Siril/Documentation - English @ Siril translation

Component	Unfinished	Unfinished words	Unfinished characters	Untranslated	Checks	Suggestions	Comments
Documentation-1.2 GFDL-1.3	0	0	0	0	519	0	0

Documentation-1.4 GFDL-1.3	0	0	0	0	1,033	0	0

Documentation-1.6 GFDL-1.3	0	0	0	0	528	0	0

Overview

Project website	siril.org

17 hours ago

String statistics

	Strings percent	Hosted strings	Words percent	Hosted words	Characters percent	Hosted characters
Total		16,263		316,951		2,010,380
Translated	100%	16,263	100%	316,951	100%	2,010,380
Needs editing	0%	0	0%	0	0%	0
Read-only	100%	16,263	100%	316,951	100%	2,010,380
Failing checks	12%	2,080	6%	19,480	7%	142,355
Strings with suggestions	0%	0	0%	0	0%	0
Untranslated strings	0%	0	0%	0	0%	0

Last 12 months Previous

Activity in last 12 months


Jul	Aug	Sep	Oct	Nov	Dec	Jan	Feb	Mar	Apr	May	Jun

Quick numbers

Hosted words

Hosted strings

Translated

Contributors

and previous 30 days

Trends of last 30 days

—

Hosted words

—

Hosted strings

—

Translated

—

Contributors

—

None Resource updated Siril Documentation Documentation-1.6 English	The “`po/siril-documentation.pot`” file was changed. 2 weeks ago
None Source string added Siril Documentation Documentation-1.6 English	English 0 characters edited Read only Yes, we know it's a shame. In fact we would have liked to use GTK as the toolkit for writing sirilpy scripts, but unfortunately the binary wheel has been broken on Windows for a long time and doesn't show any sign of being fixed. We can't release core functionality that doesn't work on one of the target OSes. And if we could add ``gi`` we would get tickets asking for ``pyqt``, and then other packages and before you know it Siril would be a multi-gigabyte download essentially the same size as an entire OS. So we have settled on TKinter. Yes, it's a bit plain but it's simple to use, it's supported everywhere and since it's a part of Python's standard library the risk of problems with it is very low. 2 weeks ago
None Source string added Siril Documentation Documentation-1.6 English	English 0 characters edited Read only Dependencies Note that both torch and jax use GPU library pypi packages, so installation of them may pull in very large chains of dependencies - the full set of NVIDIA packages (CUDA, CuDNN etc.) is around 1GB. So when using these in scripts be sure to warn users that they may incur a large and potentially slow download on first use! 2 weeks ago
None Source string added Siril Documentation Documentation-1.6 English	English 0 characters edited Read only jax - Not a dedicated ML framework as such but a means of offloading computation to the GPU. jax provides a mostly numpy-compatible module called jax.numpy which can greatly accelerate many mathematical array processing operations by offloading them to the GPU and also provides functions for gradient computation that numpy does not. Bearing in mind that the slow part of GPU processing is the data transfer between system memory and GPU memory, jax provides a jit compiler to precompile its offloaded workloads but it still requires careful algorithm design to keep data on the GPU for as long as possible, only returning the final result. Jax has good GPU support on Linux but only CPU support on Apple and Windows, although experimental Metal support and Windows GPU support is in progress. This is therefore not recommended for actual scripts at present, but is highlighted as an interesting framework to develop against in the (hopefully) near future when its support broadens. A ``JaxHelper()`` class is under development that will help with selection of the correct jax package to install and testing to confirm it works properly. 2 weeks ago
None Source string added Siril Documentation Documentation-1.6 English	English 0 characters edited Read only torch - This is a heavyweight among ML frameworks and where GPU support was once very much CUDA-based it is now broadening its coverage: suupport for NVidia and Apple Silicon GPUs is good, and support for AMD GPUs (using ROCm) and Intel GPUs is in development (ROCm is stable enough to use on Linux, not yet on Windows). Unfortunately torch has some dependency issues (it is excessively strict about the version of CuDNN it requires) which causes problems using it in the same venv as some other modules (including jax - see below) that pull in the nvidia-cudnn package. A ``TorchHelper()`` class is under development that will help with selection of the correct torch package to install and testing to confirm it works properly. 2 weeks ago
None Source string added Siril Documentation Documentation-1.6 English	English 0 characters edited Read only onnxruntime - used in GraXpert and Riccardo's AberrationRemover.py script, this package is available for all the target OSes. There are different backends to support different GPUs, but in some cases these are fragile. A helper class ``ONNXHelper()`` is available to help with identification and installation of the optimum package as well as ensuring correct runtime fallback to CPU. Note that currently on Windows the ONNXHelper will select the DirectML runtime for all supported GPUs. Even though in theory the CUDA-based onnxruntime-gpu package might be a bit faster, it unfortunately depends heavily on system libraries being installed and configured correctly and the installation is considered too fragile to be reliable in the kind of automated python environment used by Siril. If users wish to use it and are confident they can get it to work on their system they can install it manually using ``sirilpy.ensure_installed("onnxruntime-gpu") but this is not supported and related issues will be closed. 2 weeks ago
None Source string added Siril Documentation Documentation-1.6 English	English 0 characters edited Read only A variety of machine learning modules exist which can be used to write "AI" scripts. However the state of support for heterogenous computing (i.e. GPU use) is patchy and often requires system libraries such as CUDA or ROCm to be installed. The following modules are regarded as key for GPU acceleration and either are supported with helper classes or have helper classes in development. 2 weeks ago
None Source string added Siril Documentation Documentation-1.6 English	English 0 characters edited Read only Machine Learning Modules 2 weeks ago
None Source string added Siril Documentation Documentation-1.6 English	English 0 characters edited Read only In filefilter specifications ``tkfilebrowser`` requires multiple extensions to be provided separated by \| whereas filedialog requires space-separated extensions: ``sirilpy.tkfilebrowser`` has been adapted to accept either format. 2 weeks ago
None Source string added Siril Documentation Documentation-1.6 English	English 0 characters edited Read only ``sirilpy.tkfilebrowser.askdirectory()`` has been added as an alias for ``tkfilebrowser.askopendirname()`` to maintain compatibility with ``tk.filedialog.askdirectory()``. 2 weeks ago

