There is a really cool recent blog post @ Julia Computing about how easily custom loss functions can be implemented in XGBoost w/ Zygote. They use the Titanic survival data from Kaggle & compare scores on the Leaderboard. I wasn’t able to replicate their results & there was no place to post a comme…

Here is how I automatically store all (49) learners from sk: using AutoMLPipeline sk= AutoMLPipeline.SKLearners.learner_dict; sk= keys(sk) |> collect |> x-> sort(x,lt=(x,y)->lowercase(x)<lowercase(y)) I realize some of these learners apply to regression others classification, but let’s put that as…

this is awesome, I’ll try to turn this into a tutorial when I have some time :slight_smile:

yeah. all the problems i’m working right now is for classification tasks. but it’s easy to add the regression metric for crossvalidation. you can look at the skcrossvalidator.jl source ( AutoMLPipeline.jl/skcrossvalidator.jl at master · IBM/AutoMLPipeline.jl · GitHub ). I’m using these metrics: 3.3. M…

Here’s what I had in mind applied to the Iris data: using AutoMLPipeline, DataFrames #Get models. sk= AutoMLPipeline.SKLearners.learner_dict |> keys |> collect; sk= sk |> x-> sort(x,lt=(x,y)->lowercase(x)<lowercase(y)); iris = AutoMLPipeline.Utils.getiris(); X = iris[:,1:4]; Y = iris[:,end] |> Vec…

thanks for the sample code! this is what i hope to achieve with automlpipeline, i.e., coding the feature extraction, feature engineering and modeling in a simple easy to understand manner such that it becomes easy to attack ML problems using elegant syntax borrowed from unix pipes. please feel fre…

using Distributed nprocs() == 1 && addprocs() @everywhere using AutoMLPipeline, DataFrames #Get models. sk= AutoMLPipeline.SKLearners.learner_dict |> keys |> collect; sk= sk |> x-> sort(x,lt=(x,y)->lowercase(x)<lowercase(y)); m_cl= sk[occursin.("Classifier", sk)]; m_cl= m_cl ∪ sk[occursin.("NB", sk…

Nice. Suppose I train/crossvalidate the models on a subset of rows train, is there an easy way to score the best cross-validated model on separate test rows that were not used for training/cross-validation?

crossvalidation purpose is to run the models with same set of parameters several times to get the average performance. each fold starts fresh with a new copy of the model using the same set of parameters. if for example among the models, SVC is the best, then you can use the fit! to train the SVC mo…

you can use BestLearner to automatically pick the best learner using internal cross-validation among a group of learners and uses the best learner output as it’s final output learners = SKLearner.(["AdaBoostClassifier","BaggingClassifier","SGDClassifier","SVC","LinearSVC"]) blearner = BestLearner(l…

I probably wasn’t being clear about my objective. Consider a Lasso model, w/ one hyper-parameter λ. Suppose I have a grid of possible values for λ: G=[0.0, 0.1, …, 1.0] (Sometimes sklearn includes a default grid, sometimes I make my own grid, sometimes I use another tuning method…) My objective:…

Custom XGBoost Loss function w/ Zygote. Julia Computing blog post

Specific Domains Machine Learning

ppalmes April 8, 2020, 5:54pm 33

please raise issue or PR in https://github.com/IBM/AutoMLPipeline.jl so that I can track and update the development there.

Topic		Replies	Views
Automate training MLJ models Machine Learning machine-learning , mlj	14	2362	February 17, 2020
[ANN] BetaML.jl.. yet an other (simple) Machine Learning Package Package Announcements package , announcement , machine-learning	16	3613	May 15, 2024
Problems with Lathe TrainTestSplit New to Julia	11	1718	September 18, 2020
AutoMLPipeline.jl makes it easy to create complexed ML pipeline structures Package Announcements machine-learning	23	2981	March 9, 2020
What is the relation between MLJ and Flux? Machine Learning	17	6905	February 10, 2021

Custom XGBoost Loss function w/ Zygote. Julia Computing blog post

Related topics