pyhton 3.8
ipykernel==6.9.1 pandas==1.4.1 scikit-learn==1.0.2 scipy==1.8.0 numpy==1.22.2 wxPython==4.1.1 matplotlib==3.5.1
Only needed to run TMTT test code
tensorflow==2.8.0
Clone the repository or download the source code.
Create a virtual environment, activate it and install the dependencies. For a faster wxPyhton installation you can check their wheels
$ python3.8 -m venv <path to your venv>
$ . <path to your venv>/bin/activate
<your venv>$ python -m pip install --upgrade pip
<your venv>$ pip install -r requirements.txtTensorflow is needed to RUN ANNmodels, but it is not required to run the tool.
The data available at the moment are:
- spiral octogonal transformer in a 65nm - 1:1, 2:1, 1:2 and 1:1 with primary overlapping secondary (_balun).
- spiral octogonal inductor in 350nm - 1, 2, 3, 4, 5 turns
Before creating the models, the data must be preprocessed. This is done using the data_prepare_tmtt_transf.ipynb and data_prepare_tmtt_indRG.ipynb notebooks.
Once the data is reorganized, the models cab be created as indicated in the notebook. Notebooks model_tmtt* make comparisson between modelings apreaoches and strategies.
Once the model is done execute PACOSYT.
<your venv>$ python src/pacosyt.py- Include data-prepare + model generation in GUI
- Formalize testing and CI
- Pypi packaging
F. Passos et al., "PACOSYT: A Passive Component Synthesis Tool Based on Machine Learning and Tailored Modeling Strategies Towards Optimal RF and mm-Wave Circuit Designs," in IEEE Journal of Microwaves, doi: 10.1109/JMW.2023.3237260.
This research is funded by the European Union�s Horizon 2020 research and innovation program under the MSCA grant agreement No. 892431 and also by the Instituto de Telecomunicações internal research projects LAY(RF)2 (X-0002-LX-20) and HAICAS (X-0009-LX-20). This work was supported by grant PID2019-103869RB-C31 funded by MCIN/AEI/10.13039/ 501100011033.