Repository navigation
Releases
Releases · hiddenstateresearch/visualtorch
v1.4.1
v1.4.0
What's Changed
- feat: add animated GIF rendering for graph_view, flow_view, and lenet_view by @willyfh in #167
- docs: simplify docs homepage title by @willyfh in #169
- feat: deprecate graph_view/flow_view/lenet_view in favor of render() by @willyfh in #171
- Add FTIR CNN-KAN paper to Research Showcase by @Tsubashimo-Nanato in #166
- chore: bump pre-commit hook versions (pre-commit-hooks, mypy, markdownlint, ruff) by @pre-commit-ci[bot] in #168
- docs: demonstrate level_gap for nested skip connections by @danhdoan in #173
- docs: demonstrate animation timing options by @cnYui in #174
- feat: replace placeholder fire-emoji icon with an original logo by @willyfh in #177
- docs: pin README logo image to its actual commit SHA by @willyfh in #178
- fix: use a clean, tightly-cropped logo with real transparency by @willyfh in #179
- fix: README logo alignment, docs sidebar blur, and SHA pin by @willyfh in #180
- fix: dark-mode wordmark, docs sidebar sizing, and README cleanup by @willyfh in #181
- fix: use scoped PAT for star history API auth by @willyfh in #182
- fix: use a square favicon instead of stretching the tall flame icon by @willyfh in #183
- Add legends to graph and LeNet views by @Tsubashimo-Nanato in #175
- Build a robust client-neutral VisualTorch MCP by @Tariolle in #176
- fix: add legend option to MCP manifest for graph/lenet styles by @willyfh in #184
- Add legend positioning to graph and LeNet views by @mgalore in #186
- [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci[bot] in #188
- docs: clarify Becoming a Maintainer with explicit path subsections by @willyfh in #189
- Bump version to v1.4.0 by @willyfh in #190
New Contributors
- @pre-commit-ci[bot] made their first contribution in #168
- @danhdoan made their first contribution in #173
- @cnYui made their first contribution in #174
- @mgalore made their first contribution in #186
Full Changelog: v1.3.0...v1.4.0
v1.3.0
What's Changed
- Add GOVERNANCE.md and CODEOWNERS, rename LICENSE to LICENSE.md by @willyfh in #111
- Ask first-time contributors to claim one issue at a time by @willyfh in #116
- Smooth star history curve with quarterly-resampled interpolation by @willyfh in #118
- Enable Dependabot and pre-commit.ci auto-fix by @willyfh in #120
- Bump actions/setup-python from 2 to 6 by @dependabot[bot] in #121
- Bump actions/checkout from 2 to 7 by @dependabot[bot] in #122
- Update aggdraw requirement from >=1.3.11 to >=1.3.19 by @dependabot[bot] in #125
- Update setuptools requirement from >=42 to >=83.0.0 by @dependabot[bot] in #126
- Remove Dependabot version updates, keep security alerts only by @willyfh in #128
- feat: add connector_fill and connector_width to flow_view and lenet_view (#115) by @Maqbool61 in #119
- Invite people to share their work by @willyfh in #131
- Add test coverage for the visualtorch.layered module import path by @willyfh in #130
- feat: add type_ignore support to graph_view (#117) by @Maqbool61 in #132
- Invite third-party contributions to the Research Showcase by @willyfh in #134
- Support integer/bool input_dtype for models starting with nn.Embedding by @willyfh in #137
- Add configurable flow legend position by @Tsubashimo-Nanato in #133
- docs: add draw_funnel examples by @iamdhrv in #144
- Add maintainer promotion/demotion rules to GOVERNANCE.md by @willyfh in #145
- feat(style): add outline_width parameter to flow, lenet, and graph views (#112) by @Maqbool61 in #143
- Add custom font examples for all styles by @Tsubashimo-Nanato in #146
- docs: add dedicated examples for connector_fill and connector_width by @turab-usmani in #147
- Rename lenet connector-style example to match naming convention by @willyfh in #150
- Auto-assign PRs to their author when opened by @willyfh in #152
- fix: apply connector style to regular funnels by @w3lld1 in #151
- Fix silently dropped edges when a model's final op is an unconsumed merge by @willyfh in #155
- Attribute ongoing copyright to VisualTorch Contributors by @willyfh in #157
- Document release process in README and add CI coverage for previous PyTorch release by @willyfh in #158
- feat: add optional VisualTorch MCP integration by @Tariolle in #148
- Mention pull requests as a valid maintainer-nomination venue by @willyfh in #161
- docs(examples): add show_neurons=False example for graph_view by @minagayid in #142
- Switch maintainer nomination to lazy consensus by @willyfh in #162
- Add Tariolle as a maintainer, CODEOWNER for visualtorch_mcp by @willyfh in #159
- docs: demonstrate max_channels in lenet view by @w3lld1 in #156
- Add optional graph connector arrowheads by @hasanfaesal in #160
- docs: fix show_arrows docstring position to match signature order by @willyfh in #163
- chore: bump version to 1.3.0 by @willyfh in #164
New Contributors
- @Maqbool61 made their first contribution in #119
- @Tsubashimo-Nanato made their first contribution in #133
- @iamdhrv made their first contribution in #144
- @turab-usmani made their first contribution in #147
- @w3lld1 made their first contribution in #151
- @Tariolle made their first contribution in #148
- @minagayid made their first contribution in #142
- @hasanfaesal made their first contribution in #160
Full Changelog: v1.2.1...v1.3.0
v1.2.1
What's Changed
- Add deprecation shims for renamed API instead of hard-removing it (#109): restores backward compatibility for
layered_view,InputDummyLayer, andone_dim_orientation, all of which previously hit hard errors instead of a deprecation warning.
