Building end-to-end Machine Learning Systems
Joined March 2023
- Tweets 3,195
- Following 459
- Followers 203
- Likes 2,651
322 Photos and videos
Pinned Tweet
Apr 20
Telecom Customer Churn.
An end to end MLOps Projects.
Built using
MLflow
Airflow
DVC
Streamlit
Fastapi
AWS
Github actions
Docker
All the updates and down below.
3
1
12
653
Apr 20
Telecom Customer Churn.
An end to end MLOps Projects.
Built using
MLflow
Airflow
DVC
Streamlit
Fastapi
AWS
Github actions
Docker
All the updates and down below.
3
1
12
653
May 2
May 2
Successfully Ran the Airflow DAG.
No Errors. each steps executed smoothly.
As scheduled it's running the project in every 15 min.
It ran the project, gave the output - Recall : 0.96.
MLFlow tracking the project assigned airflow in it.
1
19
May 27
May 27
I built a ML pipeline using Airflow MLFlow Fastapi Streamlit. containerized in Docker
Took 10 days to learn and implement.
Each service working successfully.
Airflow -> MLFlow -> FastAPI -> Streamlit
Using Docker compose for different containers, different dockerfile
6
May 27
I built a ML pipeline using Airflow MLFlow Fastapi Streamlit. containerized in Docker
Took 10 days to learn and implement.
Each service working successfully.
Airflow -> MLFlow -> FastAPI -> Streamlit
Using Docker compose for different containers, different dockerfile
1
3
84
May 27
> In Docker the first service i'm using is Airflow.
It is running the project.
> MLflow is saving tha experiments and models
> FastAPI , picking the latest model trained by airflow from saved models in MLFlow
> Streamlit, user interface for predicting the outout using FastAPI
2
1
41
May 27
it is working successfully, now i have new things to do
> PostgreSQL
> AWS
> Github Action (CI/CD)
i'll go with postgresql and AWS
1
33
May 18
Raw journaling while building a project is the best things to do, daily journaling writing
what you doing and what not,
what made me question,
the issues i got and how i solved it, process i followed
the problem: i'm stuck understanding docker,airflow,mlflow internal working
3
30
May 17
i'm not getting what is Volumes,Ports and how they in docker with MLflow, i want to understand it by creating a new toy project which would be using mlflow and it will me understand how experiments, tracking and data works in mlflow
3
28
May 17
after 10 days i came back , didn't touch my mac keyboard, now i'm goosebumming with it
1
14
Rishav retweeted
May 12
For LA
- 3Blue1Brown Essence of Linear Algebra
- Maths for ML LA by Imperial College London
For Calculus
- Essence of Calculus series
- How to Master Calculus 1 by Ace Tutors
Learned Statistics & Probability
- CampusX playlist Maths for ML
- Statistics Fundamental By StatQuest
10
2
39
1,552
May 2
After spending 5 days on model training, data cleaning, preprocessing,evaluation,testing. as backened and uses a button to predict Churned/not churned.
Feels machine learning is 10% and 90% engineering about how to use that prediction and make it reach to customer and get result
1
89
May 2
Airflow schedules the projects and runs in every 15min/hour/day.
No need to run the project manually, it automates and orchestrates the project, each step is executed after each one.
MLFlow tracks the performance, params, evaluation metrics of project store. Logged in
May 2
Successfully Ran the Airflow DAG.
No Errors. each steps executed smoothly.
As scheduled it's running the project in every 15 min.
It ran the project, gave the output - Recall : 0.96.
MLFlow tracking the project assigned airflow in it.
1
2
297
Rishav retweeted
May 2
POV: everyone’s got a macbook
and you walk in with a gaming laptop 😂
104
722
11,143
1,016,846
May 2
Project update guys
Telecom customer churn.
May 2
Successfully Ran the Airflow DAG.
No Errors. each steps executed smoothly.
As scheduled it's running the project in every 15 min.
It ran the project, gave the output - Recall : 0.96.
MLFlow tracking the project assigned airflow in it.
1
29
May 2
Successfully Ran the Airflow DAG.
No Errors. each steps executed smoothly.
As scheduled it's running the project in every 15 min.
It ran the project, gave the output - Recall : 0.96.
MLFlow tracking the project assigned airflow in it.
1
379
May 2
Errors i got and how i solved then
the very first error i got
Segmentation fault error:
the error was coming cause the file path was not correct
gave it the right path it ran.
Then it gave the error again.
i changed the n_jobs=-1 to n_jobs=4, using less cores in Airflow.
1
23
May 2
Simply i read docs understood the topics and concepts and executed it used very simple code
made it much simpler just to orchestrate and schedule the project locally.
All the files and codes are running locally
16
Apr 28
Saved more time playing 5 min Blitz
thinking > winning.
i would play 300 elo for 20 times.
understand, handle every situation than just increasing elo and playing higher and higher.
thinking during the game matters the most
1
26