I have finished the refactoring the code for vaex and also writtten test cases to compare the spectrum calculated using pandas with the dataframe calculated using vaex dataframe . Also, Various spectroscopic quantities as absorbance , emissitivity is also compared for the both the dataframams. Also, there was many issues that was raised by the maintainers and I have resolved almost all of these , and commented on the other issues to discuss the problem and discuss some possible solution .Issues raised by the maintainers was mainly related to make changes more matainable and easy to understand and simple programming logic is preferred inplace of using some complex code without explaining that in detail. Also , the issue was to ensure a light test suite , that is test cases which takes less resources and time . Initialy , I didn’t focused on this thing and focused on testing the code and changes more elaborately by writing the test cases that cover many areas of code .

For testing specturm produce using vaex and pandas for non-equilibrium calculations are same , the code similar to equilibrium calculations is used

I passed the midterm evaluation, Next i have to refactor part of code in vaex which is used in non-equilibrium calculations.

Though Vaex reduced memory use by RADIS to compute specturm but it is slow for smaller databank and in our case when the number of lines in the databank is very less . The slow performance of Vaex for smaller dataframes is due to three main reasons for our implementation of RADIS

After completing all the changes to compute Spectrum using Vaex , I compared the memory used during the execution of the program . I used tracemalloc to compute memory uses to compute Spectrum .

After the community bonding period , I continued refactoring the code for vaex . I have the following the things in this period