Full Changelog: v1.2.0...v1.2.1
v1.2.0
What's Changed
- Make Research Showcase paper links clickable by @willyfh in #96
- Fix typos in CONTRIBUTING.md and document adding gallery examples by @willyfh in #97
- Add higher-resolution examples for graph, flow, and lenet styles by @willyfh in #98
- Show every output tensor's shape, not just the first by @willyfh in #99
- Fix Research Showcase page title duplicating its own sidebar caption by @willyfh in #100
- Credit torchview and refine attribution/citation wording by @willyfh in #101
- Add named color palette presets (palette= option) by @willyfh in #102
- Backfill missing example topics across graph/flow/lenet_style by @willyfh in #103
- Fix low-dim shape depth mapping, rename one_dim_orientation by @willyfh in #104
- Bump version to 1.2.0, fix README banner on PyPI by @willyfh in #105
Full Changelog: v1.1.0...v1.2.0
v1.1.0
What's Changed
Breaking
InputDummyLayerrenamed toInput. If you customized the synthetic input box's color viacolor_map[visualtorch.InputDummyLayer], update it tocolor_map[visualtorch.Input].
New features
show_inputoption ongraph_view/flow_view/lenet_view(defaultTrue) - a single-input model's input box is now shown by default, fixing lopsided-looking diagrams for autoencoders, U-Net, and similar encoder/decoder architectures.- New gallery examples: ResNet18 (via torchvision), U-Net, and Vision Transformer, each in both
graphandflowstyle. - Added a Research Showcase page and
CITATION.cfffor GitHub's native "Cite this repository" button.
Fixes
- Skip-connection detour routing is now collision-aware, only detouring around boxes it would actually visually cross (#93).
- Fixed skip-connector nesting order so a wider, enclosing skip connection (e.g. U-Net's outer skip) routes farther out than a narrower nested one, instead of the reverse.
- Fixed the 3D-style skip connector's attachment point so it no longer visually cuts through a box's own fill.
Full Changelog: v1.0.0...v1.1.0
v1.0.0
VisualTorch 1.0.0
This is a major release: the API has been substantially reworked since v0.2.5. 1.0.0 marks the render() API as the stable public surface going forward.
Breaking changes
- Replaced three separate functions (
graph_view,layered_view,lenet_view) with a single entry point:visualtorch.render(model, input_shape, style="graph"|"flow"|"lenet", **kwargs). Per-style kwargs are validated via typed dataclasses (GraphStyleOptions/FlowStyleOptions/LenetStyleOptions/CommonOptions) - a typo'd kwarg now raisesTypeErrorinstead of being silently ignored. - Renamed style
"layered"to"flow"everywhere. - Dropped
index_ignore(position-based layer hiding) - no unambiguous meaning under the new topology-based layout.type_ignore(hide by class) still works.
New capabilities
- Multi-input model support: models whose
forward()takes multiple separate input tensors are now supported - pass a tuple of per-tensor shapes asinput_shape, e.g.((1, 3, 224, 224), (1, 10)). - Unified backend: all three styles now share one real topology tracer -
flow/lenetgained genuine branching and skip-connection support they never had before, andgraphdropped its old hardcodedLinear/Conv2d-only restriction. - Skip/residual connections now route as a visible detour line above the diagram in every style, instead of drawing invisible/collinear.
nn.MultiheadAttention(and anyTransformerEncoderLayer/TransformerDecoderLayer) no longer vanishes from the diagram.- RNN/LSTM/GRU output shapes no longer get misread as CNN channel counts.
show_dimensionsupport added across all three styles (including a new toggle forlenet, to avoid label overlap on parallel branches).legendsupport added tograph_view.- Fixed crashes on multi-input shapes (now a clear
ValueError) and on non-Sequential/ModuleListcontainer modules (e.g. timm'sFeatureListNet).
Visual / quality
- Default color palette switched to Okabe-Ito (colorblind-safe) across all three styles.
- Fixed gallery example thumbnails being blindly stretched into a fixed canvas (blur) - now sized proportionally to their native resolution.
- New examples: residual block, multi-branch (Inception-style) merge, multi-input (siamese-style) model, dark background usage - each available in all three styles.
- Refreshed the README and readthedocs homepage banners.
Upgrading
If you depend on the pre-1.0 API, pin to the last compatible release: pip install visualtorch==0.2.5.
Full changelog: v0.2.5...v1.0.0
v0.2.5
What's Changed
- Bump version by @willyfh in #66
- Improve BibTeX formatting by @willyfh in #67
- Add regression tests for issues #63, #68, #69 by @willyfh in #70
- Fix shape-handling crashes for multi-output containers and multi-input shapes by @willyfh in #71
- Expand test coverage for previously untested parameters by @willyfh in #72
Full Changelog: v0.2.4...v0.2.5
v0.2.4
v0.2.3
What's Changed
- Bump version by @willyfh in #41
- Create CODE_OF_CONDUCT.md by @willyfh in #42
- Create Contributing Guidelines by @willyfh in #43
- Add link to the contributing guideline by @willyfh in #44
- Remove unnecessary print by @willyfh in #45
- Fix link to the license by @willyfh in #46
- Fix link to the license (#46) by @willyfh in #47
- Add email contact address in Code of Conduct by @willyfh in #50
- Update documentation based on the bullet points of JOSS review checklist by @willyfh in #52
- Add copyrights by @willyfh in #51
- Merge from main to joss-review branch by @willyfh in #53
- Merge JOSS paper review branch by @willyfh in #54
- Update LICENSE by @willyfh in #55
- Merge from Joss-review (#56) by @willyfh in #57
- Add ORCID and update some sentences by @willyfh and @paulgavrikov in #58
Full Changelog: v.0.2.2...v0.2.3