/ML-

First neural network using tensorflow

Primary LanguageJupyter Notebook

ML-

First neural network using tensorflow

TODO: /

  1. IMPLEMENT LOGISTIC REGRESSION [/ ]

  2. IMPLEMENT KNN .

  3. IMPORT PANDAS , NUMPY PRACTICE FILES

  4. SKLEARN PRACTICE FILES

  5. SKLEARN PIPELINES

  6. CUSTOM DATASETS

  7. IMPLEMENT CNN build a custom dataset and perform image processing

  8. Build a basics Graph Neural Network - lvl1 lvl2 - Tune the whole model to a production ready version .

Project Idea Data Engineering: Build Large custom datasets by implementing fetcher and preprocessing units to periodically retieve data from various sources using python and pandas . Implement and maintain a robust and dynamic PostgresSQL database using SQLAlchemy

Neural Network Modelling : Develop UNET object detection and for detecting defects on wafers using Pytorch

Implement Reward Constrained Policy Optimization (RCPO) into stable baseline3 implementation of proximal policy optimization (ppo) using pytorch

Try publishing papers

Build a Machine learning API

RLHF : Implement a Reinforcement learning paper .

Recommender systems .

Build large datasets using pyspark .

Data Mining

Sequential Modelling predictive models

Q and A:

  1. Why are regularization techniques useful in generational models
  • Prevent overfitting and improve model's ability to generalize to new data

  • dropout -> Disables some neurons during training , focrcing model to rely on mutiple features

  • weight decay -> Penalizes large weights - ensuring simpler and more general features are learned

  • spectral normalization ->