Dmitry Serykh
Dmitry Serykh
I don't have the time to test the ROC implementation right now. Have you tried using the branch? I am planning a big overhaul of the whole package in the...
Yes, it is possible. You should create an "image" with only 1 pixel by re-shaping your dataset: `image_data = 1d_data.reshape(-1, 1, 1)`
1) Correct 2) The BFASTMonitor algorithm only detects one breakpoint, if you wish to detect more breakpoints, try using the BFAST algorithm. An experimental prototype for BFAST (python backend only)...
x should be a vector of floating point time-steps, e.g. 1 January 2001 would be represented by 2001.0, while 31 December 2001 would be something like 2001.99. You can use...
I don't know if I understood your question completely, but this call: ``` arr = r_style_interval((2016, 1), (2016, 365), 365) ``` would return a vector of 365 values: ``` [2016....
Do you have NaNs in your dataset? Can you include the whole error message and the code snippet that you have ran?
Yes, the setup script in general is a bit of a mess. It needs to be looked at.
This is definitely not normal. The OpenCL backend should be much faster than the Python backend. What GPU are you using? What version of the code are you using?
I have just tried running the version from the master branch on my laptop and a GPU machine with NVIDIA A100. On both machines, the OpenCL version is at least...
It is probably the case. Furthermore, I don't know if the parallelization approach that you are using will work on multiple GPUs, since BFASTMonitor would always choose the OpenCL device...