Leo Auri
Leo Auri
I can't compare with CUDA right now, no device here, but for me `.to(device)` lands on `mps`: ``` from kan import * device = 'mps' model = KAN(width=[1,1]) model.to(device) [(n,...
Yeah, I did `pip install -e .` on `0c79f78`.
+1 Any updates on releasing training code? Would be helpful for related research.. Reproducibility 😅
I tried adding these lines back into the `CMakeLists.txt` manually. The build then failed with the weird error: ``` ** BUILD FAILED ** The following build commands failed: PhaseScriptExecution Generate\...
* cmake version 3.20.5 * FRUT 43e9f045 * JUCE 0c56fa8fa
Which commits of FRUT & JUCE have you successfully compiled with?
The instructions don't say what to do with the downloaded traced models, perhaps that is missing?
I found a hardcoded path to traced model in `micro-tcn/realtime/plugin/Source/PluginProcessor.cpp` ``` model = torch::jit::load(".../models/traced_1-uTCN-300__causal__4-10-13__fraction-0.01-bs32.pt"); ``` which I replaced, but same error...
Absolutely, possibly link it to the locale. Very disconcerting if you don't think dates in that direction