Felipe Fernandez

Data Engineering and MLOps.

Still figuring it out, let's figure it out together

🇨🇴 in 🇨🇦

[LatinX in AI (LXAI) @_LXAI]

Dive into the fascinating world of image classification with this captivating article by #LatinXinAI community member Felipe Fernandez! Learn how #AI distinguishes between animals and anime characters.
#MachineLearning #DeepLearning #Pytorch
https://t.co/DYA2TE0TXy

[LatinX in AI (LXAI) @_LXAI]

Learn how to build and deploy an object detection model using YOLOv8, Roboflow, and HuggingFace! Dive into this detailed guide by #LatinXinAI community member @Learningdata20 and improve your skills in #ObjectDetection
#DeepLearning #ArtificialIntelligence
https://t.co/BHOzzSdBzH

[Felix Zumstein @felixzumstein]

Microsoft has released Data Wrangler, a VS Code extension that gives you a UI for data viewing & cleaning and that generates clean Python/pandas code. I really hope that they rip out Power Query in Excel and replace it with this! I'd go as far a saying that this should have been the first deliverable for Python in Excel...https://t.co/qKtSjt3iL2 #pythoninexcel #vscode #pandas #microsoftexcel

[DeepLearning.AI @DeepLearningAI]

In Building Agentic RAG with @Llama_Index, you'll move beyond the basic RAG pipeline.

Let’s build a router for handling QA and summarization over a single doc, an agent reasoning loop, and an agent capable of managing multiple documents.

Enroll for free: https://t.co/EWGvffus9A

[Pau Labarta Bajo @paulabartabajo_]

Looking for real-world ML project ideas? 💡

Let's build an ML system that predicts loan defaults.

With 𝗳𝘂𝗹𝗹 𝘀𝗼𝘂𝗿𝗰𝗲 𝗰𝗼𝗱𝗲👷🏾‍♀️👷↓

[Vadym Kazulkin @VKazulkin]

"How to Be a Better Software Engineering Leader" https://t.co/ErQL6Ca5se

[Sam Altman @sama]

using technology to create abundance--intelligence, energy, longevity, whatever--will not solve all problems and will not magically make everyone happy.

but it is an unequivocally great thing to do, and expands our option space.

to me, it feels like a moral imperative.

[merve @mervenoyann]

Everything open-source LLMs at @huggingface
https://t.co/Kg5JRmoBNU 💬

[merve @mervenoyann]

community-led computer vision course just dropped to @huggingface Hub 📖🤍

[Christoph Molnar 🦋 christophmolnar.bsky.social @ChristophMolnar]

Anyone here using shapash?

(Python library for explainable AI based on SHAP)

I just had a nice chat with the developers behind shapash and they have a typical open source developer problem - they don't know who is actually using the library.

[Bojan Tunguz @tunguz]

Not a cell phone in sight, just two friends, present in the moment, enjoying a meal together.

[LangChain @LangChain]

🚀Build a Streamlit Chatbot using Langchain, ColBERT, Ragatouille, and ChromaDB

Great resource on how to use ColBERT + Ragatouille +
@trychroma + mixtral-8x7b-instruct-v0.1 + our EnsembleRetriever

s/o @aigeek__ for building!

Code: https://t.co/Yh2Ti1eV29

[Alexey Grigorev @Al_Grigor]

We started the 2nd module of our Data Engineering Zoomcamp with @mage_ai!

We'll cover:

🔸 ETL: API to Postgres, API to GCS, GCS to BigQuery
🔸 Parameterized Execution
🔸 Deployment and Advanced Blocks (Optional)

Join module 2:
https://t.co/9M8hHCD82L

[Valeriy M., PhD, MBA, CQF @predict_addict]

🚀 **Exciting news in #TimeSeries #Forecasting!** 🚀

Meet MFLES – a new Python library developed by Tyler Blume that's transforming the landscape of forecasting. It's not just another tool; it's a leap forward for data enthusiasts and professionals!

Let's dive into what MFLES offers:
✅ **Multiple Seasonality Support**: Navigate through complex data patterns with ease, from daily fluctuations to yearly trends.
✅ **Conformal Prediction Intervals**: Embrace precision with MFLES. These intervals aren't just predictions; they're a reliable gauge of future scenarios, ensuring you're always a step ahead.
✅ **Seasonality Decomposition**: Decode the hidden rhythms in your data, uncovering underlying trends that are key to insightful forecasts.
✅ **Parameter Optimization**: Fine-tune your forecasts to achieve unparalleled accuracy, making each prediction more relevant and precise.
🔜 **Custom Exogenous Models** (Coming Soon): Gear up to incorporate external insights, adding another layer of sophistication to your forecasting models.

And here's the clincher: The benchmark results are in! MFLES has been rigorously tested against renowned models like Nixtla's AutoETS and MSTL, especially in scenarios involving multiple seasonality. The outcome is a testament to its robust capabilities. 📊

For anyone in the realm of forecasting, this is a tool to watch!

#DataScience #MachineLearning #AIInnovation#ForecastingFuture #timeseries #forecasting

Benchmark showcasing MFLES against Nixtla's AutoETS and MSTL (for multiple seasonality comparison). 📊

[SkalskiP @skalskip92]

parking occupancy analysis

calculation of percentage occupancy in individual parking zones

all this was done with supervision: https://t.co/xXMRaS3Guk

btw, @UenoLeo is cooking a blog post covering this project, so stay tuned!

↓ read more

[Carlos Santana @DotCSV]

🔴 ¡NUEVO MODELO de META!

Meta vuelve a la carga con el primer regalo del año ―Code Llama 70B― Un modelo de la familia Llama orientado a programación que podremos descargar e integrar en nuestras herramientas 🔥⛈️

[Felipe Fernandez @Learningdata20]

Today, I worked on a new project, it's simple but effective to start showcasing my knowledge in certain subjects.

More convinced than ever of the importance of personal branding

Here we go.

#DataScience #Machinelearning

[Dan Kornas @DanKornas]

👉 Machine Learning Interviews

This repo aims to serve as a guide to prepare for Machine Learning (AI) engineer interviews for roles at big tech companies (in particular FAANG).

🔗 https://t.co/fwZoAAq4Rc

[Carlos Santana @DotCSV]

Pues ya se empiezan a conseguir resultados chulos en transferencia de estilo a vídeos! ✨