@ramikrispin@mstdn.social
@ramikrispin@mstdn.social avatar

ramikrispin

@ramikrispin@mstdn.social

Data science and engineering senior manager at ๏ฃฟ | #rstats & #Python | ๐Ÿ“ฆ dev | โค๏ธ time-series analysis & forecasting | Author. Opinions are my own | https://linktr.ee/ramikrispin

This profile is from a federated server and may be incomplete. Browse more on the original instance.

ramikrispin, to python
@ramikrispin@mstdn.social avatar

Thanks to @medium Staff for selecting my recent article - Introduction to Multi-Stage Image Build for Python ๐Ÿ, for a boost โค๏ธ!

This tutorial provides a step-by-step guide for converting a regular Python Dockerfile into a multi-stage build ๐Ÿš€.

๐Ÿ”—: https://medium.com/towards-data-science/introduction-to-multi-stage-image-build-for-python-41b94ebe8bb3

#python #docker #mlops #medium #datascience

ramikrispin, to python
@ramikrispin@mstdn.social avatar

(1/4) ๐ˆ๐ง๐ญ๐ซ๐จ๐๐ฎ๐œ๐ญ๐ข๐จ๐ง ๐ญ๐จ ๐Œ๐ฎ๐ฅ๐ญ๐ข-๐’๐ญ๐š๐ ๐ž ๐ˆ๐ฆ๐š๐ ๐ž ๐๐ฎ๐ข๐ฅ๐ ๐Ÿณ ๐Ÿ๐จ๐ซ ๐๐ฒ๐ญ๐ก๐จ๐ง ๐Ÿ

The size of the Docker image could quickly increase during the build time. I became more mindful of the image size when I started to deploy on Github Actions. The bigger the image size, the longer the run time and the higher the runtime cost.

This is when you should consider using a multi-stage build ๐Ÿš€.

๐Ÿงต๐Ÿ‘‡๐Ÿผ

#docker #mlops #python #DataScience #medium

ramikrispin,
@ramikrispin@mstdn.social avatar

(2/4) ๐–๐ก๐š๐ญ ๐ข๐ฌ ๐š ๐ฆ๐ฎ๐ฅ๐ญ๐ข-๐ฌ๐ญ๐š๐ ๐ž ๐›๐ฎ๐ข๐ฅ๐?
Multi-stage is a build approach that includes the following two steps:
:l1: The first build - installing the required decencies and building the binaries. This image is also called the builder image
:l2: The second build - starting from a new base image and copying from the builder image binaries applications

ramikrispin,
@ramikrispin@mstdn.social avatar

(3/4) ๐–๐ก๐ž๐ง ๐ฌ๐ก๐จ๐ฎ๐ฅ๐ ๐ฒ๐จ๐ฎ ๐ฎ๐ฌ๐ž ๐š ๐ฆ๐ฎ๐ฅ๐ญ๐ข-๐ฌ๐ญ๐š๐ ๐ž ๐›๐ฎ๐ข๐ฅ๐?
You should consider moving your build to a multi-stage build when the build-required dependencies are no longer needed after the build is completed. A classic example is when building a binary application. Also, this is effective when setting up a dockerized Python environment using a virtual environment.

#python #docker

ramikrispin,
@ramikrispin@mstdn.social avatar

(4/4) I created the following tutorial for setting up a dockerized Python environment using a multi-stage approach ๐Ÿ‘‡๐Ÿผ

https://medium.com/towards-data-science/introduction-to-multi-stage-image-build-for-python-41b94ebe8bb3

Happy Build! ๐Ÿณ๐Ÿ—๏ธ

#Docker #MLops #Python #DataScience #medium

ramikrispin, to ComputerScience
@ramikrispin@mstdn.social avatar

(1/2) Data Compression: Theory and Applications - Stanford Course ๐Ÿ‘‡๐Ÿผ

Stanford University released a new course on data compression methods taught by Prof.Tsachy Weissman, Shubham Chandak, and Pulkit Tandon. As the demand for data increases at an exponential rate, data compression plays a pivotal role in providing efficient storage solutions. The course focuses on the foundations and theory of data compression.

ramikrispin,
@ramikrispin@mstdn.social avatar

(2/2) This full semester course has 18 lectures and covers the following topics:
โœ… Lossless compression fundamentals
โœ… Lossy compression
โœ… Special topics

