diff --git a/LICENSE.md b/LICENSE.md
new file mode 100644
index 0000000000000000000000000000000000000000..21fb194ddd89c0b6beb9beefefaf728332808c3b
--- /dev/null
+++ b/LICENSE.md
@@ -0,0 +1,12 @@
+# Licences
+
+The folders in this repository hold different models under different
+licences. Each folder carries its own licence and notice files, which must
+stay with its ONNX files when they are redistributed.
+
+| folder | models | licence | files |
+| --- | --- | --- | --- |
+| `tabicl/` | TabICLv2 (Soda team, Inria) | BSD-3-Clause | `tabicl/LICENSE` |
+| `kumo/` | Kumo Tabular small, medium and large (NVIDIA) | OpenMDW-1.1 (weights); Apache-2.0 (structured-data-models code); BSD-3-Clause (TabICLv2 code notice) | `kumo/LICENSE`, `kumo/NOTICE`, `kumo/LICENSE.structured-data-models` |
+| `rfm/` | Kumo Relational (NVIDIA), fine-tuned from TabICLv2 | OpenMDW-1.1 (weights); BSD-3-Clause (TabICLv2 notice); Apache-2.0 (structured-data-models code) | `rfm/LICENSE`, `rfm/THIRD-PARTY-NOTICES.txt`, `rfm/LICENSE.structured-data-models` |
+| `tsfm/` | Chronos-2 and Chronos-2 small (Amazon), Kumo-Anomaly (NVIDIA) | Apache-2.0; MIT (DPM-Solver notice) | `tsfm/LICENSE`, `tsfm/NOTICE`, `tsfm/THIRD-PARTY-NOTICES.txt`, `tsfm/LICENSE.structured-data-models` |
diff --git a/README.md b/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..a75d9fefcd62161d11002280199fc17b0eca8713
--- /dev/null
+++ b/README.md
@@ -0,0 +1,105 @@
+---
+license: other
+license_name: bsd-3-clause-openmdw-1.1-apache-2.0
+license_link: LICENSE.md
+library_name: onnx
+base_model:
+ - jingang/TabICL
+ - nvidia/Kumo-Tabular
+ - nvidia/Kumo-Relational
+ - amazon/chronos-2
+ - autogluon/chronos-2-small
+ - nvidia/Kumo-Anomaly
+base_model_relation: quantized
+tags:
+ - onnx
+ - onnxruntime-web
+ - webgpu
+ - tabular-classification
+ - tabular-regression
+ - time-series-forecasting
+ - anomaly-detection
+ - tabular-foundation-model
+ - relational-foundation-model
+ - in-context-learning
+---
+
+# ONNX models for Tabular Data Workbench
+
+Unofficial ONNX conversions of six open foundation models, made for Tabular
+Data Workbench, a browser app that runs them with onnxruntime-web on WebGPU
+or WebAssembly. The weights are the upstream authors': the `fp32` files hold
+them unchanged, and the `f16w` files store the larger ones as float16. The
+upstream authors have not reviewed or endorsed these conversions.
+
+| folder | models | upstream | licence | size (fp32 + f16w) |
+| --- | --- | --- | --- | --- |
+| [`tabicl/`](tabicl/README.md) | TabICLv2 classifier and regressor | [jingang/TabICL](https://huggingface.co/jingang/TabICL) (Soda team, Inria) | BSD-3-Clause | 339 MB |
+| [`kumo/`](kumo/README.md) | Kumo Tabular small, medium and large, classifier and regressor | [nvidia/Kumo-Tabular](https://huggingface.co/nvidia/Kumo-Tabular) (NVIDIA) | OpenMDW-1.1 | 1.2 GB (f16w only) |
+| [`rfm/`](rfm/README.md) | Kumo Relational classifier and regressor | [nvidia/Kumo-Relational](https://huggingface.co/nvidia/Kumo-Relational) (NVIDIA) | OpenMDW-1.1 | 367 MB |
+| [`tsfm/`](tsfm/README.md) | Chronos-2, Chronos-2 small, Kumo-Anomaly | [amazon/chronos-2](https://huggingface.co/amazon/chronos-2), [autogluon/chronos-2-small](https://huggingface.co/autogluon/chronos-2-small) (Amazon), [nvidia/Kumo-Anomaly](https://huggingface.co/nvidia/Kumo-Anomaly) (NVIDIA) | Apache-2.0 | 901 MB |
+
+Each folder holds a `manifest.json` (every file's size and sha256, the
+graphs' inputs and outputs, model settings and parity figures), the licence
+and notice files for its models, and each graph in two variants. `fp32` has
+the weights as released. `f16w` stores every float32 weight of 1,024 or more
+elements as float16 and casts it back to float32 on load, which halves the
+download and keeps compute in float32. `kumo/` has the `f16w` variant only.
+The README in each folder describes its graphs, how they were made and how
+closely they match the reference implementation.
+
+The graphs are the networks only. Preprocessing, ensembling, sampling and
+decoding stay outside them, as in the reference packages, and the app
+implements them in TypeScript. The `models/` folder of Tabular Data Workbench
+has the export scripts and the numpy references for these steps.
+
+## Use in the app
+
+The app reads `//manifest.json` and the files it lists, and
+checks each file's sha256 before caching it. To serve the models from this
+repository, build the app with the base URL pinned to a commit:
+
+```sh
+VITE_MODELS_BASE_URL=https://huggingface.co/ScriptSmith/tabular-data-workbench-onnx/resolve// pnpm build
+```
+
+## How they were made
+
+The scripts are in the `models/` folder of Tabular Data Workbench, one uv
+workspace. They use PyTorch 2.14 on CPU and the `torch.onnx` dynamo exporter
+at opset 18 (opset 20 for `kumo/`), and check the graphs with onnxruntime
+1.30:
+
+```sh
+# from the root of the Tabular Data Workbench source
+(cd models/tabicl && uv run export.py)
+(cd models/kumo && uv run export.py)
+node models/rfm/dump_shop.mjs # once: the sample tables rfm's checks run on
+(cd models/rfm && uv run export.py)
+(cd models/tsfm && uv run export_chronos2.py && uv run export_kumo.py)
+```
+
+Every export checks its graphs against the reference package (tabicl,
+structured-data-models, chronos-forecasting) before it writes the manifest.
+The exports also remove the exporter's node metadata and doc strings, which
+hold stack traces with the paths of the machine that ran them.
+
+## Licences
+
+The folders are under different licences, listed in [`LICENSE.md`](LICENSE.md):
+
+- `tabicl/`: BSD-3-Clause, Copyright (c) 2025, Soda team @ Inria (`tabicl/LICENSE`).
+- `kumo/`: OpenMDW-1.1, Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES
+ (`kumo/LICENSE`), with a notice of origin and changes (`kumo/NOTICE`, which
+ also carries the BSD-3-Clause notice for the TabICLv2-derived code) and the
+ Apache-2.0 licence of structured-data-models (`kumo/LICENSE.structured-data-models`).
+- `rfm/`: OpenMDW-1.1, Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES
+ (`rfm/LICENSE`), with NVIDIA's notice for the TabICLv2 weights it was
+ fine-tuned from (`rfm/THIRD-PARTY-NOTICES.txt`, BSD-3-Clause) and the
+ Apache-2.0 licence of structured-data-models (`rfm/LICENSE.structured-data-models`).
+- `tsfm/`: Apache-2.0 (`tsfm/LICENSE`), with Amazon's and NVIDIA's notices
+ and a statement of the changes (`tsfm/NOTICE`), and DPM-Solver's MIT
+ licence (`tsfm/THIRD-PARTY-NOTICES.txt`).
+
+Keep each folder's licence and notice files with its ONNX files when
+redistributing them.
diff --git a/kumo/LICENSE b/kumo/LICENSE
new file mode 100644
index 0000000000000000000000000000000000000000..161a8f01c0e68899bf4175a5c641c5fa0e9b11db
--- /dev/null
+++ b/kumo/LICENSE
@@ -0,0 +1,52 @@
+Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+
+
+OpenMDW License Agreement, version 1.1 (OpenMDW-1.1)
+
+By exercising rights granted to you under this agreement, you accept and agree
+to its terms.
+
+As used in this agreement, "Model Materials" means the materials provided to
+you under this agreement, consisting of: (1) one or more machine learning
+models (including architecture and parameters); and (2) all related artifacts
+(including associated data, documentation and software) that are provided to
+you hereunder.
+
+Subject to your compliance with this agreement, permission is hereby granted,
+free of charge, to deal in the Model Materials without restriction, including
+under all copyright, patent, database, and trade secret rights included or
+embodied therein.
+
+If you distribute any portion of the Model Materials, you shall retain in your
+distribution (1) a copy of this agreement, and (2) all copyright notices and
+other notices of origin included in the Model Materials that are applicable to
+your distribution.
+
+If you file, maintain, or voluntarily participate in a lawsuit against any
+person or entity asserting that the Model Materials directly or indirectly
+infringe any patent or copyright, then all rights and grants made to you
+hereunder are terminated, unless that lawsuit was in response to a
+corresponding lawsuit first brought against you.
+
+This agreement does not impose any restrictions or obligations with respect to
+any use, modification, or sharing of any outputs generated by using the Model
+Materials.
+
+THE MODEL MATERIALS ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
+OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE, TITLE, NONINFRINGEMENT, ACCURACY, OR THE
+ABSENCE OF LATENT OR OTHER DEFECTS OR ERRORS, WHETHER OR NOT DISCOVERABLE, ALL
+TO THE GREATEST EXTENT PERMISSIBLE UNDER APPLICABLE LAW.
+
+YOU ARE SOLELY RESPONSIBLE FOR (1) CLEARING RIGHTS OF OTHER PERSONS THAT MAY
+APPLY TO THE MODEL MATERIALS OR ANY USE THEREOF, INCLUDING WITHOUT LIMITATION
+ANY PERSON'S COPYRIGHTS OR OTHER RIGHTS INCLUDED OR EMBODIED IN THE MODEL
+MATERIALS; (2) OBTAINING ANY NECESSARY CONSENTS, PERMISSIONS OR OTHER RIGHTS
+REQUIRED FOR ANY USE OF THE MODEL MATERIALS; OR (3) PERFORMING ANY DUE
+DILIGENCE OR UNDERTAKING ANY OTHER INVESTIGATIONS INTO THE MODEL MATERIALS OR
+ANYTHING INCORPORATED OR EMBODIED THEREIN.
+
+IN NO EVENT SHALL THE PROVIDERS OF THE MODEL MATERIALS BE LIABLE FOR ANY CLAIM,
+DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
+OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE MODEL MATERIALS, THE
+USE THEREOF OR OTHER DEALINGS THEREIN.
\ No newline at end of file
diff --git a/kumo/LICENSE.structured-data-models b/kumo/LICENSE.structured-data-models
new file mode 100644
index 0000000000000000000000000000000000000000..001a3281169d6ade30cd1a44544029254ba490a8
--- /dev/null
+++ b/kumo/LICENSE.structured-data-models
@@ -0,0 +1,204 @@
+Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+
+
+ Apache License
+ Version 2.0, January 2004
+ http://www.apache.org/licenses/
+
+ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+ 1. Definitions.
+
+ "License" shall mean the terms and conditions for use, reproduction,
+ and distribution as defined by Sections 1 through 9 of this document.
+
+ "Licensor" shall mean the copyright owner or entity authorized by
+ the copyright owner that is granting the License.
+
+ "Legal Entity" shall mean the union of the acting entity and all
+ other entities that control, are controlled by, or are under common
+ control with that entity. For the purposes of this definition,
+ "control" means (i) the power, direct or indirect, to cause the
+ direction or management of such entity, whether by contract or
+ otherwise, or (ii) ownership of fifty percent (50%) or more of the
+ outstanding shares, or (iii) beneficial ownership of such entity.
+
+ "You" (or "Your") shall mean an individual or Legal Entity
+ exercising permissions granted by this License.
