50+ days of Machine Learning

 Hello :) Today is Day 140! A quick summary of today: Ch 1 - calculus review and plain vanilla options Ch 2 - improper integrals, numerical integration, interest rates and bonds Ch 3 - probability, Black-Scholes formula, Greeks and hedging  From the books I found yesterday I thought that starting from this one is the best because of its pure math + real application format. Each chapter consists of some math/stats and also how they are used in the real world. Ch 1 - calculus review and plain vanilla options Ch 2 - improper integrals, numerical integration, interest rates and bonds Ch 3 - probability, Black-Scholes formula, Greeks and hedging  On another note ~ today I realised that calling this whole blog project self-study into AI might not be the best. Because math, probability and stats are at the root of AI. So maybe some kind of renaming would be appropriate.  That is all for today! See you tomorrow :) 

 Hello :) Today is Day 139! A quick summary of today: learned how to host a mlflow server on AWS I managed to get my card working and finally finished the set up on my AWS account, so I could finish the last bit of module 2 from the MLOps zoomcamp . First, go to EC2 console And set up an instance We need to select instance type (for me it is the free one) Then we need to create a key-pair for logging in And it is done  Next we need to create an S3 bucket Next, we create a postgresql db Some configs to choose And I got a link for the db on port 5432 (which I later use for the mlflow server). We need to give access to EC2 to the database And we have the console Where we have to install some dependencies Install dependencies: pip3 install mlflow boto3 psycopg2-binary I had to create a user and add some permissions so that it can connect to S3, EC2 and RDS Then I got it running So just like when doing it locally we have the db (ofc run on aws), and the mlflow hosted as well To mak...

 Hello :) Today is Day 138! A quick summary of today: fine-tuned microsoft's speech T5 to create a Glaswegian TTS and published it on huggingface  (using our project's dataset ) I started the day by meeting with my collaborator (as I am writing this I remembered I forgot to ask them if I can mention them by name today ...) for an hour discussing my progress last week on the Scottish phrases audio clips and how we can move forward. After the short clips, next we can move forward with transcribing podcasts, or him/friends doing voice recordings for our dataset because the 1st option might be too much manual labour.   During the call M asked me how the whole model operation happens and about embedding the text and audio, and after the meeting I decided to dig a bit and see how *once* we have a good dataset, how do we actually use it. I found this  course from huggingface that introduces working with audio data. One of the units is specifically about TTS. From there...

 Hello :) Today is Day 137! A quick summary of today: attended 'data analysis' and 'business tech' sessions during Day 2 of AWS Summit in Seoul I tried to take more notes with my laptop during the session today. First session - 'Data strategy for successful GenAI on AWS' A GenAI app is like an iceberg.  Under the water it consists of among other things data storage, database, data lake, data manipulation, data governance.  The session introduced us to RAG, and how we can either use an out-of-the-box language model, pre-train a model, or train a model from scratch.  As for a vector search AWS offers plenty of options: Amazon OpenSearch Service, OpenSearch Serverless, Aurora PostgreSQL, RDS for PostgreSQL, DocumentDB, DynamoDB via zero ETL, MemoryDB for redis, Neptune. As for a database, there are plenty of structured and nonstructured dbs.  Amazon Bedrock - allows for using out-of-the-box models for GenAI applications. It gives us the ability to use a simple API ...

 Hello :) Today is Day 136! A quick summary of today: Attended day 1 of AWS summit Seoul and sat in the 'banking, finance and fintech' talks - everything was related to AI (not surprisingly haha) It was all in Korean, so below I will try to give a general overview of the attended sessions + some extra activities that I did.  Something funny and interesting. There was a booth where you took a picture and your face was put on a medieval/pre-historic/renaissance figure. It is an event in Korea so I was curious whether it can recognise white people faces, and below is supposed to be me Haha. Interesting ^^ There was also a football kick booth where I queued for around 30mins to shoot twice and to try to kick and score with the highest speed (I placed rank 24 with 72km/h kick on target). They also had a model that takes a picture and puts a football jersey on the person Next, about the sessions.  They were all in Korean (the slides as well) and even though I took pictures putt...

 Hello :) Today is Day 135! A quick summary of today: covered module 2 of the mlops-zoomcamp by DataTalks club about experiment tracking and model management cut 2 more videos for the Scottish dataset project Firstly, about using mlflow for MLOps Maybe this is because I am starting to learn about MLOps for the 1st time and I don't know other tools, but WOW mlflow is amazing. Below are my notes from the module 2 lectures. Full code on my github repo. First, some important concepts ML experiment: the process of building an ML model experiment run: each trial in an ML experiment run artifact: any file that is associated with an ML run experiment metadata What's experiment tracking? the process of keeping track of all the relevant info from an ML experiment (could include source code, environment, data, model, hyperparams, metrics, other - these can vary) Why is experiment tracking important? reproducability organization optimization Why can't we just use an excel spreadsheet...