antoninkriz / timeseriesclassification.jl Goto Github PK
View Code? Open in Web Editor NEWTime Series Classification algorithms in Julia - Bachelor thesis project
License: MIT License
Time Series Classification algorithms in Julia - Bachelor thesis project
License: MIT License
In examples/knndtw.jl the line
ucr_archive_datasets = DataSets.list_available_datasets(UCRArchive)
is leading to this exception:
failed to find end of centeral directory record
error(::String)@error.jl:35
_find_sigoffset(::IOStream, ::UInt32)@ZipFile.jl:288
_find_enddiroffset@ZipFile.jl:293[inlined]
ZipFile.Reader(::IOStream, ::Bool)@ZipFile.jl:106
Reader@ZipFile.jl:120[inlined]
unzip(::String, ::String)@utils.jl:25
download_datasets(::Nothing, ::Nothing, ::Bool)@UCRArchive.jl:222
var"#load_dataset_metadata#5"(::Bool, ::typeof(TimeSeriesClassification.DataSets._Loader.load_dataset_metadata), ::Type{TimeSeriesClassification.DataSets.Loaders._UCRArchiveLoader.UCRArchive}, ::Symbol, ::Nothing, ::Nothing)@UCRArchive.jl:284
load_dataset_metadata@UCRArchive.jl:277[inlined]
load_dataset_metadata(::Type{TimeSeriesClassification.DataSets.Loaders._UCRArchiveLoader.UCRArchive}, ::Symbol)@UCRArchive.jl:277
top-level scope@[Local: 1](http://localhost:1234/edit?id=906ce0f6-5bfd-11ee-1049-9184a6a60f1c#)[inlined]
I'm staring to look over the MLJ interface. A fundamental issue is the current form of input data, which I understand should be in the form (X_train, :column_based)
or (X_train, :row_based)
where X
is what exactly?
Passing the row/column flag in this way is quite non-standard for MLJ models. In any case, it means the currently declared input_scitype
does not match the data requirements.
You can pass metadata like this flag as a third argument, as in machine(model, X, y, flag)
but before going that route, can you say more about what data you can accept for X
and why you need this distinction? Can we instead deal with row-versus-column issue by having the user provide a lazy transpose
, if her data does not conform to your preferred format?
It doesn't look like one needs all of MLJ for testing. I think MLJBase should suffice.
If you are using any measures (metrics) from MLJ, then now that this has just merged, you will additionally need to add StatisticalMeasures.jl as a test dependency and add using StatisticalMeasures
to your tests.
This issue is used to trigger TagBot; feel free to unsubscribe.
If you haven't already, you should update your TagBot.yml
to include issue comment triggers.
Please see this post on Discourse for instructions and more details.
If you'd like for me to do this for you, comment TagBot fix
on this issue.
I'll open a PR within a few hours, please be patient!
Part of the output from ?KNNDTWModel
:
Hyperparameters
=================
K=1: Number of neighbors weights=:uniform: Either :uniform or :distance based weights of
the neighbors. - :uniform: All neighbors are weighted equally. - :distance: Each
neighbor is weighted by it's distance. distance=DTW(): DTW distance struct, for example
DTWSakoeChiba or pure DTW. - DTW(): Dynamic Time Warping without any constraints. -
DTWSakoeChiba(): Dynamic Time Warping with Sakoe Chiba bound constraint. - DTWItakura():
Dynamic Time Warping without Itakura Parallelogram constraint. - You can provide your
own metric by subtyping DTWType. bounding=LBNone(): Lower bounding of the distance using
methods like LBKeogh(). - LBNone(): NO-OP, no lower bouning is being done. - LBKeogh():
Estimating distance lower bound of the distance using the LB_Keogh method
(https://www.cs.ucr.edu/~eamonn/LB_Keogh.htm). - You can provide your own methofs by
subtyping LBType.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.