100 Days of Machine Learning

Various machine learning related codes for my 100 days of machine learning project.

The installations needed for all the upcoming projects. This I learnt from my Udacity Self Driving Car course.

Learn editing markdown file here

Preparing system for projects

1. Install miniconda
2. Create a new conda environment
3. Activate the environment before staring your work

You can also create a conda env ml_100 by copying the environment.yml from this repo and running

conda env create -f environment.yml

This will install all necessary packages. To see a list of all of your environments

conda env list

After installing cleanup downloaded files using

conda clean -tp

To activate ml_100

source activate ml_100

To deactivate an env

source deactivate

To delete an env

conda env remove -n ml_100

Hopefully all went well. So after activating the env start the Jupyter notebook in the background

jupyter notebook&

It will open the ipynb notebook in the browser. You are all set to start coding.

Week 1 (not neccessarily in order)

1. Deep Learning by Udacity
2. Computer Vision Class by Devi Parikh
3. Computer Vision by Udacity

Good Reads:

1. Hyperparameters and model validation
2. Receptive field arithmetic for CNN
3. How to handle imbalanced classification problem and this
4. Interesting questions for Data Science
5. Understand the maths behind gradient from Khan Academy
6. Andrej Karpathy's cs231 who doesn't know this, seriously?
7. A list of all courses available for any concepts related to deep learning, machine learning, computer vision and so on, thanks to the post in reddit. Deep learning drizzle
8. I find Waymo's blog very interesting to read as you get to know all latest technologies that are coming up for self driving cars.

Papers from recent ML

1. AutoML
2. Learning transferable architecture for scalable image recognition
3. SING, NeurIPS 2018
4. Deep anomaly detection using geometric transofrmation, NeurIPS 2018