Icon

05_​Deploy_​Trained_​DL_​Model_​solution

05 Deploy Trained DL Model

This workflow is part of a collection of exercise/solution materials used at a hands on workshop held at German Conference for Bioinformatics (GCB-2020). The title of the workshop is "Binding Preference Prediction using KNIME Analytics Platform and its Keras Deep Learning Integration"

Exercise 05: Deploy Trained Deep Learning Model Activity I: Save Trained Model This time we start from part of the workflow from the previous exercise, where the model training is already done. 1. Add a Keras Network Writer node and connect it to the executed Keras Network Learner node and save the trained network to disk for future use. Use filepath "knime://knime.workflow/../../data/Keras_Trained_Model.h5" in the Keras Network Writer node configuration 2. Execute the node to write the trained model to disk. Activity II: Use Deployed Model Now we can use the saved model. 1. Use the provided Keras Network Reader node to read back the saved model. Use the same filepath as above. 2. Select one of the test FASTA files under data/iDeepS_PARCLIP/test/ using the provided Simple File Reader node. 3. Execute Keras Network Executor node to see the predicted results. You can repeat the process (step 4 & 5) for different FASTA files. The data used in this workflow are from the following publication:Xiaoyong Pan, Peter Rijnbeek, Junchi Yan, Hong-Bin Shen. Prediction of RNA-protein sequence and structure binding preferences using deep convolutional and recurrent neuralnetworks. BMC Genomics, 2018, 19:511.Specifically: https://github.com/xypan1232/iDeepS/tree/master/datasets/clip Load NW architectureRMSProp30,000 vs10,000 Load DataNode 359Node 369Load Trained ModelKeras NetworkReader Simple File Reader Keras NetworkLearner Partitioning Table Reader Keras NetworkExecutor Keras NetworkWriter Keras NetworkReader Process Fasta File Exercise 05: Deploy Trained Deep Learning Model Activity I: Save Trained Model This time we start from part of the workflow from the previous exercise, where the model training is already done. 1. Add a Keras Network Writer node and connect it to the executed Keras Network Learner node and save the trained network to disk for future use. Use filepath "knime://knime.workflow/../../data/Keras_Trained_Model.h5" in the Keras Network Writer node configuration 2. Execute the node to write the trained model to disk. Activity II: Use Deployed Model Now we can use the saved model. 1. Use the provided Keras Network Reader node to read back the saved model. Use the same filepath as above. 2. Select one of the test FASTA files under data/iDeepS_PARCLIP/test/ using the provided Simple File Reader node. 3. Execute Keras Network Executor node to see the predicted results. You can repeat the process (step 4 & 5) for different FASTA files. The data used in this workflow are from the following publication:Xiaoyong Pan, Peter Rijnbeek, Junchi Yan, Hong-Bin Shen. Prediction of RNA-protein sequence and structure binding preferences using deep convolutional and recurrent neuralnetworks. BMC Genomics, 2018, 19:511.Specifically: https://github.com/xypan1232/iDeepS/tree/master/datasets/clip Load NW architectureRMSProp30,000 vs10,000 Load DataNode 359Node 369Load Trained ModelKeras NetworkReader Simple File Reader Keras NetworkLearner Partitioning Table Reader Keras NetworkExecutor Keras NetworkWriter Keras NetworkReader Process Fasta File

Nodes

Extensions

Links