MACHINE LEARNING

→ View original post on X — @erictopol

Plasma metabolic signature predicts longevity cardiovascular cancer mortality

By

@erictopol

–

15 February 2024 18h40

How a healthy lifestyle is associated with longevity, less cardiovascular and cancer-related mortality: A plasma metabolic signature of lipid pathways assessed in 4 US cohorts with 28-year follow-up accounts for ~50% https://
cell.com/med/fulltext/S
2666-6340(24)00040-0
… @ajtessierRD @MartaGuasch1

→ View original post on X — @abacusai

Smaug-72B Tops Hugging Face Leaderboard as Best Open-Source Model

By

@abacusai

–

15 February 2024 18h02

Two weeks ago we released Smaug-72B – which topped the Hugging Face LLM leaderboard and it’s the first model with an average score of 80, making it the world’s best open-source foundation model. We applied several techniques on a fine tune derived from a Qwen-72B for this model.

→ View original post on X — @sassoftware

AI adoption in fraud detection lags despite strong plans

By

@sassoftware

–

15 February 2024 17h46

83% of fraud fighters plan to add #GenAI in the next 2 years. But the reality can lag the vision: Use of AI & ML in combating #fraud has grown by just 5% in 5 years. Why? And what are the most promising anti-fraud uses for AI? New survey from @TheACFE and SAS has answers

Python ML MLOps CV NLP LLMs Daily Tutorials

By

–

15 February 2024 15h31

If you are interested in: – Python – Machine Learning – MLOps – CV/NLP – LLMs Find me → @Sumanth_077 Everyday, I share tutorials on above topics! Like/RT the first tweet to help this reach more people!

Implementing _predict_single Method for KNN Classification

By

–

15 February 2024 15h31

Finally a '_predict_single' method as mentioned it predicts the class label for a single data point This is done by: 1. Calculating the distance between one point to all training examples 2. Selecting the k-nearest Neighbours, & returning the most common class label among them

Predict Method for Class Labels Using Single Point Prediction

By

–

15 February 2024 15h31

A 'predict' method that predicts the class labels for a set of data points. It calls '_predict_single' method that actually predicts the class label for a single data point.

Implementing fit Method and Euclidean Distance in KNN

By

–

15 February 2024 15h31

Next define the "fit" method that takes the training data(X) and it's corresponding labels(y) and stores them. This is the reason why the computation cost is high as it stores all the training data! And also a method to calculates the euclidean distance between two data points.

KNNClassifier Implementation: Choosing the Right k Value

By

–

15 February 2024 15h31

Define the KNNClassifier class and initialize the number of Neighbours (k) to be considered in __init__ method Defining the K value is critical in KNN! A small value of k means noise will have a higher influence on the result and a large value makes it computationally expensive