When using pandas for standard data science: all the functions When using pandas for chemical data transformation: yeah everything's too complicated; df.apply() on every line so I can use my custom functions 🐼 - ThreadSky

supersciencegrl.co.uk • 21 days ago

When using pandas for standard data science: all the functions

When using pandas for chemical data transformation: yeah everything's too complicated; df.apply() on every line so I can use my custom functions 🐼

Comments

kizzythechemist.bsky.social•21 days ago

oh don't remind me.
df.[dataframe].str.replace("A","B")... ad nauseam. I learnt really quickly why questionnaires typically don't use free text boxes much!

supersciencegrl.co.uk•21 days ago

...Because people can f up free text responses even if they are told precisely what to write in the box? 😩

kizzythechemist.bsky.social•21 days ago

for this specific data set: people have so many more ways of describing their own identity than you realise.

popher.bsky.social•21 days ago

We have a form going out soon which has a *total* word limit of 300, but it has four separate input sections... I can guarantee that folks will put 300 words in each section.
(I am not making the form, its not my fault).

cheminformania.com•21 days ago

I made pdchemchain for reusable and easy dataprocessing of dataframes with a chemical view. I strived to make it easy to make new reusable classes while still getting all the goodies of self documentation and serialization for command line reusage. Take a look, and let me know if we should chat.

supersciencegrl.co.uk•21 days ago

I confess I hadn't seen it! I will take a look for sure!

cheminformania.com•21 days ago

If you make some cool custom links, be sure to reach out. It could use a bit of expansion with more links. Is it reaction transforms or what are you working on?

supersciencegrl.co.uk•21 days ago

The link worked fine for me on Android native Bluesky app!
I'm actually just looking up chemical data from often very limited, free text input, and trying to augment it into something useful!

cheminformania.com•21 days ago

Oh. now its there😄

cheminformania.com•21 days ago

Somehow bluesky eats the link, but its easy to google

cheminformania.com•21 days ago

https://github.com/EBjerrum/pdchemchain

georgekarag.bsky.social•21 days ago

lambdas are your friend. RDKit also has PandasTools to help but sometimes it's just easier to use .apply

supersciencegrl.co.uk•21 days ago

I'm not dealing with structures (other than relieving them from PubChem) yet so no chance to use rdkit here! Everything is .apply() 😂

cosimofedeli.bsky.social•21 days ago

Not a chemist, but df.apply() has nevertheless been my companion in many an adventure.

popher.bsky.social•21 days ago

At least you're smart enough to do apply.

I just write some code with
for index, data in df.iterrows():

and then when I realise its way too slow, have lab AI refactor the code to be vectorised.

supersciencegrl.co.uk•21 days ago

After quite some time, I am *now* 😂😂😂

Comments

Posting Rules

Reply