+
+ "Source" form shall mean the preferred form for making modifications,
+ including but not limited to software source code, documentation
+ source, and configuration files.
+
+ "Object" form shall mean any form resulting from mechanical
+ transformation or translation of a Source form, including but
+ not limited to compiled object code, generated documentation,
+ and conversions to other media types.
+
+ "Work" shall mean the work of authorship, whether in Source or
+ Object form, made available under the License, as indicated by a
+ copyright notice that is included in or attached to the work
+ (an example is provided in the Appendix below).
+
+ "Derivative Works" shall mean any work, whether in Source or Object
+ form, that is based on (or derived from) the Work and for which the
+ editorial revisions, annotations, elaborations, or other modifications
+ represent, as a whole, an original work of authorship. For the purposes
+ of this License, Derivative Works shall not include works that remain
+ separable from, or merely link (or bind by name) to the interfaces of,
+ the Work and Derivative Works thereof.
+
+ "Contribution" shall mean any work of authorship, including
+ the original version of the Work and any modifications or additions
+ to that Work or Derivative Works thereof, that is intentionally
+ submitted to Licensor for inclusion in the Work by the copyright owner
+ or by an individual or Legal Entity authorized to submit on behalf of
+ the copyright owner. For the purposes of this definition, "submitted"
+ means any form of electronic, verbal, or written communication sent
+ to the Licensor or its representatives, including but not limited to
+ communication on electronic mailing lists, source code control systems,
+ and issue tracking systems that are managed by, or on behalf of, the
+ Licensor for the purpose of discussing and improving the Work, but
+ excluding communication that is conspicuously marked or otherwise
+ designated in writing by the copyright owner as "Not a Contribution."
+
+ "Contributor" shall mean Licensor and any individual or Legal Entity
+ on behalf of whom a Contribution has been received by Licensor and
+ subsequently incorporated within the Work.
+
+ 2. Grant of Copyright License. Subject to the terms and conditions of
+ this License, each Contributor hereby grants to You a perpetual,
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+ copyright license to reproduce, prepare Derivative Works of,
+ publicly display, publicly perform, sublicense, and distribute the
+ Work and such Derivative Works in Source or Object form.
+
+ 3. Grant of Patent License. Subject to the terms and conditions of
+ this License, each Contributor hereby grants to You a perpetual,
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+ (except as stated in this section) patent license to make, have made,
+ use, offer to sell, sell, import, and otherwise transfer the Work,
+ where such license applies only to those patent claims licensable
+ by such Contributor that are necessarily infringed by their
+ Contribution(s) alone or by combination of their Contribution(s)
+ with the Work to which such Contribution(s) was submitted. If You
+ institute patent litigation against any entity (including a
+ cross-claim or counterclaim in a lawsuit) alleging that the Work
+ or a Contribution incorporated within the Work constitutes direct
+ or contributory patent infringement, then any patent licenses
+ granted to You under this License for that Work shall terminate
+ as of the date such litigation is filed.
+
+ 4. Redistribution. You may reproduce and distribute copies of the
+ Work or Derivative Works thereof in any medium, with or without
+ modifications, and in Source or Object form, provided that You
+ meet the following conditions:
+
+ (a) You must give any other recipients of the Work or
+ Derivative Works a copy of this License; and
+
+ (b) You must cause any modified files to carry prominent notices
+ stating that You changed the files; and
+
+ (c) You must retain, in the Source form of any Derivative Works
+ that You distribute, all copyright, patent, trademark, and
+ attribution notices from the Source form of the Work,
+ excluding those notices that do not pertain to any part of
+ the Derivative Works; and
+
+ (d) If the Work includes a "NOTICE" text file as part of its
+ distribution, then any Derivative Works that You distribute must
+ include a readable copy of the attribution notices contained
+ within such NOTICE file, excluding those notices that do not
+ pertain to any part of the Derivative Works, in at least one
+ of the following places: within a NOTICE text file distributed
+ as part of the Derivative Works; within the Source form or
+ documentation, if provided along with the Derivative Works; or,
+ within a display generated by the Derivative Works, if and
+ wherever such third-party notices normally appear. The contents
+ of the NOTICE file are for informational purposes only and
+ do not modify the License. You may add Your own attribution
+ notices within Derivative Works that You distribute, alongside
+ or as an addendum to the NOTICE text from the Work, provided
+ that such additional attribution notices cannot be construed
+ as modifying the License.
+
+ You may add Your own copyright statement to Your modifications and
+ may provide additional or different license terms and conditions
+ for use, reproduction, or distribution of Your modifications, or
+ for any such Derivative Works as a whole, provided Your use,
+ reproduction, and distribution of the Work otherwise complies with
+ the conditions stated in this License.
+
+ 5. Submission of Contributions. Unless You explicitly state otherwise,
+ any Contribution intentionally submitted for inclusion in the Work
+ by You to the Licensor shall be under the terms and conditions of
+ this License, without any additional terms or conditions.
+ Notwithstanding the above, nothing herein shall supersede or modify
+ the terms of any separate license agreement you may have executed
+ with Licensor regarding such Contributions.
+
+ 6. Trademarks. This License does not grant permission to use the trade
+ names, trademarks, service marks, or product names of the Licensor,
+ except as required for reasonable and customary use in describing the
+ origin of the Work and reproducing the content of the NOTICE file.
+
+ 7. Disclaimer of Warranty. Unless required by applicable law or
+ agreed to in writing, Licensor provides the Work (and each
+ Contributor provides its Contributions) on an "AS IS" BASIS,
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+ implied, including, without limitation, any warranties or conditions
+ of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+ PARTICULAR PURPOSE. You are solely responsible for determining the
+ appropriateness of using or redistributing the Work and assume any
+ risks associated with Your exercise of permissions under this License.
+
+ 8. Limitation of Liability. In no event and under no legal theory,
+ whether in tort (including negligence), contract, or otherwise,
+ unless required by applicable law (such as deliberate and grossly
+ negligent acts) or agreed to in writing, shall any Contributor be
+ liable to You for damages, including any direct, indirect, special,
+ incidental, or consequential damages of any character arising as a
+ result of this License or out of the use or inability to use the
+ Work (including but not limited to damages for loss of goodwill,
+ work stoppage, computer failure or malfunction, or any and all
+ other commercial damages or losses), even if such Contributor
+ has been advised of the possibility of such damages.
+
+ 9. Accepting Warranty or Additional Liability. While redistributing
+ the Work or Derivative Works thereof, You may choose to offer,
+ and charge a fee for, acceptance of support, warranty, indemnity,
+ or other liability obligations and/or rights consistent with this
+ License. However, in accepting such obligations, You may act only
+ on Your own behalf and on Your sole responsibility, not on behalf
+ of any other Contributor, and only if You agree to indemnify,
+ defend, and hold each Contributor harmless for any liability
+ incurred by, or claims asserted against, such Contributor by reason
+ of your accepting any such warranty or additional liability.
+
+ END OF TERMS AND CONDITIONS
+
+ APPENDIX: How to apply the Apache License to your work.
+
+ To apply the Apache License to your work, attach the following
+ boilerplate notice, with the fields enclosed by brackets "[]"
+ replaced with your own identifying information. (Don't include
+ the brackets!) The text should be enclosed in the appropriate
+ comment syntax for the file format. We also recommend that a
+ file or class name and description of purpose be included on the
+ same "printed page" as the copyright notice for easier
+ identification within third-party archives.
+
+ Copyright [yyyy] [name of copyright owner]
+
+ Licensed under the Apache License, Version 2.0 (the "License");
+ you may not use this file except in compliance with the License.
+ You may obtain a copy of the License at
+
+ http://www.apache.org/licenses/LICENSE-2.0
+
+ Unless required by applicable law or agreed to in writing, software
+ distributed under the License is distributed on an "AS IS" BASIS,
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ See the License for the specific language governing permissions and
+ limitations under the License.
diff --git a/kumo/NOTICE b/kumo/NOTICE
new file mode 100644
index 0000000000000000000000000000000000000000..9eaca1d6aa6ed1c0111941f89dd64c8ebd0627da
--- /dev/null
+++ b/kumo/NOTICE
@@ -0,0 +1,63 @@
+KumoTabular (small, medium and large; classifier and regressor) converted to
+ONNX for Tabular Data Workbench.
+
+This is an unofficial conversion. NVIDIA has not reviewed or endorsed it.
+
+Kumo Tabular
+ Source: nvidia/Kumo-Tabular, tag v1.0.1
+ (revision 3c3e10bbdb590ace29e7026847f92db3c603096d)
+ https://huggingface.co/nvidia/Kumo-Tabular
+ Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+ The weights are licensed under the OpenMDW License Agreement, version 1.1
+ (LICENSE).
+
+Model code
+ The graphs were exported from the KumoTabular implementation in
+ structured-data-models 0.2.1 (https://github.com/NVIDIA/structured-data-models),
+ Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES, licensed under the
+ Apache License, Version 2.0 (LICENSE.structured-data-models where it is
+ distributed with these files). In that implementation the cell embedding is
+ derived from TabFM (https://github.com/google-research/tabfm, Copyright 2026
+ Google LLC, Apache License 2.0), and the attention scaling contains portions
+ derived from TabICLv2 (https://github.com/soda-inria/tabicl), whose licence
+ follows at the end of this file.
+
+Changes
+ - -.f16w.onnx: the KumoTabular network (structured-data-models'
+ _KumoTabular.forward, without the preprocessing recipe or the ensemble)
+ exported with torch.onnx (dynamo, opset 20). The cell embedding's column
+ grouping, imputation and Fourier features are written without int64
+ arithmetic or scatters; what they compute is unchanged. Every weight of
+ 1,024 or more elements is stored as float16 and cast back to float32 on
+ load; the other weights are unchanged.
+
+TabICLv2
+
+BSD 3-Clause License
+
+Copyright (c) 2025, Soda team @ Inria
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+1. Redistributions of source code must retain the above copyright notice, this
+ list of conditions and the following disclaimer.
+
+2. Redistributions in binary form must reproduce the above copyright notice,
+ this list of conditions and the following disclaimer in the documentation
+ and/or other materials provided with the distribution.
+
+3. Neither the name of the copyright holder nor the names of its
+ contributors may be used to endorse or promote products derived from
+ this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
diff --git a/kumo/README.md b/kumo/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..4692e6c8ed19f5ec256c2e22b2c93bedaa8876a2
--- /dev/null
+++ b/kumo/README.md
@@ -0,0 +1,125 @@
+---
+license: openmdw-1.1
+library_name: onnx
+base_model: nvidia/Kumo-Tabular
+base_model_relation: quantized
+pipeline_tag: tabular-classification
+tags:
+ - onnx
+ - onnxruntime-web
+ - webgpu
+ - structured-data-models
+ - tabular-classification
+ - tabular-regression
+ - tabular-foundation-model
+ - in-context-learning
+---
+
+# KumoTabular for onnxruntime-web
+
+An unofficial ONNX conversion of NVIDIA's [Kumo Tabular](https://huggingface.co/nvidia/Kumo-Tabular)
+(tag `v1.0.1`) in its three sizes, made for Tabular Data Workbench, which runs
+it in the browser with onnxruntime-web on WebGPU or WebAssembly. The weights
+are NVIDIA's, with the larger ones stored as float16 (see below). NVIDIA has
+not reviewed or endorsed this conversion.
