https://github.com/colcarroll/ai_talk
WIP Slides for Boston AI Meetup
https://github.com/colcarroll/ai_talk
Last synced: 10 months ago
JSON representation
WIP Slides for Boston AI Meetup
- Host: GitHub
- URL: https://github.com/colcarroll/ai_talk
- Owner: ColCarroll
- License: mit
- Created: 2017-01-24T16:58:58.000Z (over 9 years ago)
- Default Branch: master
- Last Pushed: 2023-04-06T22:59:57.000Z (about 3 years ago)
- Last Synced: 2025-07-28T16:36:59.963Z (11 months ago)
- Language: Jupyter Notebook
- Size: 2.7 MB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Slides for 2-21-17 talk at the Boston AI Meetup
## Looking at the slides
1. Clone this repo
### ...with Python
2. Run `python -m http.server` (`python -m SimpleHTTPServer` on 2.x)
3. Open http://localhost:8000.
### ... with Node (4.0.0 or later) (also allows hot reloading)
2. `npm install`
3. `npm. start`
4. Open http://localhost:8000.
## Running the code
These slides started as an interactive Jupyter notebook, using the
[RISE project](https://github.com/damianavila/RISE). See website for
installation.
There is a `requirements.txt` in the `talk_code` folder -- all
code was written in python 3.6, but will probably work with 2.7, 3.4, 3.5.
Also required is data from
[Kaggle](https://www.kaggle.com/c/march-machine-learning-mania-2017) --
the scripts expect to see the two files
- RegularSeasonDetailedResults.csv
- Teams.csv
in the same directory (`talk_code/`) as the scripts.
Open `slides.ipynb` to run the code. Some calculations are done lazily and
cached, so it may be awful slow the first time through.