polars
db-benchmark
Our great sponsors
polars | db-benchmark | |
---|---|---|
144 | 91 | |
25,837 | 319 | |
5.3% | 0.9% | |
10.0 | 0.0 | |
5 days ago | 10 months ago | |
Rust | R | |
MIT License | Mozilla Public License 2.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
polars
-
Why Python's Integer Division Floors (2010)
This is because 0.1 is in actuality the floating point value value 0.1000000000000000055511151231257827021181583404541015625, and thus 1 divided by it is ever so slightly smaller than 10. Nevertheless, fpround(1 / fpround(1 / 10)) = 10 exactly.
I found out about this recently because in Polars I defined a // b for floats to be (a / b).floor(), which does return 10 for this computation. Since Python's correctly-rounded division is rather expensive, I chose to stick to this (more context: https://github.com/pola-rs/polars/issues/14596#issuecomment-...).
- Polars
-
Stuff I Learned during Hanukkah of Data 2023
That turned out to be related to pola-rs/polars#11912, and this linked comment provided a deceptively simple solution - use PARSE_DECLTYPES when creating the connection:
- Polars 0.20 Released
- Segunda linguagem
- Polars: Dataframes powered by a multithreaded query engine, written in Rust
- Summing columns in remote Parquet files using DuckDB
- Polars 0.34 is released. (A query engine focussing on DataFrame front ends)
db-benchmark
- Database-Like Ops Benchmark
-
Polars
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/
-
DuckDB performance improvements with the latest release
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.
-
Show HN: SimSIMD vs. SciPy: How AVX-512 and SVE make SIMD cleaner and ML faster
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"
-
Pandas vs. Julia โ cheat sheet and comparison
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.
-
Any faster Python alternatives?
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
-
Pandas 2.0 (with pyarrow) vs Pandas 1.3 - Performance comparison
The syntax has similarities with dplyr in terms of the way you chain operations, and itโs around an order of magnitude faster than pandas and dplyr (thereโs a nice benchmark here). Itโs also more memory-efficient and can handle larger-than-memory datasets via streaming if needed.
-
Pandas v2.0 Released
If interested in benchmarks comparing different dataframe implementations, here is one:
- Database-like ops benchmark
-
Python "programmers" when I show them how much faster their naive code runs when translated to C++ (this is a joke, I love python)
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.
What are some alternatives?
vaex - Out-of-Core hybrid Apache Arrow/NumPy DataFrame for Python, ML, visualization and exploration of big tabular data at a billion rows per second ๐
arrow-datafusion - Apache Arrow DataFusion SQL Query Engine
modin - Modin: Scale your Pandas workflows by changing a single line of code
Apache Arrow - Apache Arrow is a multi-language toolbox for accelerated data interchange and in-memory processing
databend - ๐๐ฎ๐๐ฎ, ๐๐ป๐ฎ๐น๐๐๐ถ๐ฐ๐ & ๐๐. Modern alternative to Snowflake. Cost-effective and simple for massive-scale analytics. https://databend.com
DataFrames.jl - In-memory tabular data in Julia
DataFramesMeta.jl - Metaprogramming tools for DataFrames
datatable - A Python package for manipulating 2-dimensional tabular data structures
sktime - A unified framework for machine learning with time series
arrow2 - Transmute-free Rust library to work with the Arrow format