pandas
SoftwareWikipedia
39
MENTIONS
17
EPISODES
14
PODCASTS
Search complete. 39 mentions across 17 episodes found for "pandas".
Nov 10, 2026
Network ChuckHOST
First of all, you would need some hardware, servers with GPUs.
Network ChuckHOST
Then you would also need a bunch of tools and libraries and SDKs like PyTorch and TensorFlow, Pandas, NumPy, Scikit-learn, Transformers and FastAI.
Network ChuckHOST
The list goes on.
Network ChuckHOST
You need lots of tools and resources in order to fine tune an LLM.
Nathan GoldbaumGUEST
It lets, it's how NumPy works.
Nathan GoldbaumGUEST
It's how Pandas, like basically every, almost every important project that you use that does heavy number crunching, it's talking to C Python over the C API or the C API.
Nathan GoldbaumGUEST
Yeah.
Nathan GoldbaumGUEST
And, and, They can do that through one of two ways.
35 MINS LATER
Christopher BaileyHOST
Yeah.
Nathan GoldbaumGUEST
And they still do, actually.
Nathan GoldbaumGUEST
One of the main purposes of the grant was to enable Pandas to switch to string D type, but then the timing worked out such that Arrow development came along It's been
Christopher BaileyHOST
confusing to me trying to even like parse it myself.
OliverHOST
By the end of these 15 minutes, four things will be true.
OliverHOST
You'll load a real CSV into a Pandas data frame.
OliverHOST
You'll inspect it properly, head, info, describe, value counts.
OliverHOST
You'll spot a real problem sitting in the file.
AnastasiaUNKNOWN
It eats whatever you feed it and hands you a number that looks official.
AnastasiaUNKNOWN
That's why we look.
AnastasiaUNKNOWN
Our tool for looking is a library called Pandas.
AnastasiaUNKNOWN
Pandas gives you a data frame.
Hood ChathamGUEST
Right.
Hood ChathamGUEST
You can load Pandas.
Hood ChathamGUEST
You can load, you know, whatever the Parquet package that is popular now is.
Hood ChathamGUEST
And, you know, I mean, just the packages, you know, if you start, you know, like, for instance, SciPy is just too big, too horrible, and it Doesn't really work to import sci fi.
Hood ChathamGUEST
I mean, sci fi when you when you bring it in, it has like over 100 different tier libraries, you know, um, yeah, sci fi has been like, like a real journey for us in piety.
Hood ChathamGUEST
Like, it's really like the most horrible package.
Hood ChathamGUEST
But But if it's not sci-fi, Pandas works and you can just load your Parquet file and show your graph as long as it's not too big.
Gyeongjae ChoiGUEST
Well, in my opinion, Pandas or DuckDB or other data processing libraries are not very optimized for Python workers.
Calvin Hendryx-ParkerHOST
And so, I mean, there's some major projects that are guilty of this.
Calvin Hendryx-ParkerHOST
I think the frustration and the ones that were listed were things like SQL Alchemy, Pandas, NumPy, and Django.
Calvin Hendryx-ParkerHOST
Yeah.
Calvin Hendryx-ParkerHOST
Those are big projects.
Michiko OkuboGUEST
Before the workflow begins, I load the library needed here.
Michiko OkuboGUEST
GeoPandas handles the map data.
Michiko OkuboGUEST
pyproj manages coordinate transformations between projections.
Michiko OkuboGUEST
Shapefree applies those transformations to the geometries, and Pandas helps manage the attribute data.
Christopher BaileyHOST
So pretty drastic change there.
Christopher BaileyHOST
For the Parquet file, though, pulling in only the columns that are needed there, Pollers is 6.6 milliseconds, DuckDB is really close, 6.9 milliseconds, and Pandas is four times slower, about 23.8 milliseconds.
Christopher BaileyHOST
CSV files are text.
Christopher BaileyHOST
Any values that are in there, they need to be converted into a number through parsing.
speaker_0HOST
Okay.
speaker_0HOST
So the source walks through using a Python library called Pandas Data Reader to gather this VIX data.
speaker_1HOST
Right.
speaker_1HOST
It's an API call.
speaker_0HOST
Doesn't that completely destroy the timeline of the market? Like in a system where momentum and the distance between events really matter, aren't we blinding the machine to the actual passage of time?
speaker_1HOST
It's a really valid concern, and it highlights a crucial vulnerability in proof processing.
speaker_1HOST
Now, when you use a Python library like Pandas and you invoke a function like DROPNA to delete those NANs.
speaker_1HOST
Right, DROPNA.
Artificial IntelligenceNARRATOR
Labels tell you what you found.
Artificial IntelligenceNARRATOR
Tested with Pandas 2.
Artificial IntelligenceNARRATOR
X. SciPy 1.
Artificial IntelligenceNARRATOR
X. Scikit Learn 1.
Samridhi VatsGUEST
So yeah, primarily SQL or Python if possible.
Samridhi VatsGUEST
your company prefers to do, like you need to know how to do data wrangling with Python, knowing NumPy, Pandas, at least all of that, not going into the advanced programming with Python, but definitely data wrangling using NumPy, Pandas, Matplotlib, but at least SQL, if not Python.
Samridhi VatsGUEST
I recognize that this was a mandatory language to know, be it you're interviewing for a data analyst, business analyst, or any role in the analytics field.
Samridhi VatsGUEST
So having identified that, I kind of upskilled myself or learned SQL writing queries, all the logics of SQLs, all the functions and all of that.
7 more episodes mention pandas.
Create an account to see the whole feed, search across every transcript, and follow the entities you care about.