Adamized
@crikvoke
21 | AI/ML undergrad | Currently doing Deep Learning | blogs on http://medium.com/@paryagsahni2
will try to deploy this by the end of the day . lets go!
Built an end-to-end ML project to predict student exam scores - Used the "Student Performance" dataset from Kaggle - Designed a modular PIPELINE for ingestion, preprocessing, training, and evaluation - Created a Flask-based UI for real-time score prediction
Me getting ready to watch Breaking Bad at 3 am after finally pushing code to github
Grind your ass off anons,the world doesn't talk about Losers Good Morning.
"Unfortunately we will not be moving forward with your application, and will be choosing to hire another candidate"
Imagine you’re learning guitar. If you keep messing up chord A, a teacher will start correcting you less on it (you already got lots of correction). But if you rarely touch chord G, and you mess up, the teacher gives more attention to it. That’s Adagrad.
You’re training a model, and the training accuracy is 95%, but the validation accuracy is only 70%. What’s likely happening, and how might you fix it?
DL update - Softmax: turning logits into class probabilities - Loss functions: Regression - MSE, MAE, RMSE, Huberloss Classification - BCE, CCE, Sparse CCE - Optimizers: Gradient Descent , sgd , mini batch sgd Up next: Adagrad and Adam - implemented chapter 4
I needed this
Let’s address the elephant in the room. Do applied data scientists or AI/ML engineers need deep, research-level knowledge of machine learning? Or is a basic understanding and intuition (without memorizing formulas) enough? Applied ML engineers do not need deep, research-level…
The 🐐obviously
Name a modern day cricketer that gives the best interviews?
A project covers the entire lifecycle from data ingestion to model deployment which includes data preprocessing, model training, evaluation, saving artifacts, and optionally exposing the model via an API or UI.
What does end to end machine learning project look like?