Buy @ Amazon

Search This Blog

January 20, 2019

Conda environments not showing up in Jupyter Notebook

This post is based on my experience of fixing this issue in Ubuntu OS. You may have to make some tweaks depending on your OS to see if this process fixes the issue.

I wasted a couple of hours in fixing this. If you think this has saved your time, don't forget to share it in your circles to help save other's time. Cheers!


January 18, 2019

Concatenate Datasets in Python

Often in data-science or machine learning hackathons, you may want to concatenate the rows of the train and test datasets for easy data exploration and  wrangling. And when you want to do that, this pro-tip comes handy.

Check the Pandas Documentation for more on this API.

January 10, 2019

Choosing Subset of Columns from a Pandas Dataframe

It is so common a situation in data-science to select a subset of columns from a data-set. In Python the usual approach is to select a set of columns using List Comprehension or using pandas df.drop() method.

I typically employ using the List Comprehension method of choosing a subset of columns. The drawback of this approach is that it is verbose over its drop() counterpart. However, there is a distinct advantage with this approach which is that this approach guarantees idempotency.

Being a huge fanboy of idempotency coming from a mathematical background and  a hatred for verbosity and duplication, I came end-up using the utility methods shown in this blog post. Clearly, this utility function gives the advantage of brevity and idempotency. 

Yay, I win! 

January 1, 2019

Service Orchestration And Service Choreography

Service Orchestration

When you think Service Orchestration, visualize the picture above. You will have an Orchestrator who controls the individual components in the live performance. An Orchestrator typically is present in the live performance to control the individuals' performance.