Course lectures ๐Ÿ“ฝ๏ธ: https://www.youtube.com/playlist?list=PLoROMvodv4rPj4uhbgUAaEKwNNak8xgkz
Course website ๐Ÿ”—: https://stanforddatacompressionclass.github.io/Fall23/

ramikrispin, to datascience
@ramikrispin@mstdn.social avatar

Forecasting Time Series with Gradient Boosting โค๏ธ

The skforecast Python ๐Ÿ library provides ML applications for time series forecasting using different regression models from the scikit-learn library. Here is a tutorial by Joaquรญn Amat Rodrigo and Javier Escobar Ortiz for time series forecasting with the skforecast using XGBoost, LightGBM, Scikit-learn, and CatBoost models ๐Ÿš€.

๐Ÿ“–๐Ÿ”—: https://cienciadedatos.net/documentos/py39-forecasting-time-series-with-skforecast-xgboost-lightgbm-catboost

image/png
image/png

ramikrispin, to llm
@ramikrispin@mstdn.social avatar

I am excited to present at the OSDC East conference next week about using LLM to create language to SQL code generator.

https://odsc.com/speakers/data-automation-with-llm/

ramikrispin, to llm
@ramikrispin@mstdn.social avatar

Llama 3 is already available on Ollama ๐Ÿš€๐Ÿ‘‡๐Ÿผ

https://ollama.com/library/llama3

#llm #llama3 #ollama #python #DataScience

ramikrispin, to llm
@ramikrispin@mstdn.social avatar

(1/3) Llama 3 is out! ๐Ÿš€

Meta released today Llama 3, the next generation of the Llama model. LLama 3 is a state-of-the-art open-source large language model. Here are some of the key features of the model: ๐Ÿงต๐Ÿ‘‡๐Ÿผ

#llama #llama3 #llm #python #DataScience #MachineLearning #deeplearning

video/mp4

ramikrispin,
@ramikrispin@mstdn.social avatar

(2/3) โœ… The new model comes with integration into the core cloud providers such as AWS, GCP, Azure, Databricks, Hugging Face, etc.
โœ… The model supports a variety of hardware architectures, such as AMD, AWS, Dell, Intel, NVIDIA, and Qualcomm.
โœ… Supporting safety and cyber tools
โœ… The model comes in two versions - 7B and 80B

ramikrispin,
@ramikrispin@mstdn.social avatar

(3/3)
More details are available on the release post ๐Ÿ‘‡๐Ÿผ
https://ai.meta.com/blog/meta-llama-3/

Playground: https://www.meta.ai/

ramikrispin, to llm
@ramikrispin@mstdn.social avatar

RAG from Scratch with LangChain ๐Ÿฆœ๐Ÿ‘‡๐Ÿผ

FreeCodeCamp released today a new course on building RAG from scratch with LangChain. The course, which is by Lance Martin from LangChain, focuses on the foundations of Retrieval Augmented Generation (RAG).

Course ๐Ÿ“ฝ๏ธ: https://www.youtube.com/watch?v=sVcwVQRHIc8
Code ๐Ÿ”—: https://github.com/langchain-ai/rag-from-scratch

ramikrispin, to datascience
@ramikrispin@mstdn.social avatar

New release to Ollama ๐ŸŽ‰

A major release to Ollama - version 0.1.32 is out. The new version includes:
โœ… Improvement of the GPU utilization and memory management to increase performance and reduce error rate
โœ… Increase performance on Mac by scheduling large models between GPU and CPU
โœ… Introduce native AI support in Supabase edge functions

More details on the release notes ๐Ÿ‘‡๐Ÿผ
https://github.com/ollama/ollama/releases

Image credit: release notes

ramikrispin, to datascience
@ramikrispin@mstdn.social avatar

Gemini API Cookbook ๐Ÿš€

Google released a new repo with a collection of guides and examples for the Gemini API. This includes a set of guides for prompt engineering and examples of the API features ๐Ÿ‘‡๐Ÿผ

๐Ÿ”— https://github.com/google-gemini/cookbook

#DataScience #MachineLearning #llm #deeplearning #Python

ramikrispin, to datascience
@ramikrispin@mstdn.social avatar

(1/3) Learn R Through Examples ๐Ÿš€๐Ÿ‘‡๐Ÿผ

The Learn R Through Examples by Xijin Ge, Jianli Qi, and Rong Fan provides an introduction to data analysis with R. The book covers the core topics of data analysis using different datasets, from simple and clean datasets to messy and big datasets. ๐Ÿงต๐Ÿ‘‡๐Ÿผ

