Browser-native OMX / HDF5 matrix explorer, now inside VS Code. Open .omx, .h5, or .hdf5 files directly in the editor — no server, no upload, no external viewer. Everything runs locally via WebAssembly inside the extension's webview.

Also available as a standalone web app: omx-viewer on GitHub Pages.

Features

Native custom editor — right-click any .omx/.h5/.hdf5 file and choose Open With → OMX Viewer, or just open it directly; it's the default editor for these extensions
Virtualized matrix grid — smooth scrolling on matrices up to 10,000 × 10,000, using chunked/lazy loading so the full matrix is never held in memory at once
Matrix picker — every matrix in the file is available from a dropdown, top-right of the editor
Aggregation summary — SUM / MIN / MAX / MEAN / MEDIAN / STD DEV / COUNT NON-ZERO, by row or by column, across the active matrix or all matrices at once
CSV export — export a summary table via the toolbar button or the OMX: Export Summary as CSV command (Ctrl+Shift+P / Cmd+Shift+P), which saves through VS Code's native save dialog
Matrix arithmetic — element-wise add/subtract/multiply/divide between two matrices in the file, opened as a temporary result tab
Cell inspector — click a cell to pin it; see its value across every matrix in the file at the same [row, col] coordinate
Native theming — colors, spacing, and icons follow your active VS Code theme automatically; no separate light/dark toggle to manage

Requirements

None — no external binaries, no Python, no network access. The HDF5 reader (h5wasm) is bundled as WebAssembly inside the extension.

Known Limitations

Read-only: this is a viewer, not an editor for OMX file contents
Recommended up to ~700 MB per file; very large files may be limited by available memory
Element-wise arithmetic loads both operand matrices into memory at once (~2× matrix size)

About OMX

OMX (Open Matrix Format) is an HDF5-based convention for storing one or more named matrices plus optional zone lookup tables, widely used in transportation demand modeling.