calad0i
calad0i
A# Description In the current version of hls4ml, if one use the key "KerasH5" and not supplying the json model file to load the Keras model, the framework will try...
# Requires #973 and `python>=3.10` now. # Description This PR adds support for a special layer type `FixedPointQuantizer` used in [calad0i/HGQ/hls4ml-integration](https://github.com/calad0i/HGQ/tree/hls4ml-integration) to facilitate conversion from HGQ model to hls4ml model...
## Precision override moved to Requires #855 now. Expect tests to fail without. Include and superceeds #917. On the same issue: #947 This PR does the following, for the `vivado`...
A# Description No test at the moment - Fix a series of strange `accum_t` behaviors for pooling layers in the current version. Potential conflict with #855, would need to converge...
## Prerequisites Please make sure to check off these prerequisites before submitting a bug report. - [x] Test that the bug appears on the current version of the master branch....
# Description Implement manually unrolled dense/conv layers for latency models. Switch to enable this optimization is not implemented. Currently, by importing `import hls4ml.optimization.fused_dotp.optimizer_pass.vitis`, all `Dense`, `Conv1/2D` are unrolled for `vivado/vitis`...
A# Description With using the Quartus backend, with conv2d with kernel size 3x3 or conv1d with kernel size 3, the framework makes the default implementation `Winograd` instad of `im2col`. As...
# Description As discussed in one of the dev meetings, we should probably remove this because: - to avoid running the slow training in tests - this test is almost...