Arthur comments

Results 7 comments of


                                            Arthur

keras tf layer Dot operation not supported?

Hi @green-cabbage, have you been able to successfully deploy your GCN model? I am also trying to implement a GCN model using hls4ml, but no success so far :/

Conv layer changes other layers to Resource unless specifically set

Hi, I have a quick question related to this. When I create my HLS model with `hls4ml.converters.keras_to_hls`, I get the following warning: `WARNING: Cannot use "Latency" model strategy for conv2d...

Conv layer changes other layers to Resource unless specifically set

Thanks for the reply @jmitrevs! The warning is gone. However, when compiling my HLS model I'm having weird errors that I've not seen before. ``` In file included from firmware/myproject.cpp:22:...

Conv layer changes other layers to Resource unless specifically set

Hi, I'm having the same issue with Dense layers. `WARNING: Cannot use "Latency" model strategy for dense1 layer. Switching to "Resource" strategy.` So this bug might not be limited to...

Multidimensional input support for Dense layer, Resource vs resource in Conv2D layer, lower reusefactor results in almost-0 resource usage

> Dense on 3D input is supported. Behind the scenes, it will result in exactly the same code as in your 2nd alternative, i.e., pointwise Conv2D. The reason you don't...

Multidimensional input support for Dense layer, Resource vs resource in Conv2D layer, lower reusefactor results in almost-0 resource usage

> You can see in the report exactly what is the DSP used for. When using the resource strategy, the weights will be stored in BRAM and there will be...

Multidimensional input support for Dense layer, Resource vs resource in Conv2D layer, lower reusefactor results in almost-0 resource usage

Hi, thank you so much for the insightful reply @vloncar ! However, by being creative with the supported Keras layers it seems like I will not be able to succesfully...