Repository for Jupyter notebooks as part of BG ML learning day. Initially covering these topics:
- Clustering
- Random Forests
Instructions for running jupyter notebooks:
- Pull these notebooks from this repo to your working directory (Or simply copy from /scratch/cbarth/BG_ML if unfamiliar with git)
- From working directory run the following in the terminal (e.g. for the Random Forest notebook):
module load scitools
jupyter notebook Random_Forest_Tutorial.ipynb
(if asked for password credentials, use your normal Windows username/password)
This should load up a Firefox page with the Jupyter notebook which you can interactively work through.
Work through the notebook by running the cells of code. (A shortcut to run a cell is to click on it and press 'Shift+Enter')
Have a play with changing hyperparameters etc. in the code to see what difference they make. You can rerun any cell to overwrite previously saved data.
Any questions, just ask Karen or Claire or post them on the MS Teams Machine Learning learning channel under BG Science and Consultancy.
- Deep Learning, Goodfellow, Bengio and Courville: https://www.deeplearningbook.org
- Neural Networks and Deep Learning, Nielsen: http://neuralnetworksanddeeplearning.com
- Scikit-learn User Guide: https://scikit-learn.org/stable/user_guide.html
- Python Data Science Handbook (with Jupyter notebooks): https://jakevdp.github.io/PythonDataScienceHandbook/
- My neural network isn't working! What should I do? http://theorangeduck.com/page/neural-network-not-working
- Towards Data Science (Medium): https://towardsdatascience.com
- Keras blog: e.g: https://blog.keras.io/how-convolutional-neural-networks-see-the-world.html
- Distil.pub: https://distill.pub
- Andrew Ng's Machine Learning course (YouTube): https://www.youtube.com/playlist?list=PLLssT5z_DsK-h9vYZkQkYNWcItqhlRJLN
- Or the Stanford lecture course (also Andrew Ng): https://www.youtube.com/playlist?list=PLA89DCFA6ADACE599
- Caltech "Learning from Data" course: https://work.caltech.edu/lectures.html
- Google's Machine Learning crash course: https://developers.google.com/machine-learning/crash-course/
- NCAR's Machine Learning Summer School lectures: https://www.youtube.com/playlist?list=PLbelYhZAAHEIr4iC1FNcPXUUYXI0zg_96
- Data Science Community of Practice Yammer and Sharepoint: https://metoffice.sharepoint.com/sites/MetOfficeDataScienceCommunity
- Most papers in the ML community are on arXiv