+
+## Files
+
+One graph per size and head, each a single file:
+
+| file | parameters | size |
+| --- | --- | --- |
+| `small-classifier.f16w.onnx` | 27.5M | 55 MB |
+| `small-regressor.f16w.onnx` | 28.5M | 57 MB |
+| `medium-classifier.f16w.onnx` | 61.5M | 124 MB |
+| `medium-regressor.f16w.onnx` | 62.5M | 126 MB |
+| `large-classifier.f16w.onnx` | 213.7M | 428 MB |
+| `large-regressor.f16w.onnx` | 215.7M | 432 MB |
+
+Every graph has the same interface, that of onnx-community's export of the
+small size:
+
+| input / output | shape | type |
+| --- | --- | --- |
+| `x` | (R, C), context rows first, missing values as NaN | float32 |
+| `y_context` | (Rc) | int32 class codes (classifier), float32 (regressor) |
+| `categorical_mask` | (C) | bool |
+| `logits` | (R - Rc, 10) logits or (R - Rc, 999) quantiles | float32 |
+
+The graph is the network only (structured-data-models'
+`_KumoTabular.forward`): it imputes missing cells itself, so NaN must reach
+it unchanged. The preprocessing recipe, the ensemble and the output
+aggregation stay outside it; the app implements them in TypeScript.
+
+In `f16w`, every float32 weight of 1,024 or more elements is stored as
+float16 and cast back to float32 on load, which halves the download and keeps
+compute in float32. The other weights are as NVIDIA released them.
+
+`manifest.json` lists every file with its size, sha256, largest weight and
+total weight bytes (as float32), the int64 shape nodes to keep on the CPU
+under WebGPU (`cpuNodes`), each size's architecture, a memory estimate per
+row, onnxruntime CPU timings, and the parity figures below.
+
+## How they were made
+
+Source: `nvidia/Kumo-Tabular` at `3c3e10bbdb590ace29e7026847f92db3c603096d`
+(tag `v1.0.1`), loaded through NVIDIA's
+[structured-data-models](https://github.com/NVIDIA/structured-data-models)
+0.2.1.
+
+```sh
+cd models/kumo
+uv run export.py # writes public/models/kumo/
+```
+
+`export.py` (in `models/kumo/` of Tabular Data Workbench) exports with
+`torch.onnx` (dynamo, opset 20, PyTorch 2.14 on CPU) and removes the
+exporter's node metadata and doc strings. The cell embedding is
+rewritten so that grouping each column with its neighbours, imputing missing
+cells and building the Fourier features need no int64 Mod, int64 sums or
+scatters; the result matches structured-data-models' forward pass. The
+graphs keep a few int64 shape computations, listed in the manifest.
+
+## Parity
+
+onnxruntime 1.30 (CPU) against structured-data-models' forward pass in
+PyTorch, maximum absolute difference over network inputs captured from its
+recipe (4 ensemble members, 540 context and 100 query rows of a synthetic
+table with missing values and two categorical columns) and over random
+inputs of five shapes:
+
+| graph | float32 graph, recipe inputs | f16w, recipe inputs | f16w, random inputs | largest output |
+| --- | --- | --- | --- | --- |
+| small classifier (logits) | 5.2e-5 | 1.9e-2 | 1.8e-2 | 14.5 |
+| small regressor (quantiles) | 2.2e-5 | 6.1e-3 | 1.8e-2 | 4.6 |
+| medium classifier | 4.2e-5 | 1.5e-2 | 1.8e-2 | 30.1 |
+| medium regressor | 9.1e-6 | 1.4e-3 | 6.5e-3 | 3.7 |
+| large classifier | 2.4e-5 | 1.1e-2 | 2.4e-2 | 8.3 |
+| large regressor | 6.9e-5 | 2.7e-3 | 4.2e-3 | 4.5 |
+
+The float32 graphs (built for this check, not published) show the export's
+own error; the rest is the rounding of the weights to float16. Regressor
+outputs are in units of the standardised target.
+
+Through the whole recipe (8 ensemble members; breast cancer with and without
+missing values, wine, diabetes), the f16w graphs of every size move class
+probabilities by at most 4.1e-4 and regression quantiles by at most 9e-4
+standard deviations from structured-data-models' predictions, with the same
+accuracy and R² within 2e-5.
+
+## Licence
+
+The weights are released by NVIDIA under the OpenMDW License Agreement 1.1
+(`LICENSE`, Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES). It allows
+use, modification and redistribution provided that the agreement and the
+notices of origin travel with the weights, and it ends the rights of anyone
+who sues claiming the model infringes a patent or copyright.
+
+The graphs were exported from structured-data-models' model code, licensed
+under Apache-2.0 (`LICENSE.structured-data-models`), whose cell embedding is
+derived from TabFM (Google, Apache-2.0) and whose attention scaling contains
+portions of TabICLv2 (Soda team, Inria, BSD-3-Clause). `NOTICE` states the
+origin and the changes and carries the TabICLv2 licence.
+
+Keep `LICENSE`, `NOTICE` and `LICENSE.structured-data-models` with the
+graphs when redistributing them.
diff --git a/kumo/large-classifier.f16w.onnx b/kumo/large-classifier.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..b7d3b331d339e1d2f21713c75187872780cbfa1d
--- /dev/null
+++ b/kumo/large-classifier.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d832eee622716e42ef30c33e76e0cb49b09298cacddfc28f7d315a4f4f7a8cb9
+size 428161256
diff --git a/kumo/large-regressor.f16w.onnx b/kumo/large-regressor.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..b44b14d77c325f9d32bb33fbe550cf2f18155044
--- /dev/null
+++ b/kumo/large-regressor.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b144ae2177a19b16e0db9232d66f7be242cee476d4f44cca54dad89915d62d0
+size 432193733
diff --git a/kumo/manifest.json b/kumo/manifest.json
new file mode 100644
index 0000000000000000000000000000000000000000..6ab3211a5cb67a78aa3b9c97f7879399aa9394f6
--- /dev/null
+++ b/kumo/manifest.json
@@ -0,0 +1,3249 @@
+{
+ "model": "KumoTabular",
+ "source": {
+ "repo": "nvidia/Kumo-Tabular",
+ "revision": "3c3e10bbdb590ace29e7026847f92db3c603096d",
+ "tag": "v1.0.1",
+ "sdm": "0.2.1",
+ "license": "OpenMDW-1.1"
+ },
+ "opset": 20,
+ "exporter": "torch.onnx (dynamo) 2.14.0+cpu, onnx 1.23.2, onnxscript 0.7.2",
+ "limits": {
+ "maxClasses": 10,
+ "numQuantiles": 999,
+ "maxColumns": 500
+ },
+ "sizes": {
+ "small": {
+ "parameters": {
+ "classifier": 27458266,
+ "regressor": 28466231
+ },
+ "dims": {
+ "cellChannels": 128,
+ "embeddingLayers": 4,
+ "embeddingHeads": 4,
+ "inducingPoints": 128,
+ "groupSize": 3,
+ "frequencies": 32,
+ "readoutTokens": 4,
+ "iclWidth": 512,
+ "iclLayers": 12,
+ "heads": 8,
+ "keyValueHeadsForQueries": 8
+ },
+ "timeFactor": 1.0,
+ "memory": {
+ "perRowBytes": {
+ "perColumn": 10240,
+ "perContextRow": 64,
+ "constant": 49152,
+ "perTokenSquared": 32
+ },
+ "largestTensorBytes": {
+ "perRowPerColumn": 2048,
+ "perRowPerContextRow": 32,
+ "perRowPerTokenSquared": 16
+ },
+ "measuredRatio": {
+ "min": 0.5,
+ "max": 0.86
+ }
+ },
+ "memoryCorrected": {
+ "perRowBytes": {
+ "perColumn": 25600,
+ "perContextRow": 64,
+ "constant": 49152,
+ "perTokenSquared": 32
+ },
+ "largestTensorBytes": {
+ "perRowPerColumn": 2048,
+ "perRowPerContextRow": 32,
+ "perRowPerTokenSquared": 16
+ },
+ "measuredRatio": {
+ "min": 1.04,
+ "max": 1.28
+ }
+ },
+ "measured": {
+ "onnxruntime": "1.30.0 CPU, f16w classifier, 64 query rows, no arena or memory pattern",
+ "runs": [
+ {
+ "seconds": 1.2154279148671776,
+ "firstSeconds": 1.2268668899778277,
+ "peakExtraBytes": 477560832,
+ "contextRows": 500,
+ "columns": 30,
+ "queryRows": 64,
+ "threads": 1
+ },
+ {
+ "seconds": 0.5068663719575852,
+ "firstSeconds": 0.5414570381399244,
+ "peakExtraBytes": 481984512,
+ "contextRows": 500,
+ "columns": 30,
+ "queryRows": 64,
+ "threads": "default"
+ },
+ {
+ "seconds": 2.560364630073309,
+ "firstSeconds": 2.559679195052013,
+ "peakExtraBytes": 884776960,
+ "contextRows": 1000,
+ "columns": 30,
+ "queryRows": 64,
+ "threads": 1
+ },
+ {
+ "seconds": 1.1203733100555837,
+ "firstSeconds": 1.1665117950178683,
+ "peakExtraBytes": 884326400,
+ "contextRows": 1000,
+ "columns": 30,
+ "queryRows": 64,
+ "threads": "default"
+ },
+ {
+ "seconds": 6.963859544834122,
+ "firstSeconds": 6.509852007962763,
+ "peakExtraBytes": 1587236864,
+ "contextRows": 2000,
+ "columns": 30,
+ "queryRows": 64,
+ "threads": 1
+ },
+ {
+ "seconds": 3.3934620078653097,
+ "firstSeconds": 7.248737532878295,
+ "peakExtraBytes": 1584603136,
+ "contextRows": 2000,
+ "columns": 30,
+ "queryRows": 64,
+ "threads": "default"
+ },
+ {
+ "seconds": 1.6311015160754323,
+ "firstSeconds": 1.6794085390865803,
+ "peakExtraBytes": 1142321152,
+ "contextRows": 500,
+ "columns": 80,
