This repo is a collection of Microsoft ML experiment Jupyter notebooks on the Microsoft ML Server. I will continue adding to this repo whenever I create a new experiment notebook.
To use this repo, you need to have Microsft Machine Learning Server installed.
If you're too lazy to install all of it, you can use this Docker image to open up a Jupyter notebook instance with docker run -p 8888:8888 -v pwd:/home/app -it johnpaulada/microsoftml after doing steps 1 and 2 below.
- Clone this repo by running
git clone https://github.com/johnpaulada/microsoftml-experiment-notebooks.giton your console. - Go into the directory by running
cd microsoftml-experiment-notebookson your console. - Open up a Jupyter notebook.
- Import these notebooks.
- Run them and there you go! You can modify them and try things out! 🎉
- Comment Flagging (Sentiment Analysis) - This experiment uses the
get_sentimentMicrosoft ML function to flag negative comments from a csv file. - BASS Philippine Mobile Connection Recommender (Pandas & Numpy) - Recommends the fastest mobile connection in the area (Smart, Globe, Sun, etc.). Uses BASS data.
MIT