Cannot reach required slope

Course Q&A TensorFlow Developer Professional Natural Language Processing in TensorFlow
1

I tried everything, using bidirection LSTM/bidirectional GRU/conv1D with MaxPooling/AvgPoolin,g including dropout layer with 0.2 to 0.4, changing number of neuros in Dense Layer and so on, yet I cannot get to the required val_loss slope. Please help!

This is the architechture I use:

[code removed - moderator]

PLEASE HELP!

2

Please click my name and message the following:

  1. Text document containing model architecture and results of all your trials.
  2. Assignment notebook in ipynb format.
3

I took the model in your notebook and changed the dropout rate. The slope of your validation loss curve is -0.00026 was observed.

Please see this link to learn more about Dropout.

4

Yes, now I finally got a slope of 0.00012 after changing the dropout rate to 0.5 in the droupout layer.
Thank you very much for you help.