+ "queryRows": 64,
+ "threads": "default"
+ },
+ {
+ "seconds": 2.7221560010220855,
+ "firstSeconds": 2.73818067391403,
+ "peakExtraBytes": 1841098752,
+ "contextRows": 500,
+ "columns": 130,
+ "queryRows": 64,
+ "threads": "default"
+ },
+ {
+ "seconds": 1.5351381269283593,
+ "firstSeconds": 1.5033495931420475,
+ "peakExtraBytes": 593227776,
+ "contextRows": 2000,
+ "columns": 5,
+ "queryRows": 64,
+ "threads": "default"
+ },
+ {
+ "seconds": 2.6551022299099714,
+ "firstSeconds": 2.6324840888846666,
+ "peakExtraBytes": 1045094400,
+ "contextRows": 3000,
+ "columns": 5,
+ "queryRows": 64,
+ "threads": "default"
+ }
+ ],
+ "timeFactorDefaultThreads": 1.0
+ }
+ },
+ "medium": {
+ "parameters": {
+ "classifier": 61485274,
+ "regressor": 62492087
+ },
+ "dims": {
+ "cellChannels": 256,
+ "embeddingLayers": 6,
+ "embeddingHeads": 4,
+ "inducingPoints": 256,
+ "groupSize": 3,
+ "frequencies": 32,
+ "readoutTokens": 4,
+ "iclWidth": 512,
+ "iclLayers": 24,
+ "heads": 8,
+ "keyValueHeadsForQueries": 2
+ },
+ "timeFactor": 3.67,
+ "memory": {
+ "perRowBytes": {
+ "perColumn": 20480,
+ "perContextRow": 64,
+ "constant": 73728,
+ "perTokenSquared": 32
+ },
+ "largestTensorBytes": {
+ "perRowPerColumn": 4096,
+ "perRowPerContextRow": 32,
+ "perRowPerTokenSquared": 16
+ },
+ "measuredRatio": {
+ "min": 0.48,
+ "max": 0.92
+ }
+ },
+ "memoryCorrected": {
+ "perRowBytes": {
+ "perColumn": 51200,
+ "perContextRow": 64,
+ "constant": 73728,
+ "perTokenSquared": 32
+ },
+ "largestTensorBytes": {
+ "perRowPerColumn": 4096,
+ "perRowPerContextRow": 32,
+ "perRowPerTokenSquared": 16
+ },
+ "measuredRatio": {
+ "min": 1.06,
+ "max": 1.3
+ }
+ },
+ "measured": {
+ "onnxruntime": "1.30.0 CPU, f16w classifier, 64 query rows, no arena or memory pattern",
+ "runs": [
+ {
+ "seconds": 4.842576633905992,
+ "firstSeconds": 5.1058256949763745,
+ "peakExtraBytes": 878211072,
+ "contextRows": 500,
+ "columns": 30,
+ "queryRows": 64,
+ "threads": 1
+ },
+ {
+ "seconds": 2.275825258111581,
+ "firstSeconds": 2.3033246230334044,
+ "peakExtraBytes": 893087744,
+ "contextRows": 500,
+ "columns": 30,
+ "queryRows": 64,
+ "threads": "default"
+ },
+ {
+ "seconds": 9.402626937953755,
+ "firstSeconds": 9.402824718970805,
+ "peakExtraBytes": 1586716672,
+ "contextRows": 1000,
+ "columns": 30,
+ "queryRows": 64,
+ "threads": 1
+ },
+ {
+ "seconds": 4.430133221903816,
+ "firstSeconds": 4.394938342971727,
+ "peakExtraBytes": 1575854080,
+ "contextRows": 1000,
+ "columns": 30,
+ "queryRows": 64,
+ "threads": "default"
+ },
+ {
+ "seconds": 20.64435661304742,
+ "firstSeconds": 20.65433070808649,
+ "peakExtraBytes": 2884071424,
+ "contextRows": 2000,
+ "columns": 30,
+ "queryRows": 64,
+ "threads": 1
+ },
+ {
+ "seconds": 9.811896333936602,
+ "firstSeconds": 9.840921485098079,
+ "peakExtraBytes": 2871320576,
+ "contextRows": 2000,
+ "columns": 30,
+ "queryRows": 64,
+ "threads": "default"
+ },
+ {
+ "seconds": 5.872162852901965,
+ "firstSeconds": 5.954272715141997,
+ "peakExtraBytes": 2156363776,
+ "contextRows": 500,
+ "columns": 80,
+ "queryRows": 64,
+ "threads": "default"
+ },
+ {
+ "seconds": 9.386592728085816,
+ "firstSeconds": 9.514258791925386,
+ "peakExtraBytes": 3423678464,
+ "contextRows": 500,
+ "columns": 130,
+ "queryRows": 64,
+ "threads": "default"
+ },
+ {
+ "seconds": 3.6557655630167574,
+ "firstSeconds": 3.689454659121111,
+ "peakExtraBytes": 761860096,
+ "contextRows": 2000,
+ "columns": 5,
+ "queryRows": 64,
+ "threads": "default"
+ },
+ {
+ "seconds": 6.346925284946337,
+ "firstSeconds": 6.677469216054305,
+ "peakExtraBytes": 1235943424,
+ "contextRows": 3000,
+ "columns": 5,
+ "queryRows": 64,
+ "threads": "default"
+ }
+ ],
+ "timeFactorDefaultThreads": 3.45
+ }
+ },
+ "large": {
+ "parameters": {
+ "classifier": 213668250,
+ "regressor": 215683191
+ },
+ "dims": {
+ "cellChannels": 256,
+ "embeddingLayers": 6,
+ "embeddingHeads": 4,
+ "inducingPoints": 256,
+ "groupSize": 3,
+ "frequencies": 32,
+ "readoutTokens": 4,
+ "iclWidth": 1024,
+ "iclLayers": 24,
+ "heads": 16,
+ "keyValueHeadsForQueries": 2
+ },
+ "timeFactor": 5.05,
+ "memory": {
+ "perRowBytes": {
+ "perColumn": 20480,
+ "perContextRow": 128,
+ "constant": 98304,
+ "perTokenSquared": 32
+ },
+ "largestTensorBytes": {
+ "perRowPerColumn": 4096,
+ "perRowPerContextRow": 64,
+ "perRowPerTokenSquared": 16
+ },
+ "measuredRatio": {
+ "min": 0.53,
+ "max": 0.98
+ }
+ },
+ "memoryCorrected": {
+ "perRowBytes": {
+ "perColumn": 51200,
+ "perContextRow": 128,
+ "constant": 98304,
+ "perTokenSquared": 32
+ },
+ "largestTensorBytes": {
+ "perRowPerColumn": 4096,
+ "perRowPerContextRow": 64,
+ "perRowPerTokenSquared": 16
+ },
+ "measuredRatio": {
+ "min": 1.06,
+ "max": 1.37
+ }
+ },
+ "measured": {
+ "onnxruntime": "1.30.0 CPU, f16w classifier, 64 query rows, no arena or memory pattern",
+ "runs": [
+ {
+ "seconds": 6.23878475301899,
+ "firstSeconds": 6.328202632023022,
+ "peakExtraBytes": 870662144,
+ "contextRows": 500,
+ "columns": 30,
+ "queryRows": 64,
+ "threads": 1
+ },
+ {
+ "seconds": 2.7787703690119088,
+ "firstSeconds": 2.7496005680877715,
+ "peakExtraBytes": 872816640,
+ "contextRows": 500,
+ "columns": 30,
+ "queryRows": 64,
+ "threads": "default"
+ },
+ {
+ "seconds": 12.92174591589719,
+ "firstSeconds": 12.876223447965458,
+ "peakExtraBytes": 1568690176,
+ "contextRows": 1000,
+ "columns": 30,
+ "queryRows": 64,
+ "threads": 1
+ },
+ {
+ "seconds": 5.7077240131329745,
+ "firstSeconds": 5.823087724857032,
+ "peakExtraBytes": 1583984640,
+ "contextRows": 1000,
+ "columns": 30,
+ "queryRows": 64,
+ "threads": "default"
+ },
+ {
+ "seconds": 28.87913363100961,
+ "firstSeconds": 29.218742611818016,
+ "peakExtraBytes": 2914562048,
+ "contextRows": 2000,
+ "columns": 30,
+ "queryRows": 64,
+ "threads": 1
+ },
+ {
+ "seconds": 12.427342433016747,
+ "firstSeconds": 12.593973094131798,
+ "peakExtraBytes": 2911326208,
+ "contextRows": 2000,
+ "columns": 30,
+ "queryRows": 64,
+ "threads": "default"
+ },
+ {
+ "seconds": 6.2925732030998915,
+ "firstSeconds": 6.2246922550257295,
+ "peakExtraBytes": 2176503808,
+ "contextRows": 500,
+ "columns": 80,
+ "queryRows": 64,
+ "threads": "default"
+ },
+ {
+ "seconds": 9.900329439900815,
+ "firstSeconds": 9.813142587197945,
+ "peakExtraBytes": 3500654592,
+ "contextRows": 500,
+ "columns": 130,
+ "queryRows": 64,
+ "threads": "default"
+ },
+ {
+ "seconds": 6.504102888982743,
+ "firstSeconds": 6.624696749961004,
+ "peakExtraBytes": 1194524672,
+ "contextRows": 2000,
+ "columns": 5,
+ "queryRows": 64,
+ "threads": "default"
+ },
+ {
+ "seconds": 11.218624849803746,
+ "firstSeconds": 11.851737375138327,
+ "peakExtraBytes": 1834246144,
+ "contextRows": 3000,
+ "columns": 5,
+ "queryRows": 64,
+ "threads": "default"
+ }
+ ],
+ "timeFactorDefaultThreads": 4.23
+ }
+ }
+ },
+ "files": [
+ {
+ "size": "small",
+ "task": "classifier",
+ "variant": "f16w",
+ "file": "small-classifier.f16w.onnx",
+ "bytes": 55301219,
+ "sha256": "1e51425be1d80433b2709c2cda2eda8dd6d2f3190aae7afd9e4f58b8db5fe97d",
+ "largestInitializerBytes": 3145728,
+ "weightBytesFp32": 109834084,
+ "nodes": 2188,
+ "cpuNodes": [
+ "node_sym_size_int_71",
+ "node_sym_size_int_72",
+ "node_sym_size_int_73",
+ "node_add_6743",
+ "node_Concat_271",
+ "node_Gather_284",
+ "node_Range_285",
+ "node_Unsqueeze_286",
+ "node_Slice_288",
+ "node_Unsqueeze_289",
+ "node_Gather_310",
+ "node_Range_311",
+ "node_Slice_315",
+ "node_Unsqueeze_316",
+ "node_Concat_326",
+ "node_Concat_359",
+ "node_Concat_389",
+ "node_Concat_408",
+ "node_Slice_432",
+ "node_Slice_433",
+ "node_Slice_435",
+ "node_Concat_437",
+ "node_Concat_440",
+ "node_Concat_487",
+ "node_Concat_519",
+ "node_Slice_555",
+ "node_Slice_556",
+ "node_Slice_558",
+ "node_Concat_560",
+ "node_Concat_563",
+ "node_Concat_582",
+ "node_Reshape_607",
+ "node_Concat_610",
+ "node_Concat_677",
+ "node_Slice_713",
+ "node_Slice_714",
+ "node_Slice_716",
+ "node_Concat_718",
+ "node_Concat_721",
+ "node_Concat_740",
+ "node_Concat_753",
+ "node_Slice_851",
+ "node_Slice_852",
+ "node_Slice_854",
+ "node_Concat_856",
+ "node_Concat_859",
+ "node_Slice_974",
+ "node_Slice_975",
+ "node_Slice_977",
+ "node_Concat_979",
+ "node_Concat_982",
+ "node_Slice_1131",
+ "node_Slice_1132",
+ "node_Slice_1134",
+ "node_Concat_1136",
+ "node_Concat_1139",
+ "node_Slice_1269",
+ "node_Slice_1270",
