Icon

Multiclass Machine Learning - Wine Quality

<p>Score UCI Wine Quality Dataset - multiple Targets (multiclass) with H2O.ai nodes and other models - measure results with LogLoss<br>https://archive.ics.uci.edu/ml/datasets/wine+quality<br><br>In the sub-folder /data/ there is a Python JUpyter notebook "kn_example_ml_multiclass_wine_quality_h2o_gbm.ipynb" for a Hyper parameter Grid search for a GBM model</p>

URL: KNIME and ML - also Deep Learning - (44276) forum entry https://forum.knime.com/t/i-have-a-data-set-consisting-of-5-inputs-and-6-outputs-i-entered-it-in-machine-learning-it-gave-me-medium-accuracy-i-want-high-accuracy-can-you-suggest-the-algorithm-to-me/44276/2?u=mlauber71
URL: [multiclass] Score Documents with multiple Classes? https://forum.knime.com/t/urgent-what-is-wrong-with-my-decision-tree-predictor-for-new-data/13292/10?u=mlauber71
URL: [multiclass] Model for Multiclass Targets (and explanation of Log Loss statistics) (1) https://forum.knime.com/t/any-advice-to-improve-the-performance-of-a-classification-model/12801/10?u=mlauber71
URL: [multiclass] Model for Multiclass Targets (and explanation of Log Loss statistics) (2) https://forum.knime.com/t/metrics-in-multiclass-classification/11193/3?u=mlauber71
URL: H2O Multinomial Scorer - for LogLoss stats https://kni.me/n/E7aow8HbOeggW7MY
URL: collection of multiclass data sets https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/multiclass.html
URL: UCI - Wine Quality Data Set https://archive.ics.uci.edu/ml/datasets/wine+quality
URL: vtreat for KNIME! https://win-vector.com/2020/06/28/vtreat-for-knime/
URL: Medium: Data preparation for Machine Learning with KNIME and the Python “vtreat” package https://medium.com/p/efcaf58fa783

# Run AutoML for 60 seconds or
# 300 = 5 min, 600 = 10 min, 900 = 15 min, 1800 = 30 min, 3600 = 1 hour,
# 7200 = 2 hours
# 14400 = 4 hours
# 16200 = 4.5 hours
# 18000 = 5 Stunden
# 21600 = 6 hours
# 25200 = 7 hours
# 28800 = 8 hours
# 36000 = 10 hours
Score UCI Wine Quality Dataset - multiple Targets (multiclass) with H2O.ai nodes and other models - measure results with LogLoss

https://archive.ics.uci.edu/ml/datasets/wine+quality

https://forum.knime.com/t/any-advice-to-improve-the-performance-of-a-classification-model/12801/10?u=mlauber71
https://forum.knime.com/t/i-have-a-data-set-consisting-of-5-inputs-and-6-outputs-i-entered-it-in-machine-learning-it-gave-me-medium-accuracy-i-want-high-accuracy-can-you-suggest-the-algorithm-to-me/44276/2?u=mlauber71

In the sub-folder /data/notebook/ there is a Python Jupyter notebook "kn_example_ml_multiclass_wine_quality_h2o_gbm.ipynb" for a Hyper parameter Grid search for a GBM model. Also a notebook for GLM grid search

prepare the data with the help of the "vtreat" package

Python Conda environment propagation. Please read this article for more details:


KNIME and Python — Setting up and managing Conda environments

https://medium.com/p/2ac217792539

WEKA_ADA
WEKA_LOGIT
collect resultsof severalmulticlassmodels
Concatenate
H2O_DEEP_LEARNING
H2O Predictor (Classification)
H2O_DEEP_LEARNING
H2O Multinomial Scorer
LogLossH2O_AutoML=> only GBM and Stacked Models4 folds (might change that)
H2O AutoML Learner
LogLossH2O_DEEP_LEARNING=> only deep learning Models4 folds (might change that)
H2O AutoML Learner
H2O_DEEP_LEARNING
Constant Value Column (deprecated)
GLM
Constant Value Column (deprecated)
h2o_automl_deep_learning.zip
Model Writer
GLM
H2O Multinomial Scorer
h2o_automl_deep_learning.zip
Model Reader
h2o_glm.zip
Model Reader
RowID
GLM
H2O Predictor (Classification)
LogLossASCENDING=> lowest value -> best model
Sorter
h2o_glm.zip
Model Writer
model_results.xlsx
Excel Writer
80
H2O to Table
XGBOOST
XGBoost Tree Ensemble Learner
XGBOOST
XGBoost Predictor
20
H2O to Table
GLM
H2O Generalized Linear Model Learner
xgboost.zip
Model Reader
H2O_AutoML
H2O Predictor (Classification)
bring resultsback to H2OXGBOOST
Table to H2O
https://archive.ics.uci.edu/ml/datasets/wine+quality
Prepare UCI Wine Quality
GBM
H2O Gradient Boosting Machine Learner
xgboost.zip
Model Writer
XGBOOST
Constant Value Column (deprecated)
dataset_multiclass_20.parquet=> to be used in a Jupyter notebookto develop a H2O.ai model
Parquet Writer
XGBOOST
H2O Multinomial Scorer
dataset_multiclass.parquethttps://archive.ics.uci.edu/ml/datasets/wine+quality
Parquet Reader
SMO
SMO (3.7)
dataset_multiclass_80.parquet
Parquet Writer
WEKA_SMO
Weka Predictor (3.7)
NB
H2O Naive Bayes Learner
GBM
H2O Predictor (Classification)
80 vtreat
Table to H2O
GBM
H2O Multinomial Scorer
H2O_AutoML
H2O Multinomial Scorer
h2o_automl_vtreat.zip
Model Reader
GBM
Constant Value Column (deprecated)
LogLossH2O_AutoML=> only GBM and Stacked Models4 folds (might change that)
H2O AutoML Learner
weka_smo.zip
Weka Classifier Reader (3.7)
H2O_AutoML_vtreat
H2O Multinomial Scorer
H2O_AutoML
Constant Value Column (deprecated)
weka_smo.zip
Weka Classifier Writer (3.7)
h2o_automl_vtreat.zip
Model Writer
H2O_AutoML_vtreat
Constant Value Column (deprecated)
H2O_AutoML_vtreat
H2O Predictor (Classification)
WEKA_SMO
H2O Multinomial Scorer
20
Table to H2O
bring resultsback to H2OWEKA_SMO
Table to H2O
WEKA_SMO
Constant Value Column (deprecated)
h2o_gbm.zip
Model Writer
h2o_gbm.zip
Model Reader
h2o_automl.zip
Model Writer
"Class"as the multi-class target80 vtreatvtreat for KNIME!https://win-vector.com/2020/06/28/vtreat-for-knime/
Python Script
h2o_gbm.zip
Model Reader
Table to H2O
NB
Constant Value Column (deprecated)
h2o_automl_vtreat_treatment.zip
Model Reader
H2O Local Context
NB
H2O Multinomial Scorer
"Class"as the multi-class target20 vtreatAPPLY
Python Script
NB
H2O Predictor (Classification)
h2o_automl_vtreat_treatment.zip
Model Writer
h2o_nb.zip
Model Writer
80/20split training/test
H2O Partitioning
h2o_nb.zip
Model Reader
Activate Conda Environmentbased on Operating SystemWindows or macOS
conda_environment_kaggle
edit: v_runtime_automlset the maximum runtime ofH2O.ai AutoML in SECONDS
Integer Input (legacy)

Nodes

Extensions

Links