Browse all changes for this language

Search

Advanced query builder

Query examples

Review strings changed by other users	`changed:>="1 month ago" AND NOT changed_by:anonymous`	Add
Translated strings	`state:>=translated`	Add
Strings with comments	`has:comment`	Add
Strings with any failing checks	`has:check`	Add
Strings with suggestions from others	`has:suggestion AND NOT suggestion_author:anonymous`	Add
Approved strings with suggestions	`state:approved AND has:suggestion`	Add
All untranslated strings added the past month	`added:>="1 month ago" AND state:<=needs-editing`	Add
Strings changed in the past two weeks	`changed:>="2 weeks ago"`	Add

Shortcut	Action
`?`	Open available keyboard shortcuts.
`Alt` + `Home`	Navigate to the first translation in the current search.
`Alt` + `End`	Navigate to the last translation in the current search.
`Alt` + `PageUp` or `Ctrl` + `↑` or `Alt` + `↑` or `Cmd` + `↑` or	Navigate to the last translation in the current search.
`Alt` + `PageDown` or `Ctrl` + `↓` or `Alt` + `↓` or `Cmd` + `↓` or	Navigate to the next translation in the current search.
`Ctrl` + `Enter` or `Cmd` + `Enter`	Submit current form; this works the same as pressing Save and continue while editing translation.
`Ctrl` + `Shift` + `Enter` or `Cmd` + `Shift` +`Enter`	Unmark translation as Needing edit and submit it.
`Alt` + `Enter` or `Option` + `Enter`	Submit the string as a suggestion; this works the same as pressing Suggest while editing translation.
`Ctrl` + `E` or `Cmd` + `E`	Focus on translation editor.
`Ctrl` + `U` or `Cmd` + `U`	Focus on comment editor.
`Ctrl` + `M` or `Cmd` + `M`	Shows Automatic suggestions tab.
`Ctrl` + `1` to `Ctrl` + `9` or `Cmd` + `1` to `Cmd` + `9`	Copies placeable of a given number from source string.
`Ctrl` + `M` followed by `1` to `9` or `Cmd` + `M` followed by `1` to `9`	Copy the machine translation of a given number to current translation.
`Ctrl` + `I` followed by `1` to `9` or `Cmd` + `I` followed by `1` to `9`	Ignore one item in the list of failing checks.
`Ctrl` + `J` or `Cmd` + `J`	Shows the Nearby strings tab.
`Ctrl` + `S` or `Cmd` + `S`	Focus on search field.
`Ctrl` + `O` or `Cmd` + `O`	Copy the source string.
`Ctrl` + `Y` or `Cmd` + `Y`	Toggle the Needs editing checkbox.
`→`	Browse the next translation string.
`←`	Browse the previous translation string.