+ "node_Slice_1272",
+ "node_Concat_1274",
+ "node_Concat_1277",
+ "node_Slice_1392",
+ "node_Slice_1393",
+ "node_Slice_1395",
+ "node_Concat_1397",
+ "node_Concat_1400",
+ "node_Slice_1549",
+ "node_Slice_1550",
+ "node_Slice_1552",
+ "node_Concat_1554",
+ "node_Concat_1557",
+ "node_Slice_1687",
+ "node_Slice_1688",
+ "node_Slice_1690",
+ "node_Concat_1692",
+ "node_Concat_1695",
+ "node_Slice_1810",
+ "node_Slice_1811",
+ "node_Slice_1813",
+ "node_Concat_1815",
+ "node_Concat_1818",
+ "node_Slice_1977",
+ "node_Slice_1978",
+ "node_Slice_1980",
+ "node_Concat_1982",
+ "node_Concat_1985",
+ "node_Gather_2037",
+ "node_Range_2038",
+ "node_Slice_2041",
+ "node_Unsqueeze_2042",
+ "node_Gather_2080",
+ "node_Range_2081",
+ "node_Slice_2085",
+ "node_Unsqueeze_2086",
+ "node_Slice_2191",
+ "node_Slice_2192",
+ "node_Slice_2194",
+ "node_Concat_2196",
+ "node_Concat_2199",
+ "node_Slice_2311",
+ "node_Slice_2312",
+ "node_Slice_2314",
+ "node_Concat_2316",
+ "node_Concat_2319",
+ "node_Slice_2431",
+ "node_Slice_2432",
+ "node_Slice_2434",
+ "node_Concat_2436",
+ "node_Concat_2439",
+ "node_Slice_2551",
+ "node_Slice_2552",
+ "node_Slice_2554",
+ "node_Concat_2556",
+ "node_Concat_2559",
+ "node_Slice_2671",
+ "node_Slice_2672",
+ "node_Slice_2674",
+ "node_Concat_2676",
+ "node_Concat_2679",
+ "node_Slice_2791",
+ "node_Slice_2792",
+ "node_Slice_2794",
+ "node_Concat_2796",
+ "node_Concat_2799",
+ "node_Slice_2911",
+ "node_Slice_2912",
+ "node_Slice_2914",
+ "node_Concat_2916",
+ "node_Concat_2919",
+ "node_Slice_3031",
+ "node_Slice_3032",
+ "node_Slice_3034",
+ "node_Concat_3036",
+ "node_Concat_3039",
+ "node_Slice_3151",
+ "node_Slice_3152",
+ "node_Slice_3154",
+ "node_Concat_3156",
+ "node_Concat_3159",
+ "node_Slice_3271",
+ "node_Slice_3272",
+ "node_Slice_3274",
+ "node_Concat_3276",
+ "node_Concat_3279",
+ "node_Slice_3391",
+ "node_Slice_3392",
+ "node_Slice_3394",
+ "node_Concat_3396",
+ "node_Concat_3399",
+ "node_mul_5593",
+ "node_add_6745",
+ "node_Reshape_3477",
+ "node_Concat_3480",
+ "node_Slice_3520",
+ "node_Slice_3521",
+ "node_Slice_3523",
+ "node_Concat_3525",
+ "node_Concat_3528"
+ ],
+ "int64Arithmetic": [
+ "node_add_6743 (Add)",
+ "node_mul_5593 (Mul)",
+ "node_add_6745 (Add)"
+ ],
+ "ops": {
+ "Mul": 320,
+ "Add": 249,
+ "Reshape": 205,
+ "Transpose": 152,
+ "Cast": 143,
+ "Pow": 119,
+ "Reciprocal": 119,
+ "ReduceMean": 119,
+ "Sqrt": 119,
+ "Slice": 118,
+ "MatMul": 114,
+ "Concat": 86,
+ "Split": 83,
+ "Gemm": 66,
+ "Shape": 31,
+ "Gelu": 29,
+ "Softmax": 24,
+ "Unsqueeze": 24,
+ "Expand": 17,
+ "Sub": 8,
+ "Gather": 6,
+ "Range": 5,
+ "Where": 5,
+ "ScatterND": 4,
+ "Squeeze": 4,
+ "Tanh": 4,
+ "Clip": 3,
+ "ReduceSum": 3,
+ "Cos": 2,
+ "Log": 2,
+ "Sin": 2,
+ "Div": 1,
+ "IsNaN": 1,
+ "Not": 1
+ },
+ "inputs": [
+ {
+ "name": "x",
+ "type": "float32",
+ "shape": [
+ "R",
+ "C"
+ ]
+ },
+ {
+ "name": "y_context",
+ "type": "int32",
+ "shape": [
+ "Rc"
+ ]
+ },
+ {
+ "name": "categorical_mask",
+ "type": "bool",
+ "shape": [
+ "C"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "logits",
+ "type": "float32",
+ "shape": [
+ "Rq",
+ 10
+ ]
+ }
+ ]
+ },
+ {
+ "size": "small",
+ "task": "regressor",
+ "variant": "f16w",
+ "file": "small-regressor.f16w.onnx",
+ "bytes": 57320387,
+ "sha256": "e97b5abbc5bdf71106d2768656c8d91d981e1359251a137983b711baddb745e8",
+ "largestInitializerBytes": 4091904,
+ "weightBytesFp32": 113865944,
+ "nodes": 2187,
+ "cpuNodes": [
+ "node_sym_size_int_71",
+ "node_sym_size_int_72",
+ "node_sym_size_int_73",
+ "node_add_6749",
+ "node_Concat_271",
+ "node_Gather_284",
+ "node_Range_285",
+ "node_Unsqueeze_286",
+ "node_Slice_288",
+ "node_Unsqueeze_289",
+ "node_Gather_310",
+ "node_Range_311",
+ "node_Slice_315",
+ "node_Unsqueeze_316",
+ "node_Concat_326",
+ "node_Concat_359",
+ "node_Concat_389",
+ "node_Concat_408",
+ "node_Slice_432",
+ "node_Slice_433",
+ "node_Slice_435",
+ "node_Concat_437",
+ "node_Concat_440",
+ "node_Concat_487",
+ "node_Concat_519",
+ "node_Slice_555",
+ "node_Slice_556",
+ "node_Slice_558",
+ "node_Concat_560",
+ "node_Concat_563",
+ "node_Concat_582",
+ "node_Reshape_607",
+ "node_Concat_610",
+ "node_Concat_677",
+ "node_Slice_713",
+ "node_Slice_714",
+ "node_Slice_716",
+ "node_Concat_718",
+ "node_Concat_721",
+ "node_Concat_740",
+ "node_Concat_753",
+ "node_Slice_851",
+ "node_Slice_852",
+ "node_Slice_854",
+ "node_Concat_856",
+ "node_Concat_859",
+ "node_Slice_974",
+ "node_Slice_975",
+ "node_Slice_977",
+ "node_Concat_979",
+ "node_Concat_982",
+ "node_Slice_1131",
+ "node_Slice_1132",
+ "node_Slice_1134",
+ "node_Concat_1136",
+ "node_Concat_1139",
+ "node_Slice_1269",
+ "node_Slice_1270",
+ "node_Slice_1272",
+ "node_Concat_1274",
+ "node_Concat_1277",
+ "node_Slice_1392",
+ "node_Slice_1393",
+ "node_Slice_1395",
+ "node_Concat_1397",
+ "node_Concat_1400",
+ "node_Slice_1549",
+ "node_Slice_1550",
+ "node_Slice_1552",
+ "node_Concat_1554",
+ "node_Concat_1557",
+ "node_Slice_1687",
+ "node_Slice_1688",
+ "node_Slice_1690",
+ "node_Concat_1692",
+ "node_Concat_1695",
+ "node_Slice_1810",
+ "node_Slice_1811",
+ "node_Slice_1813",
+ "node_Concat_1815",
+ "node_Concat_1818",
+ "node_Slice_1977",
+ "node_Slice_1978",
+ "node_Slice_1980",
+ "node_Concat_1982",
+ "node_Concat_1985",
+ "node_Gather_2037",
+ "node_Range_2038",
+ "node_Slice_2041",
+ "node_Unsqueeze_2042",
+ "node_Gather_2080",
+ "node_Range_2081",
+ "node_Slice_2085",
+ "node_Unsqueeze_2086",
+ "node_Slice_2191",
+ "node_Slice_2192",
+ "node_Slice_2194",
+ "node_Concat_2196",
+ "node_Concat_2199",
+ "node_Slice_2311",
+ "node_Slice_2312",
+ "node_Slice_2314",
+ "node_Concat_2316",
+ "node_Concat_2319",
+ "node_Slice_2431",
+ "node_Slice_2432",
+ "node_Slice_2434",
+ "node_Concat_2436",
+ "node_Concat_2439",
+ "node_Slice_2551",
+ "node_Slice_2552",
+ "node_Slice_2554",
+ "node_Concat_2556",
+ "node_Concat_2559",
+ "node_Slice_2671",
+ "node_Slice_2672",
+ "node_Slice_2674",
+ "node_Concat_2676",
+ "node_Concat_2679",
+ "node_Slice_2791",
+ "node_Slice_2792",
+ "node_Slice_2794",
+ "node_Concat_2796",
+ "node_Concat_2799",
+ "node_Slice_2911",
+ "node_Slice_2912",
+ "node_Slice_2914",
+ "node_Concat_2916",
+ "node_Concat_2919",
+ "node_Slice_3031",
+ "node_Slice_3032",
+ "node_Slice_3034",
+ "node_Concat_3036",
+ "node_Concat_3039",
+ "node_Slice_3151",
+ "node_Slice_3152",
+ "node_Slice_3154",
+ "node_Concat_3156",
+ "node_Concat_3159",
+ "node_Slice_3271",
+ "node_Slice_3272",
+ "node_Slice_3274",
+ "node_Concat_3276",
+ "node_Concat_3279",
+ "node_Slice_3391",
+ "node_Slice_3392",
+ "node_Slice_3394",
+ "node_Concat_3396",
+ "node_Concat_3399",
+ "node_mul_5595",
+ "node_add_6751",
+ "node_Reshape_3477",
+ "node_Concat_3480",
+ "node_Slice_3520",
+ "node_Slice_3521",
+ "node_Slice_3523",
+ "node_Concat_3525",
+ "node_Concat_3528"
+ ],
+ "int64Arithmetic": [
+ "node_add_6749 (Add)",
+ "node_mul_5595 (Mul)",
+ "node_add_6751 (Add)"
+ ],
+ "ops": {
+ "Mul": 320,
+ "Add": 249,
+ "Reshape": 205,
+ "Transpose": 152,
+ "Cast": 141,
+ "Pow": 119,
+ "Reciprocal": 119,
+ "ReduceMean": 119,
+ "Sqrt": 119,
+ "Slice": 118,
+ "MatMul": 114,
+ "Concat": 86,
+ "Split": 83,
+ "Gemm": 68,
+ "Shape": 31,
+ "Gelu": 29,
+ "Unsqueeze": 25,
+ "Softmax": 24,
+ "Expand": 17,
+ "Sub": 8,
+ "Range": 5,
+ "Where": 5,
+ "Gather": 4,
+ "ScatterND": 4,
+ "Squeeze": 4,
+ "Tanh": 4,
+ "Clip": 3,
+ "ReduceSum": 3,
+ "Cos": 2,
+ "Log": 2,
+ "Sin": 2,
+ "Div": 1,
+ "IsNaN": 1,
+ "Not": 1
+ },
+ "inputs": [
+ {
+ "name": "x",
+ "type": "float32",
+ "shape": [
+ "R",
+ "C"
+ ]
+ },
+ {
+ "name": "y_context",
+ "type": "float32",
+ "shape": [
+ "Rc"
+ ]
+ },
+ {
+ "name": "categorical_mask",
+ "type": "bool",
+ "shape": [
+ "C"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "logits",
+ "type": "float32",
+ "shape": [
+ "Rq",
+ 999
+ ]
+ }
+ ]
+ },
+ {
+ "size": "medium",
+ "task": "classifier",
+ "variant": "f16w",
+ "file": "medium-classifier.f16w.onnx",
+ "bytes": 123899156,
+ "sha256": "20217f82c506f01852c6b9da36f5d6adfb7424a441db66bce7b210b420ea04f1",
+ "largestInitializerBytes": 3145728,
+ "weightBytesFp32": 245991872,
+ "nodes": 5092,
+ "cpuNodes": [
+ "node_sym_size_int_166",
+ "node_sym_size_int_167",
+ "node_sym_size_int_168",
+ "node_add_15333",
+ "node_Concat_271",
+ "node_Gather_284",
+ "node_Range_285",
