North Korea and United States of America are the only serious global powers we have today. Who will keep their promises even amidst adversity. U.S struck Iran to protect it's buddy, Israel, against the backdrop of international outcry and opposition. N. Korea supported Russian aggression in Ukraine and sent in Arms and deployed troops despite threats from US and Europe.
IRAN IS JUST FINDING OUT THAT YOU TRUST CHINA AND RUSSIA AT YOUR OWN PERIL! Those Burkina Faso, Mali,Niger fellows, take note!
It is increasingly becoming difficult for Israel and the United States to convince the global south and much of the population of the world that Iran is the bad and evil. The moment you take unilateral actions against global rule of law you cannot be as good as you want the world to believe. The US is going to see massive anti war protests soon. The mid terms will be interesting for MAGA. Fuel prices to soar as well.
Maybe - just maybe - the aliens are waiting at our galaxy's door, watching this entire spectacle while munching on intergalactic popcorn and waiting for the movie to start.
And when, eventually, the movie ends, they will swoop in for the spoils.
Maybe.
Great friends and partners: Prime Minister of Italy @GiorgiaMeloni and President of the EU Commission @vonderleyen. Italy’s Mattei Plan commits €5.5 billion and Global Gateway of the @EU_Commission commits €150 billion to Africa. Thank you very much!🙏🏽💕
"ANY RETALIATION BY IRAN AGAINST THE UNITED STATES OF AMERICA WILL BE MET WITH FORCE FAR GREATER THAN WHAT WAS WITNESSED TONIGHT. THANK YOU! DONALD J. TRUMP, PRESIDENT OF THE UNITED STATES"
JUST IN: 🇮🇷🇺🇸 Iran is open to negotiations with the United States, NYT reports.
"A Senior Iranian official from the Foreign Ministry, who asked that his name not be used because he was not authorized to speak publicly, said that Iran would accept President Trump's offer to meet soon.
Let's build a 𝗿𝗲𝗮𝗹-𝘁𝗶𝗺𝗲 𝗠𝗟 𝘀𝘆𝘀𝘁𝗲𝗺, with Python, Rust, 𝗟𝗟𝗠𝘀 and Kubernetes.
Step by step ⬇️
𝗢𝘂𝗿 𝗴𝗼𝗮𝗹 🎯
Let's build an ML system that can predict crypto prices 5 minutes into the future.
The data sources we will use are
> Real time market prices
> Real time market news about blockchain, or any economic factor that can possibly impact the target metric we want to predict.
𝗦𝘆𝘀𝘁𝗲𝗺 𝗗𝗲𝘀𝗶𝗴𝗻 📐
Let's break down our system into 4 types pipelines
- Feature pipelines (2 of them)
- Training pipeline
- Inference pipeline
- Monitoring pipeline
Let's go one by one
1️⃣ 𝗙𝗲𝗮𝘁𝘂𝗿𝗲 𝗽𝗶𝗽𝗲𝗹𝗶𝗻𝗲𝘀
Our 2 feature pipelines transform raw data into reusable ML model features, and save them in our Feature Store Hopsworks
- Feature pipeline for technical indicators
> Ingests market trades in from a websocket API like Kraken Digital Asset Exchange
> aggregates them into 1-minute windows
> Engineers technical indicators to capture momentum in the. market, and
> Pushes them to the Feature Store
- Feature pipeline for market sentiment
> Scrapes news by pooling a website like coinbase,
> Parses the raw text into a structure sentiment score using an LLM, and
> Pushes them to the Feature Store
2️⃣ 𝗧𝗿𝗮𝗶𝗻𝗶𝗻𝗴 𝗽𝗶𝗽𝗲𝗹𝗶𝗻𝗲
We can use a boosting tree model like XGBoost to uncover any patterns between the
- current technical indicators, and
- market senttiment
and
- the price of BTC or ETH 5 minutes into the future
The final model is pushed to the model registry.
3️⃣ 𝗜𝗻𝗳𝗲𝗿𝗲𝗻𝗰𝗲 𝗽𝗶𝗽𝗲𝗹𝗶𝗻𝗲
To serve the final predictions, while making them easily accessible to our next pipeline (the monitoring pipeline) I recommend you split this step into 2 services:
- Prediction generator, that loads the model from the registry and continuously uses the latest features to generate a new prediction and push it o a persistent storage, like an Elastic Search index.
- Prediction API. This is a lightweight service that
> Receives incoming requests from client apps,
> Finds the predictions in the Elastic Search index, and
> Returns the prediction to the client.
4️⃣ 𝗠𝗼𝗻𝗶𝘁𝗼𝗿𝗶𝗻𝗴 𝗽𝗶𝗽𝗲𝗹𝗶𝗻𝗲
There are lots of things you can monitor here. The most fundamental one is the error of your predictions.
To monitor these you need to build a streaming service that
> Listens to incoming price data
> Loads the predictions from the DB
> Computes the error and
> Pushes the error into another Elastic Search index
These errors are plotted on a dashboard using Kibana, and trigger alerts to your Slack or Discord team.
BOOM!
𝗪𝗮𝗻𝗻𝗮 𝗹𝗲𝗮𝗿𝗻 𝘄𝗶𝘁𝗵 𝗺𝗲?
In March 450+ students and myself will build this system together.
Wanna join?
-> https://t.co/rBeeM1lrr0
Stupid isn't about you, it's about democracy,🤓! We pick it from President Clinton who said "Its Economy Stupid " during his campaign. Sry wasn't meant for you,my dear.
Thank you very much President @oliguinguema of the Republic of Gabon for conferring on me the prestigious national honour of the Grand Officer of the Gabonese Order of Merit. @PresidenceGA @AfDB_Group