Skip to content

Fix log/symlog tick positions and compute linthresh from data#1

Closed
patrickoleary wants to merge 197 commits intomasterfrom
fix/linthresh
Closed

Fix log/symlog tick positions and compute linthresh from data#1
patrickoleary wants to merge 197 commits intomasterfrom
fix/linthresh

Conversation

@patrickoleary
Copy link
Copy Markdown
Owner

  • Compute linthresh from min abs non-zero value in data array (zero-copy, vectorized O(n))
  • Fix symlog function to standard formulation: sign(v) * log10(1 + |v|/linthresh)
  • Fix log tick positions to use log-normalized positions on linear colorbar
  • Fix symlog tick values to use powers of 10 matching LUT breakpoints
  • Capture linear colorbar image before any log/symlog LUT transforms
  • Pass linthresh once from data through all functions that need it

jourdain and others added 30 commits February 9, 2026 10:40
Automatically generated by python-semantic-release
Automatically generated by python-semantic-release
Automatically generated by python-semantic-release
Automatically generated by python-semantic-release
Automatically generated by python-semantic-release
huiwanpnnl and others added 29 commits April 23, 2026 16:16
- Add tooltips to file browser buttons
- Add slider/text-input toggle to DataSelection toolbar
- Style hotkey badges with rounded borders
- Add primary outline to active drawer toggle buttons
- Fix tooltips for color config card

Resolves Kitware#44
Default animation_track to None instead of "timestamps" to avoid
crashing at startup before any file is loaded. Also guard against
empty state values in the track change handler.
Dimensions that lack coordinate variables (e.g., ext_cnt,
num_gas_aerosol_constituents) now appear in the slice selection
UI with index-based sliders. Labels show the actual units when
a coordinate variable exists, or "Index value:" otherwise.
Replace flatten() with reshape(-1) to avoid copies where the data
is only read. In _load_variable, use a single copy followed by
in-place fill value replacement instead of flatten + np.where
which created three copies.
Track which dimensions changed in SetSlicing and only reload
variables whose dimensions intersect the changed set. Variables
unaffected by the slice change are kept from the previous output.

For example, changing lev now skips all 2D (time, ncol) variables.
With 200 2D + 20 3D variables loaded, lev changes are ~2.5x faster.
Only copy and scan for fill values when the variable actually has a
_FillValue attribute. Most variables default to NaN, making the
comparison a no-op that still scans the entire array. Variables
without fill values now return a zero-copy view from reshape.
Automatically generated by python-semantic-release
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

7 participants