wdm0006

5 ms·

1.
by wdm0006 20d ago
I use a lot of different harnesses, so in CC i use sonnet as main w/ opus advisor, in codex i use astra all the time, in obvious I usually use the GLM models. In general I've found that having a fast main model with a big smart ad
2.
by wdm0006 1mo ago
The trick is to make, and care deeply about, fewer bigger decisions. Savviness will always matter but the things to be savvy about change.
3.
by wdm0006 9mo ago
https://mcginniscommawill.com
4.
by wdm0006 10mo ago
I think the main point of a personal blog is to write consistently about things you think are neat, and allow the world to find it, assess it, and those who also find those things neat will stick around. What is neat means different things
5.
Zero-code-change CUDA acceleration for scikit-learn(developer.nvidia.com)
3 points by wdm0006 1y ago|0 comments
6.
Stargazers: A CLI for downloading GitHub stars data(mcginniscommawill.com)
1 points by wdm0006 1y ago|0 comments
7.
Estimating the time spent on a project with git-pandas(willmcginnis.com)
1 points by wdm0006 10y ago|0 comments
8.
LIME: reasonable interpretations for black-box models(homes.cs.washington.edu)
3 points by wdm0006 10y ago|0 comments
9.
Bayesian Survival Analysis in Python with Pymc3(austinrochford.com)
2 points by wdm0006 10y ago|0 comments
10.
File details and owners with gitnoc and git-pandas(willmcginnis.com)
3 points by wdm0006 11y ago|0 comments
11.
Estimating code quality with survival analysis and git-pandas(willmcginnis.com)
1 points by wdm0006 11y ago|0 comments
12.
Overoptimizing: a story about Kaggle(willmcginnis.com)
4 points by wdm0006 11y ago|0 comments
13.
Using python's inspect module to ensure module consistency(willmcginnis.com)
2 points by wdm0006 11y ago|0 comments
14.
Plot your GitHub cumulative progress in 5 lines of python(willmcginnis.com)
6 points by wdm0006 11y ago|0 comments
15.
Pygeohash: now with basic stats(willmcginnis.com)
1 points by wdm0006 11y ago|0 comments
16.
Full-team Git dashboards with python and git-pandas(willmcginnis.com)
5 points by wdm0006 11y ago|0 comments
17.
Common data pitfalls for recurring ML systems(willmcginnis.com)
3 points by wdm0006 11y ago|0 comments
18.
Create organization-wide punchcards with python and git-pandas(willmcginnis.com)
3 points by wdm0006 11y ago|0 comments
19.
Even further beyond one-hot: feature hashing(willmcginnis.com)
1 points by wdm0006 11y ago|0 comments
20.
Gitpandas v0.0.6: python 2 support, file-wise blame, and more(willmcginnis.com)
1 points by wdm0006 11y ago|0 comments
21.
Beyond one-hot: an exploration of categorical variables(willmcginnis.com)
1 points by wdm0006 11y ago|0 comments
22.
Python examples with Apache Flink(github.com)
2 points by wdm0006 11y ago|0 comments
23.
by wdm0006 11y ago
I'm not proud of the amount of time I just spent searching for a picture of panda bear on github to use in an example. I was not successful.
24.
Analyze Git repositories with pandas: git-pandas(willmcginnis.com)
15 points by wdm0006 11y ago|2 comments
25.
by wdm0006 11y ago
Author here. Source is available at: https://github.com/wdm0006/flink-python-examples , let me know if there are any other examples or features that would be helpful.
26.
Using the new python API in Apache Flink v0.10.1(willmcginnis.com)
4 points by wdm0006 11y ago|1 comments