Utilising community bonding period Effectively!
I started to contribute to RADIS from January’23 and whenever i needed help in some issue , mentors were quite helpful in providing revelant information to resolve that issue . And this thing helped me , to develop good bonding and understanding with the mentors .
So,I thought, i should use community bonding period more productively thus, i decided to work on project from 15 May .
In community bonding period i have done following things -
- Read the RADIS and HITRAN Paper (selected parts)
- Worked on Loading the dataframe in Vaex format (task of first week)
Loading Dataframe in Vaex format
There are two main functions which are used load databank in RADIS These are :
- fetch_databank()
- load_databank()
fetch_databank() : It is used to load databank from standard databank like HITRAN,HITEMP,EXOMOL,GEISA by fetching it from their respective APIs, then parsing and processing them.
load_databank() : It is used to load databank from the local file or to load databank from user defined databank .
Upto now, i have worked on fetch_databank() function to load dataframe in Vaex format , some of the things were already implemented while at other points i needed to write code specifically for vaex dataframe format.
Similarly, for load_databank() function , the hurdle was to parse it in vaex dataframe format as virual columns which are used in vaex dataframe to reduce memory use , it was throwing error . After trying many i finally found a fix for it .
Now, I have made necessary changes to these two functions to load dataframe in Vaex format. Next week i will be working on writing test cases for these.