WARN Keras Network Learner 0:14 /home/user/anaconda3/envs/py3_knime/lib/python3.6/site-packages/keras/models.py:282: UserWarning: No training configuration found in save file: the model was not compiled. Compile it manually.
ERROR Keras Network Learner 0:14 Execute failed: Missing cell in input row ‘Row3307750’, column ‘Price’.
WARN Keras Network Learner 0:14 /home/user/anaconda3/envs/py3_knime/lib/python3.6/site-packages/keras/models.py:282: UserWarning: No training configuration found in save file: the model was not compiled. Compile it manually.
ERROR Keras Network Learner 0:14 Execute failed: Missing cell in input row ‘Row2314736’, column ‘Price’.
WARN Keras Network Learner 0:14 /home/user/anaconda3/envs/py3_knime/lib/python3.6/site-packages/keras/models.py:282: UserWarning: No training configuration found in save file: the model was not compiled. Compile it manually.
ERROR Keras Network Learner 0:14 Execute failed: Missing cell in input row ‘Row2314736’, column ‘Price’.
This warning message can safely be ignored. It just tells you that Keras did not detect any training configuration inside the network you built, when loading it in the Keras Network Learner node. That is okay since you set a training configuration in the learner node anyway (the training configuration is, e.g., the optimizer and the loss function used during training).
This means that there is at least one missing value in the “Price” column of at least one of the input tables of the learner node. Since there is no general representation for – or handling of – missing values in deep learning, you need to handle them yourself before passing the tables to the learner node. You can do this by, e.g., inserting a Missing Value node between the File Reader and the Partitioning node. An easy strategy would then be to remove the entire row of each missing cell that is encountered (“Remove Row” option in the Missing Value node’s dialog). Alternatively, you can also replace missing values by valid ones using this node.
