cog that received 27 comments
Design & document HTTP API
This is a work-in-progress implementation of #413. I was considering creating an issue to discuss the design of the HTTP API, but I figure a pull r…
+61
−0
•
27
comments
Opened 10 other pull requests in 1 repository
replicate/cog
8
merged
1
open
1
closed
- Fix output documentation
- s/build, ship, run/package/
- Update compatibility matrices
- Add python 3.6 to CI to see if it fails
- Remove references to nascent AMQP API
- Support int choices
- Add types to enums in the schema
- Remove cog.Predictor
- Make progressive output a list
- Only set custom Output object if it's not already called Output
Reviewed 7 pull requests in 2 repositories
replicate/cog
6 pull requests
pixray/pixray
1 pull request
Created an issue in replicate/cog that received 2 comments
Design: file handling
As was started to explore here, we need a way to upload and download files over the HTTP API and the queue-based API. Some of this is partially imp…
2
comments
Opened 12 other issues in 1 repository
replicate/cog
10
open
2
closed
- Check for CUDA and Nvidia driver compatibility
- Specify what GPUs a model is compatible with and check compatibility on boot
- Don't be tied hard to Tensorflow CPU packages
- Document how to use a Jupyter notebook with Cog
- Design proposal: Progressive output as a list
-
Remove
Anyoutput from getting started guide - Fix any TODOs/FIXMEs
- Espouse some of the virtues of Cog in the readme
-
Remove redundant
titlein OpenAPI schema - Only allow particular output types
-
Rename
leandgetominimumandmaximumincog.Input -
Input/Outputnaming in OpenAPI schema is fragile
47
contributions
in private repositories
Mar 1 – Mar 28