+ "node_Unsqueeze_286",
+ "node_Slice_288",
+ "node_Unsqueeze_289",
+ "node_Gather_310",
+ "node_Range_311",
+ "node_Slice_315",
+ "node_Unsqueeze_316",
+ "node_Concat_326",
+ "node_Concat_359",
+ "node_Concat_389",
+ "node_Concat_408",
+ "node_Slice_432",
+ "node_Slice_433",
+ "node_Slice_435",
+ "node_Concat_437",
+ "node_Concat_440",
+ "node_Concat_487",
+ "node_Concat_519",
+ "node_Slice_555",
+ "node_Slice_556",
+ "node_Slice_558",
+ "node_Concat_560",
+ "node_Concat_563",
+ "node_Concat_582",
+ "node_Reshape_607",
+ "node_Concat_610",
+ "node_Concat_677",
+ "node_Slice_713",
+ "node_Slice_714",
+ "node_Slice_716",
+ "node_Concat_718",
+ "node_Concat_721",
+ "node_Concat_740",
+ "node_Concat_753",
+ "node_Slice_851",
+ "node_Slice_852",
+ "node_Slice_854",
+ "node_Concat_856",
+ "node_Concat_859",
+ "node_Slice_974",
+ "node_Slice_975",
+ "node_Slice_977",
+ "node_Concat_979",
+ "node_Concat_982",
+ "node_Slice_1131",
+ "node_Slice_1132",
+ "node_Slice_1134",
+ "node_Concat_1136",
+ "node_Concat_1139",
+ "node_Slice_1269",
+ "node_Slice_1270",
+ "node_Slice_1272",
+ "node_Concat_1274",
+ "node_Concat_1277",
+ "node_Slice_1392",
+ "node_Slice_1393",
+ "node_Slice_1395",
+ "node_Concat_1397",
+ "node_Concat_1400",
+ "node_Slice_1549",
+ "node_Slice_1550",
+ "node_Slice_1552",
+ "node_Concat_1554",
+ "node_Concat_1557",
+ "node_Slice_1687",
+ "node_Slice_1688",
+ "node_Slice_1690",
+ "node_Concat_1692",
+ "node_Concat_1695",
+ "node_Slice_1810",
+ "node_Slice_1811",
+ "node_Slice_1813",
+ "node_Concat_1815",
+ "node_Concat_1818",
+ "node_Slice_1967",
+ "node_Slice_1968",
+ "node_Slice_1970",
+ "node_Concat_1972",
+ "node_Concat_1975",
+ "node_Slice_2105",
+ "node_Slice_2106",
+ "node_Slice_2108",
+ "node_Concat_2110",
+ "node_Concat_2113",
+ "node_Slice_2228",
+ "node_Slice_2229",
+ "node_Slice_2231",
+ "node_Concat_2233",
+ "node_Concat_2236",
+ "node_Slice_2385",
+ "node_Slice_2386",
+ "node_Slice_2388",
+ "node_Concat_2390",
+ "node_Concat_2393",
+ "node_Slice_2523",
+ "node_Slice_2524",
+ "node_Slice_2526",
+ "node_Concat_2528",
+ "node_Concat_2531",
+ "node_Slice_2646",
+ "node_Slice_2647",
+ "node_Slice_2649",
+ "node_Concat_2651",
+ "node_Concat_2654",
+ "node_Slice_2813",
+ "node_Slice_2814",
+ "node_Slice_2816",
+ "node_Concat_2818",
+ "node_Concat_2821",
+ "node_Gather_2869",
+ "node_Range_2870",
+ "node_Slice_2873",
+ "node_Unsqueeze_2874",
+ "node_Gather_2895",
+ "node_Range_2896",
+ "node_Slice_2900",
+ "node_Unsqueeze_2901",
+ "node_Slice_3002",
+ "node_Slice_3003",
+ "node_Slice_3005",
+ "node_Concat_3007",
+ "node_Concat_3010",
+ "node_mul_12678",
+ "node_add_15335",
+ "node_Reshape_3103",
+ "node_Concat_3106",
+ "node_Concat_3154",
+ "node_Concat_3158",
+ "node_Slice_3164",
+ "node_Slice_3165",
+ "node_Slice_3167",
+ "node_Concat_3169",
+ "node_Concat_3172",
+ "node_Slice_3292",
+ "node_Slice_3293",
+ "node_Slice_3295",
+ "node_Concat_3297",
+ "node_Concat_3300",
+ "node_Concat_3442",
+ "node_Concat_3446",
+ "node_Slice_3452",
+ "node_Slice_3453",
+ "node_Slice_3455",
+ "node_Concat_3457",
+ "node_Concat_3460",
+ "node_Slice_3580",
+ "node_Slice_3581",
+ "node_Slice_3583",
+ "node_Concat_3585",
+ "node_Concat_3588",
+ "node_Concat_3730",
+ "node_Concat_3734",
+ "node_Slice_3740",
+ "node_Slice_3741",
+ "node_Slice_3743",
+ "node_Concat_3745",
+ "node_Concat_3748",
+ "node_Slice_3868",
+ "node_Slice_3869",
+ "node_Slice_3871",
+ "node_Concat_3873",
+ "node_Concat_3876",
+ "node_Concat_4018",
+ "node_Concat_4022",
+ "node_Slice_4028",
+ "node_Slice_4029",
+ "node_Slice_4031",
+ "node_Concat_4033",
+ "node_Concat_4036",
+ "node_Slice_4156",
+ "node_Slice_4157",
+ "node_Slice_4159",
+ "node_Concat_4161",
+ "node_Concat_4164",
+ "node_Concat_4306",
+ "node_Concat_4310",
+ "node_Slice_4316",
+ "node_Slice_4317",
+ "node_Slice_4319",
+ "node_Concat_4321",
+ "node_Concat_4324",
+ "node_Slice_4444",
+ "node_Slice_4445",
+ "node_Slice_4447",
+ "node_Concat_4449",
+ "node_Concat_4452",
+ "node_Concat_4594",
+ "node_Concat_4598",
+ "node_Slice_4604",
+ "node_Slice_4605",
+ "node_Slice_4607",
+ "node_Concat_4609",
+ "node_Concat_4612",
+ "node_Slice_4732",
+ "node_Slice_4733",
+ "node_Slice_4735",
+ "node_Concat_4737",
+ "node_Concat_4740",
+ "node_Concat_4882",
+ "node_Concat_4886",
+ "node_Slice_4892",
+ "node_Slice_4893",
+ "node_Slice_4895",
+ "node_Concat_4897",
+ "node_Concat_4900",
+ "node_Slice_5020",
+ "node_Slice_5021",
+ "node_Slice_5023",
+ "node_Concat_5025",
+ "node_Concat_5028",
+ "node_Concat_5170",
+ "node_Concat_5174",
+ "node_Slice_5180",
+ "node_Slice_5181",
+ "node_Slice_5183",
+ "node_Concat_5185",
+ "node_Concat_5188",
+ "node_Slice_5308",
+ "node_Slice_5309",
+ "node_Slice_5311",
+ "node_Concat_5313",
+ "node_Concat_5316",
+ "node_Concat_5458",
+ "node_Concat_5462",
+ "node_Slice_5468",
+ "node_Slice_5469",
+ "node_Slice_5471",
+ "node_Concat_5473",
+ "node_Concat_5476",
+ "node_Slice_5596",
+ "node_Slice_5597",
+ "node_Slice_5599",
+ "node_Concat_5601",
+ "node_Concat_5604",
+ "node_Concat_5746",
+ "node_Concat_5750",
+ "node_Slice_5756",
+ "node_Slice_5757",
+ "node_Slice_5759",
+ "node_Concat_5761",
+ "node_Concat_5764",
+ "node_Slice_5884",
+ "node_Slice_5885",
+ "node_Slice_5887",
+ "node_Concat_5889",
+ "node_Concat_5892",
+ "node_Concat_6034",
+ "node_Concat_6038",
+ "node_Slice_6044",
+ "node_Slice_6045",
+ "node_Slice_6047",
+ "node_Concat_6049",
+ "node_Concat_6052",
+ "node_Slice_6172",
+ "node_Slice_6173",
+ "node_Slice_6175",
+ "node_Concat_6177",
+ "node_Concat_6180",
+ "node_Concat_6322",
+ "node_Concat_6326",
+ "node_Slice_6332",
+ "node_Slice_6333",
+ "node_Slice_6335",
+ "node_Concat_6337",
+ "node_Concat_6340",
+ "node_Slice_6460",
+ "node_Slice_6461",
+ "node_Slice_6463",
+ "node_Concat_6465",
+ "node_Concat_6468",
+ "node_Concat_6610",
+ "node_Concat_6614",
+ "node_Slice_6620",
+ "node_Slice_6621",
+ "node_Slice_6623",
+ "node_Concat_6625",
+ "node_Concat_6628",
+ "node_Slice_6748",
+ "node_Slice_6749",
+ "node_Slice_6751",
+ "node_Concat_6753",
+ "node_Concat_6756",
+ "node_Concat_6898",
+ "node_Concat_6902",
+ "node_Slice_6908",
+ "node_Slice_6909",
+ "node_Slice_6911",
+ "node_Concat_6913",
+ "node_Concat_6916",
+ "node_Slice_7036",
+ "node_Slice_7037",
+ "node_Slice_7039",
+ "node_Concat_7041",
+ "node_Concat_7044",
+ "node_Concat_7186",
+ "node_Concat_7190",
+ "node_Slice_7196",
+ "node_Slice_7197",
+ "node_Slice_7199",
+ "node_Concat_7201",
+ "node_Concat_7204",
+ "node_Slice_7324",
+ "node_Slice_7325",
+ "node_Slice_7327",
+ "node_Concat_7329",
+ "node_Concat_7332",
+ "node_Concat_7474",
+ "node_Concat_7478",
+ "node_Slice_7484",
+ "node_Slice_7485",
+ "node_Slice_7487",
+ "node_Concat_7489",
+ "node_Concat_7492",
+ "node_Slice_7612",
+ "node_Slice_7613",
+ "node_Slice_7615",
+ "node_Concat_7617",
+ "node_Concat_7620",
+ "node_Concat_7762",
+ "node_Concat_7766",
+ "node_Slice_7772",
+ "node_Slice_7773",
+ "node_Slice_7775",
+ "node_Concat_7777",
+ "node_Concat_7780",
+ "node_Slice_7900",
+ "node_Slice_7901",
+ "node_Slice_7903",
+ "node_Concat_7905",
+ "node_Concat_7908",
+ "node_Concat_8050",
+ "node_Concat_8054",
+ "node_Slice_8060",
+ "node_Slice_8061",
+ "node_Slice_8063",
+ "node_Concat_8065",
+ "node_Concat_8068",
+ "node_Slice_8188",
+ "node_Slice_8189",
+ "node_Slice_8191",
+ "node_Concat_8193",
+ "node_Concat_8196",
+ "node_Concat_8338",
+ "node_Concat_8342",
+ "node_Slice_8348",
+ "node_Slice_8349",
+ "node_Slice_8351",
+ "node_Concat_8353",
+ "node_Concat_8356",
+ "node_Slice_8476",
+ "node_Slice_8477",
+ "node_Slice_8479",
+ "node_Concat_8481",
+ "node_Concat_8484",
+ "node_Concat_8626",
+ "node_Concat_8630",
+ "node_Slice_8636",
+ "node_Slice_8637",
+ "node_Slice_8639",
+ "node_Concat_8641",
+ "node_Concat_8644",
+ "node_Slice_8764",
+ "node_Slice_8765",
+ "node_Slice_8767",
+ "node_Concat_8769",
+ "node_Concat_8772",
+ "node_Concat_8914",
+ "node_Concat_8918",
+ "node_Slice_8924",
+ "node_Slice_8925",
+ "node_Slice_8927",
+ "node_Concat_8929",
+ "node_Concat_8932",
+ "node_Slice_9052",
+ "node_Slice_9053",
+ "node_Slice_9055",
+ "node_Concat_9057",
+ "node_Concat_9060",
+ "node_Concat_9202",
+ "node_Concat_9206",
+ "node_Slice_9212",
+ "node_Slice_9213",
+ "node_Slice_9215",
+ "node_Concat_9217",
+ "node_Concat_9220",
+ "node_Slice_9340",
+ "node_Slice_9341",
+ "node_Slice_9343",
