db-benchmark VS r4ds

Real-world performance is complicated since data science covers a lot of use cases.

If you're just reading a small CSV to do analysis on it, then there will be no human-perceptible difference between Polars and Pandas. If you're reading a larger CSV with 100k rows, there still won't be much of a perceptible difference.

Per this (old) benchmark, there are differences once you get into 500MB+ territory: https://h2oai.github.io/db-benchmark/

I do think it was important for duckdb to put out a new version of the results as the earlier version of that benchmark [1] went dormant with a very old version of duckdb with very bad performance, especially against polars.

[1] https://h2oai.github.io/db-benchmark/

https://news.ycombinator.com/item?id=33270638 :

> Apache Ballista and Polars do Apache Arrow and SIMD.

> The Polars homepage links to the "Database-like ops benchmark" of {Polars, data.table, DataFrames.jl, ClickHouse, cuDF, spark, (py)datatable, dplyr, pandas, dask, Arrow, DuckDB, Modin,} but not yet PostgresML? https://h2oai.github.io/db-benchmark/ *

LLM -> Vector database: https://en.wikipedia.org/wiki/Vector_database

/? inurl:awesome site:github.com "vector database"

I agree with your conclusion but want to add that switching from Julia may not make sense either.

According to these benchmarks: https://h2oai.github.io/db-benchmark/, DF.jl is the fastest library for some things, data.table for others, polars for others. Which is fastest depends on the query and whether it takes advantage of the features/properties of each.

For what it's worth, data.table is my favourite to use and I believe it has the nicest ergonomics of the three I spoke about.

Same. Numba does wonders for me in most scenarios. Yesterday I've discovered pola-rs and looks like I will add it to the stack. It's API is similar to pandas. Have a look at the benchmarks of cuDF, spark, dask, pandas compared to it: Benchmarks

The syntax has similarities with dplyr in terms of the way you chain operations, and it’s around an order of magnitude faster than pandas and dplyr (there’s a nice benchmark here). It’s also more memory-efficient and can handle larger-than-memory datasets via streaming if needed.

If interested in benchmarks comparing different dataframe implementations, here is one:

https://h2oai.github.io/db-benchmark/

Bad examples. Both numpy and pandas are notoriously un-optimized packages, losing handily to pretty much all their competitors (R, Julia, kdb+, vaex, polars). See https://h2oai.github.io/db-benchmark/ for a partial comparison.

https://r4ds.hadley.nz is free and very good

R for Data Science teaches you the tidyverse packages, which makes data wrangling so much easier!

One of the best free resources is the R4DS book by Hadley Wickham. You should make sure you start with the in progress second edition. https://r4ds.hadley.nz/

The first edition of R4DS is quite old now. Check out the soon to be released second edition: https://r4ds.hadley.nz/

R for Data Science (2nd Ed), the updated guide from Hadley Wickham

The big skills gap you have is in practical data exploration and transformation, which will be a large part of any data-centric role. As much as people may have distaste for it, there is no avoiding data manipulation as critical foundational enabler of all inferential and predictive modeling work. SQL is the lingua franca here and well worth picking up the basics (joins, window functions, handling dates and times, etc.), plus learning how to implement similar transformations in R and Python. With appropriately transformed data, you then need to be able to visualize it effectively using tools like Tableau or ggplot2 in R. I would not necessarily seek courses or certificates in it but expect to be evaluated on them in technical interview screenings, so self-study accordingly. R for Data Science by Hadley Wickham is a great free resource for these topics for R.

I just looked and there is now a second edition! https://r4ds.hadley.nz/