Icon

kn_​deeplearn_​200_​h2o_​automl_​dl

use H2O.ai auto-machine-learning to just generate deep learning models

use H2O.ai auto-machine-learning to just generate deep learning models
to compare that to architectures you have created with Keras or Temsor Flow 2 in KNIME (cf. the other workflows in this collection)

please download the complete DeepLearning (Keras, Tensorflow, H2O.ai) Workflow group:
https://hub.knime.com/mlauber71/spaces/Public/latest/kn_example_deeplearning_keras_tensorflow_classification~G8jl-DTMCBqoxyB9/

use H2O.ai auto-machine-learning to just generate deep learning modelsto compare that to architectures you have created with Keras or Temsor Flow 2 in KNIME (cf. the other workflows in this collection)(video) - Automatic & Explainable Machine Learning with H2O: Erin LeDell, H2O.ai - "Deep Learning"https://www.youtube.com/watch?v=4fsmLRa1NqE&t=642please download the complete DeepLearning (Keras, Tensorflow, H2O.ai) Workflow group:https://hub.knime.com/mlauber71/spaces/Public/latest/kn_example_deeplearning_keras_tensorflow_classification~G8jl-DTMCBqoxyB9/ if the complicated Model Quality Metanode does not work you can stick to the"Binary Classification Inspector" node Propagate R environmentfor KNIME withMinicondaconfigure how to handle the environmentdefault = just check the names../data/deep_001_training.table../data/deep_005_test.table../data/deep_009_validate.tabletest and validation togetheredit: v_runtime_automlset the maximum runtime ofH2O.ai AutoML in SECONDSRead the MOJOmodelScore the test tableyou might also use a third table to validatethat has not been used developing themodelsolutionto string^(.*submission|solution).*$=> allow only deep learningmodels../model/adult_h2o_automl_deep_learning.zipbinary classification modelswith R. Results:/model/validation/number of Rows knime_r_environment Table Reader Table Reader Table Reader Concatenate Integer Input collect meta data Merge Variables H2O MOJO Reader H2O MOJO Predictor(Classification) Number To String Column Filter Column Rename H2O AutoML Learner H2O Local Context Table to H2O H2O Model to MOJO H2O MOJO Writer Model QualityClassification - Graphics Binary ClassificationInspector Extract TableDimension Table Columnto Variable use H2O.ai auto-machine-learning to just generate deep learning modelsto compare that to architectures you have created with Keras or Temsor Flow 2 in KNIME (cf. the other workflows in this collection)(video) - Automatic & Explainable Machine Learning with H2O: Erin LeDell, H2O.ai - "Deep Learning"https://www.youtube.com/watch?v=4fsmLRa1NqE&t=642please download the complete DeepLearning (Keras, Tensorflow, H2O.ai) Workflow group:https://hub.knime.com/mlauber71/spaces/Public/latest/kn_example_deeplearning_keras_tensorflow_classification~G8jl-DTMCBqoxyB9/ if the complicated Model Quality Metanode does not work you can stick to the"Binary Classification Inspector" node Propagate R environmentfor KNIME withMinicondaconfigure how to handle the environmentdefault = just check the names../data/deep_001_training.table../data/deep_005_test.table../data/deep_009_validate.tabletest and validation togetheredit: v_runtime_automlset the maximum runtime ofH2O.ai AutoML in SECONDSRead the MOJOmodelScore the test tableyou might also use a third table to validatethat has not been used developing themodelsolutionto string^(.*submission|solution).*$=> allow only deep learningmodels../model/adult_h2o_automl_deep_learning.zipbinary classification modelswith R. Results:/model/validation/number of Rowsknime_r_environment Table Reader Table Reader Table Reader Concatenate Integer Input collect meta data Merge Variables H2O MOJO Reader H2O MOJO Predictor(Classification) Number To String Column Filter Column Rename H2O AutoML Learner H2O Local Context Table to H2O H2O Model to MOJO H2O MOJO Writer Model QualityClassification - Graphics Binary ClassificationInspector Extract TableDimension Table Columnto Variable

Nodes

Extensions

Links