Experimentation with Papermill, and MLFlow

[techwithshadab]

Abstract: Machine Learning and AI are changing or would say have changed the way how businesses used to behave. However, the Data Science community is still lacking good practices for organizing their projects and effectively collaborating and experimenting quickly to reduce “time to market”. During this session, we will learn about such open-source tools like “papermill” and “mlflow” which can help you in experimenting, and managing the lifecycle of ML models.

Duration: 45mins

[mayankskb]

Hi Shadab,

Thank you for making an issue. Can you please help me with the line up that you are planning to cover? Also, please help me with your availability for late this month.

Thanks Mayank Mishra

[techwithshadab]

Hi Mayank, Below will be the break-out of the topics:

1. Understanding the normal process a Data Scientist/Machine Learning Engineer goes through while building models and hyperparameter tuning
2. Which tools they can use to ease the experimentation
3. Why its difficult to monitor/track metrics and model performance and how they can overcome it
4. Hands-on demo wherein they will learn about the integration of papermill and mlflow

I'm available on the last weekend of this month.

Regards, Shadab

[mayankskb]

Hi Shadab,

Let's not keep the session in this month. Lets schedule it for Nov 21. Please let me know your availability. If not on Nov 21 then later that week?

Thanks Mayank Mishra

[techwithshadab]

Sure Nov 21 works for me

[mayankskb]

Finalising it for nov 21. Please send me brief about about you and a headshot for flyer.

Thanks

[mayankskb]

Hi Shadab,

Due to some reasons, we were not able to conduct the talk earlier. Please let me know if we can schedule it for 20 March.

Thanks Mayank Mishra

[techwithshadab]

Hi Mayank, Sure, 20th March works for me.

Regards, Shadab

[mayankskb]

Hi Shadab,

Please send me brief about about you and a headshot for flyer. You can send it on mail pydata.pune@gmail.com