calad0i

Experiencing the same issue with 0.33. It seems that this bug is located somewhere in the resampling functions, as it is only triggered only when the resample=True flag is taking...

Currently 3.7.0.

`ERROR: [Board 49-71] The board_part definition was not found for tul.com.tw:pynq-z2:part0:1.0. The project's board_part property was not set, but the project's part property was set to xc7z020clg400-1. Valid board_part values...

pre-commit.ci autofix

Fixed vitis and mnist tests. Pytorch test directly crashed, need some debugging to see what is wrong.

`test/pytest/test_weight_writer.py` is not updated in this PR though... pre-commit.ci autofix

> The `test_qkeras.py::test_qgru[Vivado]` failure we can ignore. For cnn_mnist, why does this error cause the failure? We mentioned about possibly removing that test, but I want to understand why it's...

Confirmed. Will make a PR for this. More details: `replace_node` method in `ModelGraph` did not consider the case when a parent/children node has more than 1 output/inputs.

Vivado has an unroll limit of 4096 hardcoded that cannot be circumvented easily. The warning message (`Layer ... is too large (524288), are you sure you want to train?`) suggests...

It seems that you are using parallel io. In the newer version of hls4ml, conv unrolls are controlled by `PrallelizationFactor` in the config file (`hls_config`). (Dull unroll was done for...