+ "node_Concat_9345",
+ "node_Concat_9348",
+ "node_Concat_9490",
+ "node_Concat_9494",
+ "node_Slice_9500",
+ "node_Slice_9501",
+ "node_Slice_9503",
+ "node_Concat_9505",
+ "node_Concat_9508",
+ "node_Concat_9676",
+ "node_Concat_9680",
+ "node_Slice_9686",
+ "node_Slice_9687",
+ "node_Slice_9689",
+ "node_Concat_9691",
+ "node_Concat_9694"
+ ],
+ "int64Arithmetic": [
+ "node_add_15333 (Add)",
+ "node_mul_12678 (Mul)",
+ "node_add_15335 (Add)"
+ ],
+ "ops": {
+ "Mul": 732,
+ "Reshape": 582,
+ "Add": 503,
+ "Transpose": 371,
+ "Slice": 350,
+ "Pow": 253,
+ "Reciprocal": 253,
+ "ReduceMean": 253,
+ "Sqrt": 253,
+ "Cast": 247,
+ "Concat": 245,
+ "MatMul": 228,
+ "Gemm": 221,
+ "Split": 143,
+ "Expand": 96,
+ "Shape": 96,
+ "Gelu": 72,
+ "Unsqueeze": 72,
+ "Softmax": 65,
+ "Sub": 12,
+ "Gather": 6,
+ "Tanh": 6,
+ "Range": 5,
+ "Where": 5,
+ "ScatterND": 4,
+ "Squeeze": 4,
+ "Clip": 3,
+ "ReduceSum": 3,
+ "Cos": 2,
+ "Log": 2,
+ "Sin": 2,
+ "Div": 1,
+ "IsNaN": 1,
+ "Not": 1
+ },
+ "inputs": [
+ {
+ "name": "x",
+ "type": "float32",
+ "shape": [
+ "R",
+ "C"
+ ]
+ },
+ {
+ "name": "y_context",
+ "type": "int32",
+ "shape": [
+ "Rc"
+ ]
+ },
+ {
+ "name": "categorical_mask",
+ "type": "bool",
+ "shape": [
+ "C"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "logits",
+ "type": "float32",
+ "shape": [
+ "Rq",
+ 10
+ ]
+ }
+ ]
+ },
+ {
+ "size": "medium",
+ "task": "regressor",
+ "variant": "f16w",
+ "file": "medium-regressor.f16w.onnx",
+ "bytes": 125916284,
+ "sha256": "e778f2da3b522690016020edaf7dfea3683b7cfbecce67990864a2cd2d75a5e9",
+ "largestInitializerBytes": 4091904,
+ "weightBytesFp32": 250019124,
+ "nodes": 5091,
+ "cpuNodes": [
+ "node_sym_size_int_166",
+ "node_sym_size_int_167",
+ "node_sym_size_int_168",
+ "node_add_15339",
+ "node_Concat_271",
+ "node_Gather_284",
+ "node_Range_285",
+ "node_Unsqueeze_286",
+ "node_Slice_288",
+ "node_Unsqueeze_289",
+ "node_Gather_310",
+ "node_Range_311",
+ "node_Slice_315",
+ "node_Unsqueeze_316",
+ "node_Concat_326",
+ "node_Concat_359",
+ "node_Concat_389",
+ "node_Concat_408",
+ "node_Slice_432",
+ "node_Slice_433",
+ "node_Slice_435",
+ "node_Concat_437",
+ "node_Concat_440",
+ "node_Concat_487",
+ "node_Concat_519",
+ "node_Slice_555",
+ "node_Slice_556",
+ "node_Slice_558",
+ "node_Concat_560",
+ "node_Concat_563",
+ "node_Concat_582",
+ "node_Reshape_607",
+ "node_Concat_610",
+ "node_Concat_677",
+ "node_Slice_713",
+ "node_Slice_714",
+ "node_Slice_716",
+ "node_Concat_718",
+ "node_Concat_721",
+ "node_Concat_740",
+ "node_Concat_753",
+ "node_Slice_851",
+ "node_Slice_852",
+ "node_Slice_854",
+ "node_Concat_856",
+ "node_Concat_859",
+ "node_Slice_974",
+ "node_Slice_975",
+ "node_Slice_977",
+ "node_Concat_979",
+ "node_Concat_982",
+ "node_Slice_1131",
+ "node_Slice_1132",
+ "node_Slice_1134",
+ "node_Concat_1136",
+ "node_Concat_1139",
+ "node_Slice_1269",
+ "node_Slice_1270",
+ "node_Slice_1272",
+ "node_Concat_1274",
+ "node_Concat_1277",
+ "node_Slice_1392",
+ "node_Slice_1393",
+ "node_Slice_1395",
+ "node_Concat_1397",
+ "node_Concat_1400",
+ "node_Slice_1549",
+ "node_Slice_1550",
+ "node_Slice_1552",
+ "node_Concat_1554",
+ "node_Concat_1557",
+ "node_Slice_1687",
+ "node_Slice_1688",
+ "node_Slice_1690",
+ "node_Concat_1692",
+ "node_Concat_1695",
+ "node_Slice_1810",
+ "node_Slice_1811",
+ "node_Slice_1813",
+ "node_Concat_1815",
+ "node_Concat_1818",
+ "node_Slice_1967",
+ "node_Slice_1968",
+ "node_Slice_1970",
+ "node_Concat_1972",
+ "node_Concat_1975",
+ "node_Slice_2105",
+ "node_Slice_2106",
+ "node_Slice_2108",
+ "node_Concat_2110",
+ "node_Concat_2113",
+ "node_Slice_2228",
+ "node_Slice_2229",
+ "node_Slice_2231",
+ "node_Concat_2233",
+ "node_Concat_2236",
+ "node_Slice_2385",
+ "node_Slice_2386",
+ "node_Slice_2388",
+ "node_Concat_2390",
+ "node_Concat_2393",
+ "node_Slice_2523",
+ "node_Slice_2524",
+ "node_Slice_2526",
+ "node_Concat_2528",
+ "node_Concat_2531",
+ "node_Slice_2646",
+ "node_Slice_2647",
+ "node_Slice_2649",
+ "node_Concat_2651",
+ "node_Concat_2654",
+ "node_Slice_2813",
+ "node_Slice_2814",
+ "node_Slice_2816",
+ "node_Concat_2818",
+ "node_Concat_2821",
+ "node_Gather_2869",
+ "node_Range_2870",
+ "node_Slice_2873",
+ "node_Unsqueeze_2874",
+ "node_Gather_2895",
+ "node_Range_2896",
+ "node_Slice_2900",
+ "node_Unsqueeze_2901",
+ "node_Slice_3002",
+ "node_Slice_3003",
+ "node_Slice_3005",
+ "node_Concat_3007",
+ "node_Concat_3010",
+ "node_mul_12680",
+ "node_add_15341",
+ "node_Reshape_3103",
+ "node_Concat_3106",
+ "node_Concat_3154",
+ "node_Concat_3158",
+ "node_Slice_3164",
+ "node_Slice_3165",
+ "node_Slice_3167",
+ "node_Concat_3169",
+ "node_Concat_3172",
+ "node_Slice_3292",
+ "node_Slice_3293",
+ "node_Slice_3295",
+ "node_Concat_3297",
+ "node_Concat_3300",
+ "node_Concat_3442",
+ "node_Concat_3446",
+ "node_Slice_3452",
+ "node_Slice_3453",
+ "node_Slice_3455",
+ "node_Concat_3457",
+ "node_Concat_3460",
+ "node_Slice_3580",
+ "node_Slice_3581",
+ "node_Slice_3583",
+ "node_Concat_3585",
+ "node_Concat_3588",
+ "node_Concat_3730",
+ "node_Concat_3734",
+ "node_Slice_3740",
+ "node_Slice_3741",
+ "node_Slice_3743",
+ "node_Concat_3745",
+ "node_Concat_3748",
+ "node_Slice_3868",
+ "node_Slice_3869",
+ "node_Slice_3871",
+ "node_Concat_3873",
+ "node_Concat_3876",
+ "node_Concat_4018",
+ "node_Concat_4022",
+ "node_Slice_4028",
+ "node_Slice_4029",
+ "node_Slice_4031",
+ "node_Concat_4033",
+ "node_Concat_4036",
+ "node_Slice_4156",
+ "node_Slice_4157",
+ "node_Slice_4159",
+ "node_Concat_4161",
+ "node_Concat_4164",
+ "node_Concat_4306",
+ "node_Concat_4310",
+ "node_Slice_4316",
+ "node_Slice_4317",
+ "node_Slice_4319",
+ "node_Concat_4321",
+ "node_Concat_4324",
+ "node_Slice_4444",
+ "node_Slice_4445",
+ "node_Slice_4447",
+ "node_Concat_4449",
+ "node_Concat_4452",
+ "node_Concat_4594",
+ "node_Concat_4598",
+ "node_Slice_4604",
+ "node_Slice_4605",
+ "node_Slice_4607",
+ "node_Concat_4609",
+ "node_Concat_4612",
+ "node_Slice_4732",
+ "node_Slice_4733",
+ "node_Slice_4735",
+ "node_Concat_4737",
+ "node_Concat_4740",
+ "node_Concat_4882",
+ "node_Concat_4886",
+ "node_Slice_4892",
+ "node_Slice_4893",
+ "node_Slice_4895",
+ "node_Concat_4897",
+ "node_Concat_4900",
+ "node_Slice_5020",
+ "node_Slice_5021",
+ "node_Slice_5023",
+ "node_Concat_5025",
+ "node_Concat_5028",
+ "node_Concat_5170",
+ "node_Concat_5174",
+ "node_Slice_5180",
+ "node_Slice_5181",
+ "node_Slice_5183",
+ "node_Concat_5185",
+ "node_Concat_5188",
+ "node_Slice_5308",
+ "node_Slice_5309",
+ "node_Slice_5311",
+ "node_Concat_5313",
+ "node_Concat_5316",
+ "node_Concat_5458",
+ "node_Concat_5462",
+ "node_Slice_5468",
+ "node_Slice_5469",
+ "node_Slice_5471",
+ "node_Concat_5473",
+ "node_Concat_5476",
+ "node_Slice_5596",
+ "node_Slice_5597",
+ "node_Slice_5599",
+ "node_Concat_5601",
+ "node_Concat_5604",
+ "node_Concat_5746",
+ "node_Concat_5750",
+ "node_Slice_5756",
+ "node_Slice_5757",
+ "node_Slice_5759",
+ "node_Concat_5761",
+ "node_Concat_5764",
+ "node_Slice_5884",
+ "node_Slice_5885",
+ "node_Slice_5887",
+ "node_Concat_5889",
+ "node_Concat_5892",
+ "node_Concat_6034",
+ "node_Concat_6038",
+ "node_Slice_6044",
+ "node_Slice_6045",
+ "node_Slice_6047",
+ "node_Concat_6049",
+ "node_Concat_6052",
+ "node_Slice_6172",
+ "node_Slice_6173",
+ "node_Slice_6175",
+ "node_Concat_6177",
+ "node_Concat_6180",
+ "node_Concat_6322",
+ "node_Concat_6326",
+ "node_Slice_6332",
+ "node_Slice_6333",
+ "node_Slice_6335",
+ "node_Concat_6337",
+ "node_Concat_6340",
+ "node_Slice_6460",
+ "node_Slice_6461",
+ "node_Slice_6463",
+ "node_Concat_6465",
+ "node_Concat_6468",
+ "node_Concat_6610",
+ "node_Concat_6614",
+ "node_Slice_6620",
+ "node_Slice_6621",
+ "node_Slice_6623",
+ "node_Concat_6625",
+ "node_Concat_6628",
+ "node_Slice_6748",
+ "node_Slice_6749",
+ "node_Slice_6751",
+ "node_Concat_6753",
+ "node_Concat_6756",
+ "node_Concat_6898",
+ "node_Concat_6902",
+ "node_Slice_6908",
+ "node_Slice_6909",
+ "node_Slice_6911",
+ "node_Concat_6913",
+ "node_Concat_6916",
+ "node_Slice_7036",
+ "node_Slice_7037",
+ "node_Slice_7039",
+ "node_Concat_7041",
+ "node_Concat_7044",
+ "node_Concat_7186",
+ "node_Concat_7190",
+ "node_Slice_7196",
+ "node_Slice_7197",
+ "node_Slice_7199",
+ "node_Concat_7201",
+ "node_Concat_7204",
+ "node_Slice_7324",
+ "node_Slice_7325",