#RStats #DataScience #datavisualization #data

image/png

ramikrispin,
@ramikrispin@mstdn.social avatar

(2/3) This includes the following topics:
โœ… Working with data frames
โœ… Data visualization with base R and ggplot2
โœ… Data structures
โœ… Summary statistics and correlation analysis
โœ… Case studies - analyzing multiple datasets

ramikrispin,
@ramikrispin@mstdn.social avatar

(3/3) Book ๐Ÿ“š: https://gexijin.github.io/learnR/index.html

Thanks to the authors for making this book available for free online! ๐Ÿ™๐Ÿผ

Image credit: from the book

ramikrispin,
@ramikrispin@mstdn.social avatar

@AlanSill my go-to solution for parallelize execution in R is the mclapply - which enables running lapply function in parallel using multi CPUs:
https://www.rdocumentation.org/packages/parallel/versions/3.4.0/topics/mclapply

ramikrispin,
@ramikrispin@mstdn.social avatar

@dzegpi I am not familiar about books or resources focusing on forecasting with tidymodels besides the work of Matt Dancho.

ramikrispin, to datascience
@ramikrispin@mstdn.social avatar

Happy Sunday!

Backpropagation Calculus ๐Ÿš€๐Ÿš€๐Ÿš€

This short video by Grant Sanderson provides a great explanation of the math beyond the backpropagation algorithm using calculus ๐Ÿ‘‡๐Ÿผ

https://www.youtube.com/watch?v=tIeHLnjs5U8

#DataScience #math #deeplearning #MachineLearning

ramikrispin, to datascience
@ramikrispin@mstdn.social avatar

Hands-on Data Science: Complete your First Project ๐Ÿš€

This beginner crash course by Misra Turp provides an introduction to the foundations of data science by solving real-life examples. This includes the different steps of a data science project, from setting the environment to loading and analyzing the data using Python, git, Jupyter notebooks and other tools ๐Ÿ‘‡๐Ÿผ

๐Ÿ“ฝ๏ธ: https://www.youtube.com/playlist?list=PLM8lYG2MzHmTgsYKLJtdKwf6tHVbui9eE

ramikrispin, to python
@ramikrispin@mstdn.social avatar

(1/2) A new release to PyMC ๐Ÿš€๐Ÿš€๐Ÿš€

This week, PyMC version v5.13.0 was released. PyMC is one of the main ๐Ÿ libraries for ๐๐š๐ฒ๐ž๐ฌ๐ข๐š๐ง statistics โค๏ธ. It provides a framework for probabilistic programming, enabling users to build models with a simple Python API and fit them using ๐Œ๐š๐ซ๐ค๐จ๐ฏ ๐‚๐ก๐š๐ข๐ง ๐Œ๐จ๐ง๐ญ๐ž ๐‚๐š๐ซ๐ฅ๐จ (MCMC) methods ๐Ÿš€.

The new release includes new features, bug fixes ๐Ÿž, and documentation improvements ๐Ÿ“–. More details on the release notes ๐Ÿ“ ๐Ÿ‘‡

ramikrispin,
@ramikrispin@mstdn.social avatar

(2/2) ๐‹๐ข๐œ๐ž๐ง๐ฌ๐ž ๐Ÿชช: Apache 2.0 ๐Ÿฆ„

Release notes ๐Ÿ“: https://github.com/pymc-devs/pymc/releases/tag/v5.13.0
Documentation ๐Ÿ“–: https://www.pymc.io/welcome.html

stevensanderson, to github
@stevensanderson@mstdn.social avatar
ramikrispin,
@ramikrispin@mstdn.social avatar

@stevensanderson what are the applications of this function?

  • All
  • Subscribed
  • Moderated
  • Favorites
  • โ€ข
  • JUstTest
  • mdbf
  • ngwrru68w68
  • modclub
  • magazineikmin
  • thenastyranch
  • rosin
  • khanakhh
  • InstantRegret
  • Youngstown
  • slotface
  • Durango
  • kavyap
  • DreamBathrooms
  • megavids
  • GTA5RPClips
  • tacticalgear
  • normalnudes
  • tester
  • osvaldo12
  • everett
  • cubers
  • ethstaker
  • anitta
  • Leos
  • provamag3
  • cisconetworking
  • lostlight
  • All magazines