python vowpalwabbit regression for big data files with FEARTURES INTERACTION

python vowpalwabbit regression for big data files with FEARTURES INTERACTION



1 ridge
2 lasso
3 quantile for both ridge and lasso !

FEATURES BOTH CATEGORICAL AND CONTINUES

IMPORTANT to HAVE INTERACTION BETWEEN CATEGORICAL FEARTURES : SECOND ORDER AND THIRED ORDER

like
from vowpalwabbit.sklearn_vw import VW, VWClassifier, VWRegressor
vw_squared = VWRegressor(loss_function='squared' , normalized = True, interactions = 'abc')

but better to use
from vowpalwabbit import pyvw

for big data like
us-used-cars-dataset 9 GB 3ml rows 66 features predict price
https://www.kaggle.com/ananaymital/us-used-cars-dataset

but start you can from
https://www.kaggle.com/prakharrathi25/banking-dataset-marketing-targets


all calculations done in vowpalwabbit python including one hot for categorical data (not scikit learn one hot)

data has both categorical and continues features



code starter
https://github.com/matanbruker/mechine-learning-project/blob/1583c3790fe9d09608e8805e573b6493bc7bcbc1/mechine%20learning%20project/Notebooks/analcatdata_neavote.ipynb



https://medium.com/open-machine-learning-course/open-machine-learning-course-topic-8-vowpal-wabbit-fast-learning-with-gigabytes-of-data-60f750086237

https://reposhub.com/python/deep-learning/rambler-digital-solutions-criteo-1tb-benchmark.html#local-training---vowpal-wabbit--xgboost


https://github.com/szilard/benchm-ml


http://fastml.com/go-non-linear-with-vowpal-wabbit/

https://www.kite.com/python/answers/how-to-read-a-large-csv-file-in-chunks-with-pandas-in-python#:~:text=Use%20chunksize%20to%20read%20a,be%20read%20in%20per%20chunk.



https://rdrr.io/rforge/RVowpalWabbit/man/vw.html
https://search.r-project.org/CRAN/refmans/RVowpalWabbit/html/vw.html
https://github.com/peterhurford/vowpal_platypus
https://www.kaggle.com/kashnitsky/topic-8-online-learning-and-vowpal-wabbit
https://bigdatapath.wordpress.com/big-data-analytics-online-learning/

vw train_titanic.vw -f model.vw –binary –passes 20 -c -q ff –sgd –l1
0.00000001 –l2 0.0000001 –learning_rate 0.5 –loss_function logistic

https://ebookreading.net/view/book/EB9781785887215_28.html

test3 <- c("-t", system.file("test", "train-sets", "0002.dat", package="RVowpalWabbit"),
"-f", file.path(tempdir(), "0002.model"),
"--cache_file", file.path(tempdir(), "0002.cache"))


also https://ebookreading.net/view/book/EB9781785887215_28.html many example for VW

maybe >>> from vowpalwabbit.DFtoVW import DFtoVW
>>> import pandas as pd
>>> df = pd.DataFrame({"y": [1], "x": [2]})
>>> conv = DFtoVW.from_colnames(y="y", x="x", df=df)
>>> conv.convert_df()
['1 | x:2']