I can’t find a consensus for what would be a good .rdata file equivalent. I like rdata files because they are very compact and I can throw multiple objects into them. Any suggestions?

Strictly speaking, there isn’t one. Julia’s type system is orders of magnitude more complex than R’s (which basically has about 6 vector types, the rest is done by metadata composed of the same). I would recommend JSON or HDF5 for most applications.

[image] Unable to save Unitful unit into a JLD file General Usage In case you are not awre the options are Base.Serialization, BSON.jl, JLSO.jl, JLD.jl and JLD2.jl For dataframes-like you have Feather.jl Parquet.jl, JDF.jl, HDF5.jl

SQLite which could store tabular data directly as tables, and serialized vars in a special vars table with blobs

For relatively short-lived data there is serialize/deserialize - I use them as a kind of a cache to save values that can easily be recomputed. Works reliably for complex types as well.

[image] dlakelan: SQLite but can you reliable load the data back in the julia type?

you can serialize the data and squish that into a blob column

Julia save ".rdata" equivalent

New to Julia

xiaodai June 14, 2020, 3:19am 3

2 Likes

Topic		Replies	Views
Future-proof way to save and load data General Usage	12	1179	May 9, 2020
[ANN] JDF.jl - Experimental Julia DataFrames serialization format Package Announcements	3	1457	January 19, 2020
How to easily save Julia data? General Usage question , serialization	5	681	February 16, 2025
What is the preferred way to save variables? General Usage jld , hdf5 , jld2	39	18904	August 24, 2021
Simple way to save several variables New to Julia question	10	2142	July 5, 2020

Julia save ".rdata" equivalent

Related topics