Roadmap to learn AI for associates at McCarthy Lab@Next Tech Lab
- Best places to learn -
- Python
- NumPy
- Pandas
- Matplotlib and Seaborn
- scikit-learn
- SciPy
- Basic computer architecture:
- Linux
- Containers - Docker
- Bash Cheatsheet
- Git-Introduction
- PyTorch
- Use fast.ai as High level wrapper
- TensorFlow
- Use tf.keras as a High level wrapper
- Effective TensorFlow
- Stanford ALgorithms - Coursera or
- Introduction to Algorithms (MIT 6.006)
- Introduction to Computational Thinking and Data Science (MIT 6.0002)
- Learn any one programming language really well and compete on Codechef, Hackerrank, HackerEarth,etc
Note :
- Implement Machine Learning models from scratch using Python
- Once you're comfortable implementing models from scratch, learn scikit-learn and compare performance
- Practice on Kaggle to get your skiills ---> 😎
Feel free to use any of these frameworks, all are not required
- Introduction to Statstical Learning
- Elements of Statistical Learning (A little more in-depth than ISLR)
- Pattern Recognition And Machine Learning
Note : Learn from official tutorials/docs or GitHub repos which have detailed notebooks like Hvass Labs
- The Matrix Calculus You Need For Deep Learning - - Quick refresher
- Mathematics for Machine Learning - Intermediate
- Numerical Algorithms - Advanced
- Natural Language Processing by National Research University Higher School of Economics
- NLP course by Yandex Data School