Want 4 times faster numpy.full and numpy.full_like?
Performance gains on a 12-core Ryzen 5900:
| Size | Threads | np.full |
parallel_full |
Speedup |
|---|---|---|---|---|
| 0.25 MB | 1 | 0.01 ms | 0.01 ms | 1.10x |
| 1 MB | 1 | 0.02 ms | 0.02 ms | 0.94x |
| 4 MB | 1 | 0.47 ms | 0.47 ms | 0.99x |
| 8 MB | 4 | 0.91 ms | 0.57 ms | 1.60x |
A practitioner's guide based on experience setting up and operating sensor development and characterisation labs at CERN, LBNL, DESY, and the University of Bergen.
Author: Magne Lauritzen
Contact: magne.lauritzen@summacogni.com
The majority of this guide applies to any lab doing experimental physics, electronics development, or instrumentation work. Sensor-specific sections are clearly marked.
| { | |
| "title": "Unicode Keypad Processor", | |
| "rules": [ | |
| { | |
| "description": "Switch to Unicode Hex Input on Option hold", | |
| "manipulators": [ | |
| { | |
| "type": "basic", | |
| "from": { | |
| "key_code": "left_option" |
| { | |
| "title": "Unicode Keypad Processor", | |
| "rules": [ | |
| { | |
| "description": "Unicode hex input with Option key", | |
| "manipulators": [ | |
| // Handle Left Option key - convert and type on release | |
| { | |
| "type": "basic", | |
| "from": { |
| """ | |
| This Gist illustrates how you can import cupy when it may or not be installed. Also works with frozen applications. | |
| If cupy is not installed, or the CUDA_PATH system variable is not set, or the incorrect CUDA version is installed | |
| (relevant for frozen applications), is_cupy_available = False. | |
| """ | |
| import warnings | |
| is_cupy_available = False |
I hereby claim:
To claim this, I am signing this object: