The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Modeling >> Behavioral Models >> Machine-Learning-based-Behavioral Modelling
https://designers-guide.org/forum/YaBB.pl?num=1618839212

Message started by rhanna on Apr 19^th, 2021, 6:33am

Title: Machine-Learning-based-Behavioral Modelling
Post by rhanna on Apr 19^th, 2021, 6:33am

Hello,

I need your help in debugging/tips to debug a Verilog-A module that I created based on the training of a Recurrent Neural Network (RNN).

The model is supposed to solve the non-linear differential system of equations coming from the RNN representation.

I will summarize everything in steps so that it will be easier to discuss it afterward.

1. I created a testbench for a chain of inverters and got the data (Vin,Iout) in .csv format after interpolation to a timestep=1e-9 s.

Attached testbench: https://drive.google.com/file/d/1n5uyi9_nTurEUgIai_yb0RWwVqGcYBIx/view?usp=sharing
and sample of Iout: https://drive.google.com/file/d/1APHkSRnFl5gxAFEDcvrW81Ygpvv7yMO5/view?usp=sharing
and training_data.csv: https://drive.google.com/file/d/1iKuxfxAz4b-mrt8UOs0gPqNqQfSwCnre/view?usp=sharing

2.Train/test datasets are pre-processed (scaled/normalized/standardized ...) to fit the RNN model used.
After tuning model parameters, it’s tested against unseen stimuli.

Attached learning curves: https://drive.google.com/file/d/1iF3D_tp4-F8p6B3sVirlBFtKVEoFMdUk/view?usp=sharing
RNN prediction: https://drive.google.com/file/d/1ufzQZsS19DvamE_oCMx9N6YZoM43tYTD/view?usp=sharing
and parameters matrices: https://drive.google.com/file/d/1eXxqaCrzyDqMTLY8N0_7Mp8-7QM1GkNP/view?usp=sharing

3. The parameter arrays are flattened to be used as 1-D vectors in the Verilog-A module (since multi-D is not allowed in Verilog-A)

Below, you will find a listing of my code which is basically using Spectre to solve the continuous non-linear system of equations of the RNN model.

code: https://drive.google.com/file/d/1zbWbgu4UApWolGYfJDxRuryDGtLgmFV1/view?usp=sharing

*** attached summary for the equations and the equivalent circuit of the model.

https://drive.google.com/file/d/1oHXT9a7a_cxw1xfquPQ-Aiv-q2njpI7E/view?usp=sharing

4. results after compiling and running the simulation look like that:

https://drive.google.com/file/d/1wm4wRaL_BK_j0iIwpdDx5-WbwmspQ7h4/view?usp=sharing

which is not really good! I changed alpha to consider different simulation steps but this is the best result.

Sorry for the long post and the many links and thanks in advance to your time.