Happy New Year! May 2019 be better than 2018 but not as good as 2020 😁

NLP 2018 Highlights

#NLP #AI #Conversational_AI #ML #

https://github.com/omarsar/nlp_highlights/blob/master/NLP_2018_Highlights.pdf

While I never professionally worked on graph based machine learning problems, they have always been fascinating and I have tried keeping up to date with newish papers. Today I came across a really nice package called AmpliGraph (https://lnkd.in/gpXYuuQ), available on pip and running on top of TensorFlow. The API looks very clean with a number of example notebooks. Excited to play around with this.

#machinelearning #ML #datascience #graphs #MLLM #Cubonacci #AI

New tutorial!🚀

Learn how to build an Image Hashing Search Engine that scales to 1,000,000s of images using #OpenCV, #Python, and VP-Trees.

Full tutorial w/ code here: http://pyimg.co/myj11 👍 #ComputerVision #MachineLearning #ArtificialIntelligence #DataScience #AI #BigData

Awesome Graph Classification

A collection of important graph embedding, classification and representation learning papers with implementations.

GitHub, by Benedek Rozemberczki: https://lnkd.in/eErZBnh

#graph2vec #deepgraphkernels #graphattentionmodel
#graphattentionnetworks

"Mathematics For Machine Learning"

A book that is intended to help people understand the #mathematics behind the #MachineLearning techniques.

Its aim is to make people understand what goes under the hood in common ML algorithms.

The best part is that the team is also working on Jupyter notebook tutorials

Download the PDF of the book: https://lnkd.in/e-gXPRf

100% OFF in Home Delivery Asia 2019>>> https://lnkd.in/f_TxgKN

For Data Science Implementations:
Know Data Science https://lnkd.in/fMHtxYP
Understand How to answer Why https://lnkd.in/f396Dqg
Machine Learning Terminology https://lnkd.in/fCihY9W
Understand Machine Learning Implementation https://lnkd.in/f5aUbBM
Machine Learning on Retail https://lnkd.in/fihPTJf
and Marketing https://lnkd.in/fBncKiy

Detecting new knowledge in unstructured text using ML. More evidence that when you put large amounts of papers and reports together and apply OpenSource machine learning to the text - the whole can be greater than the sum of its parts. This paper focuses on Thermoelectric materials.

Vice News Article
https://lnkd.in/gkXnEXt

Nature Paper (Tshitoyan et al 2019)
https://www.nature.com/articles/s41586-019-1335-8

Podcast:

Ali Ghodsi On Building A $2.7B Business And Being Considered One Of The True Founders Of Artificial Intelligence

https://alejandrocremades.com/ali-ghodsi/

The 5 graph algorithms that you should know

Rahul Agarwal describes some of the most important graph algorithms you should know and how to implement them using Python.

Detecting and treating outliers is a necessity in any dataset as it inevitably introduces the deviation in the model estimations. It can make the difference between winning and loosing a data science competition.

https://lnkd.in/fMV6GaY

This article deals with the detection of the outliers in Time Series data using different ideas, every idea improving upon the previous one and finally treating the outliers in the best way possible.

Hint of ideas covered.....
Idea #1 — Winsorization
Idea #2 Standard deviation etc.

An article covering the case study over "Customer Transaction Prediction using LightGBM".

https://medium.com/analytics-vidhya/https-medium-com-kushagrarajtiwari-customer-transaction-prediction-3191c6c634dc

It comprehensively covers:
1. General Business Significance of this problem
2. Exploratory Data Analysis
3. Feature Engineering
4. Why use LightGBM for this problem

A good read if you want to explore problems in bank/financial domain.

Automating the end-to-end lifecycle of Machine Learning applications
#CD4ML #software_engineering #ML

Discoverable and Accessible Data
Reproducible Model Training
Model Serving (Embedded model, Model as service)
Testing and Quality in Machine Learning
Experiments Tracking
Model Deployment (Multiple models, Shadow models)
Model Monitoring and Observability

https://martinfowler.com/articles/cd4ml.html

Microsoft open-sourced scripts and notebooks to pre-train and finetune BERT natural language model with domain-specific texts

Github: https://github.com/microsoft/AzureML-BERT
Earlier: https://t.me/opendatascience/837

#Bert #Microsoft #NLP #dl

Great collection of practical rules for routine DS engineering / research job.

Machine Learning in a company is 10% Data Science & 90% other challenges, this pdf provides a great deal of principals and solutions to deal with them.

We can only recommend saving this post to your Saved Messages by forwarding it to yourself.

Link: http://martin.zinkevich.org/rules_of_ml/rules_of_ml.pdf

#cheatsheet #advice #practical #common #shouldbesaved

CS238: Decision Making under Uncertainty (AA 228)

Textbook: Decision Making Under Uncertainty: Theory and Application by Mykel J. Kochenderfer et al. (MIT Lincoln Laboratory Series)
See course materials

http://web.stanford.edu/class/aa228/