Inference a Custom Model in PYNQ-Z2

[Ba1tu3han]

Hello All,

I have trained the CNV network with a custom dataset by Brevitas. Then, I compiled it with FINN. I have the deploy folder that contains .bit, .hwh and driver files. I follow https://github.com/Xilinx/finn/blob/main/notebooks/end2end_example/bnn-pynq/cnv_end2end_example.ipynb and https://github.com/Xilinx/finn/blob/main/notebooks/end2end_example/cybersecurity/3-build-accelerator-with-finn.ipynb notebooks to inference in Pynq-Z2 board.

I could not load and give it as input correctly my validation dataset or test images in Pynq-Z2. I read the driver.py and validate.py source files however I could not understand requirements of input images. Could you help me about just inference with my custom model? Additionaly, I am open to any guideline about how to use the driver.

Bests,
Batuhan