+ "node_Slice_7327",
+ "node_Concat_7329",
+ "node_Concat_7332",
+ "node_Concat_7474",
+ "node_Concat_7478",
+ "node_Slice_7484",
+ "node_Slice_7485",
+ "node_Slice_7487",
+ "node_Concat_7489",
+ "node_Concat_7492",
+ "node_Slice_7612",
+ "node_Slice_7613",
+ "node_Slice_7615",
+ "node_Concat_7617",
+ "node_Concat_7620",
+ "node_Concat_7762",
+ "node_Concat_7766",
+ "node_Slice_7772",
+ "node_Slice_7773",
+ "node_Slice_7775",
+ "node_Concat_7777",
+ "node_Concat_7780",
+ "node_Slice_7900",
+ "node_Slice_7901",
+ "node_Slice_7903",
+ "node_Concat_7905",
+ "node_Concat_7908",
+ "node_Concat_8050",
+ "node_Concat_8054",
+ "node_Slice_8060",
+ "node_Slice_8061",
+ "node_Slice_8063",
+ "node_Concat_8065",
+ "node_Concat_8068",
+ "node_Slice_8188",
+ "node_Slice_8189",
+ "node_Slice_8191",
+ "node_Concat_8193",
+ "node_Concat_8196",
+ "node_Concat_8338",
+ "node_Concat_8342",
+ "node_Slice_8348",
+ "node_Slice_8349",
+ "node_Slice_8351",
+ "node_Concat_8353",
+ "node_Concat_8356",
+ "node_Slice_8476",
+ "node_Slice_8477",
+ "node_Slice_8479",
+ "node_Concat_8481",
+ "node_Concat_8484",
+ "node_Concat_8626",
+ "node_Concat_8630",
+ "node_Slice_8636",
+ "node_Slice_8637",
+ "node_Slice_8639",
+ "node_Concat_8641",
+ "node_Concat_8644",
+ "node_Slice_8764",
+ "node_Slice_8765",
+ "node_Slice_8767",
+ "node_Concat_8769",
+ "node_Concat_8772",
+ "node_Concat_8914",
+ "node_Concat_8918",
+ "node_Slice_8924",
+ "node_Slice_8925",
+ "node_Slice_8927",
+ "node_Concat_8929",
+ "node_Concat_8932",
+ "node_Slice_9052",
+ "node_Slice_9053",
+ "node_Slice_9055",
+ "node_Concat_9057",
+ "node_Concat_9060",
+ "node_Concat_9202",
+ "node_Concat_9206",
+ "node_Slice_9212",
+ "node_Slice_9213",
+ "node_Slice_9215",
+ "node_Concat_9217",
+ "node_Concat_9220",
+ "node_Slice_9340",
+ "node_Slice_9341",
+ "node_Slice_9343",
+ "node_Concat_9345",
+ "node_Concat_9348",
+ "node_Concat_9490",
+ "node_Concat_9494",
+ "node_Slice_9500",
+ "node_Slice_9501",
+ "node_Slice_9503",
+ "node_Concat_9505",
+ "node_Concat_9508",
+ "node_Concat_9676",
+ "node_Concat_9680",
+ "node_Slice_9686",
+ "node_Slice_9687",
+ "node_Slice_9689",
+ "node_Concat_9691",
+ "node_Concat_9694"
+ ],
+ "int64Arithmetic": [
+ "node_add_15339 (Add)",
+ "node_mul_12680 (Mul)",
+ "node_add_15341 (Add)"
+ ],
+ "ops": {
+ "Mul": 732,
+ "Reshape": 582,
+ "Add": 503,
+ "Transpose": 371,
+ "Slice": 350,
+ "Pow": 253,
+ "Reciprocal": 253,
+ "ReduceMean": 253,
+ "Sqrt": 253,
+ "Cast": 245,
+ "Concat": 245,
+ "MatMul": 228,
+ "Gemm": 223,
+ "Split": 143,
+ "Expand": 96,
+ "Shape": 96,
+ "Unsqueeze": 73,
+ "Gelu": 72,
+ "Softmax": 65,
+ "Sub": 12,
+ "Tanh": 6,
+ "Range": 5,
+ "Where": 5,
+ "Gather": 4,
+ "ScatterND": 4,
+ "Squeeze": 4,
+ "Clip": 3,
+ "ReduceSum": 3,
+ "Cos": 2,
+ "Log": 2,
+ "Sin": 2,
+ "Div": 1,
+ "IsNaN": 1,
+ "Not": 1
+ },
+ "inputs": [
+ {
+ "name": "x",
+ "type": "float32",
+ "shape": [
+ "R",
+ "C"
+ ]
+ },
+ {
+ "name": "y_context",
+ "type": "float32",
+ "shape": [
+ "Rc"
+ ]
+ },
+ {
+ "name": "categorical_mask",
+ "type": "bool",
+ "shape": [
+ "C"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "logits",
+ "type": "float32",
+ "shape": [
+ "Rq",
+ 999
+ ]
+ }
+ ]
+ },
+ {
+ "size": "large",
+ "task": "classifier",
+ "variant": "f16w",
+ "file": "large-classifier.f16w.onnx",
+ "bytes": 428161256,
+ "sha256": "d832eee622716e42ef30c33e76e0cb49b09298cacddfc28f7d315a4f4f7a8cb9",
+ "largestInitializerBytes": 12582912,
+ "weightBytesFp32": 854772920,
+ "nodes": 5239,
+ "cpuNodes": [
+ "node_sym_size_int_166",
+ "node_sym_size_int_167",
+ "node_sym_size_int_168",
+ "node_add_15330",
+ "node_Concat_271",
+ "node_Gather_284",
+ "node_Range_285",
+ "node_Unsqueeze_286",
+ "node_Slice_288",
+ "node_Unsqueeze_289",
+ "node_Gather_310",
+ "node_Range_311",
+ "node_Slice_315",
+ "node_Unsqueeze_316",
+ "node_Concat_326",
+ "node_Concat_359",
+ "node_Concat_389",
+ "node_Concat_408",
+ "node_Slice_432",
+ "node_Slice_433",
+ "node_Slice_435",
+ "node_Concat_437",
+ "node_Concat_440",
+ "node_Concat_487",
+ "node_Concat_519",
+ "node_Slice_555",
+ "node_Slice_556",
+ "node_Slice_558",
+ "node_Concat_560",
+ "node_Concat_563",
+ "node_Concat_582",
+ "node_Reshape_607",
+ "node_Concat_610",
+ "node_Concat_677",
+ "node_Slice_713",
+ "node_Slice_714",
+ "node_Slice_716",
+ "node_Concat_718",
+ "node_Concat_721",
+ "node_Concat_740",
+ "node_Concat_753",
+ "node_Slice_851",
+ "node_Slice_852",
+ "node_Slice_854",
+ "node_Concat_856",
+ "node_Concat_859",
+ "node_Slice_974",
+ "node_Slice_975",
+ "node_Slice_977",
+ "node_Concat_979",
+ "node_Concat_982",
+ "node_Slice_1131",
+ "node_Slice_1132",
+ "node_Slice_1134",
+ "node_Concat_1136",
+ "node_Concat_1139",
+ "node_Slice_1269",
+ "node_Slice_1270",
+ "node_Slice_1272",
+ "node_Concat_1274",
+ "node_Concat_1277",
+ "node_Slice_1392",
+ "node_Slice_1393",
+ "node_Slice_1395",
+ "node_Concat_1397",
+ "node_Concat_1400",
+ "node_Slice_1549",
+ "node_Slice_1550",
+ "node_Slice_1552",
+ "node_Concat_1554",
+ "node_Concat_1557",
+ "node_Slice_1687",
+ "node_Slice_1688",
+ "node_Slice_1690",
+ "node_Concat_1692",
+ "node_Concat_1695",
+ "node_Slice_1810",
+ "node_Slice_1811",
+ "node_Slice_1813",
+ "node_Concat_1815",
+ "node_Concat_1818",
+ "node_Slice_1967",
+ "node_Slice_1968",
+ "node_Slice_1970",
+ "node_Concat_1972",
+ "node_Concat_1975",
+ "node_Slice_2105",
+ "node_Slice_2106",
+ "node_Slice_2108",
+ "node_Concat_2110",
+ "node_Concat_2113",
+ "node_Slice_2228",
+ "node_Slice_2229",
+ "node_Slice_2231",
+ "node_Concat_2233",
+ "node_Concat_2236",
+ "node_Slice_2385",
+ "node_Slice_2386",
+ "node_Slice_2388",
+ "node_Concat_2390",
+ "node_Concat_2393",
+ "node_Slice_2523",
+ "node_Slice_2524",
+ "node_Slice_2526",
+ "node_Concat_2528",
+ "node_Concat_2531",
+ "node_Slice_2646",
+ "node_Slice_2647",
+ "node_Slice_2649",
+ "node_Concat_2651",
+ "node_Concat_2654",
+ "node_Slice_2813",
+ "node_Slice_2814",
+ "node_Slice_2816",
+ "node_Concat_2818",
+ "node_Concat_2821",
+ "node_Gather_2873",
+ "node_Range_2874",
+ "node_Slice_2877",
+ "node_Unsqueeze_2878",
+ "node_Gather_2916",
+ "node_Range_2917",
+ "node_Slice_2921",
+ "node_Unsqueeze_2922",
+ "node_Slice_3035",
+ "node_Slice_3036",
+ "node_Slice_3038",
+ "node_Concat_3040",
+ "node_Concat_3043",
+ "node_mul_12676",
+ "node_add_15332",
+ "node_Reshape_3140",
+ "node_Concat_3143",
+ "node_Concat_3191",
+ "node_Concat_3195",
+ "node_Slice_3201",
+ "node_Slice_3202",
+ "node_Slice_3204",
+ "node_Concat_3206",
+ "node_Concat_3209",
+ "node_Slice_3329",
+ "node_Slice_3330",
+ "node_Slice_3332",
+ "node_Concat_3334",
+ "node_Concat_3337",
+ "node_Concat_3479",
+ "node_Concat_3483",
+ "node_Slice_3489",
+ "node_Slice_3490",
+ "node_Slice_3492",
+ "node_Concat_3494",
+ "node_Concat_3497",
+ "node_Slice_3617",
+ "node_Slice_3618",
+ "node_Slice_3620",
+ "node_Concat_3622",
+ "node_Concat_3625",
+ "node_Concat_3767",
+ "node_Concat_3771",
+ "node_Slice_3777",
+ "node_Slice_3778",
+ "node_Slice_3780",
+ "node_Concat_3782",
+ "node_Concat_3785",
+ "node_Slice_3905",
+ "node_Slice_3906",
+ "node_Slice_3908",
+ "node_Concat_3910",
+ "node_Concat_3913",
+ "node_Concat_4055",
+ "node_Concat_4059",
+ "node_Slice_4065",
+ "node_Slice_4066",
+ "node_Slice_4068",
+ "node_Concat_4070",
+ "node_Concat_4073",
+ "node_Slice_4193",
+ "node_Slice_4194",
+ "node_Slice_4196",
+ "node_Concat_4198",
+ "node_Concat_4201",
+ "node_Concat_4343",
+ "node_Concat_4347",
+ "node_Slice_4353",
+ "node_Slice_4354",
+ "node_Slice_4356",
+ "node_Concat_4358",
+ "node_Concat_4361",
+ "node_Slice_4481",
+ "node_Slice_4482",
+ "node_Slice_4484",
+ "node_Concat_4486",
+ "node_Concat_4489",
+ "node_Concat_4631",
+ "node_Concat_4635",
+ "node_Slice_4641",
+ "node_Slice_4642",
+ "node_Slice_4644",
+ "node_Concat_4646",
+ "node_Concat_4649",
+ "node_Slice_4769",
+ "node_Slice_4770",
+ "node_Slice_4772",
+ "node_Concat_4774",
+ "node_Concat_4777",
+ "node_Concat_4919",
+ "node_Concat_4923",
+ "node_Slice_4929",
+ "node_Slice_4930",
+ "node_Slice_4932",
+ "node_Concat_4934",
+ "node_Concat_4937",
+ "node_Slice_5057",
+ "node_Slice_5058",
+ "node_Slice_5060",
+ "node_Concat_5062",
+ "node_Concat_5065",
+ "node_Concat_5207",
+ "node_Concat_5211",
+ "node_Slice_5217",
+ "node_Slice_5218",
+ "node_Slice_5220",
+ "node_Concat_5222",
+ "node_Concat_5225",
+ "node_Slice_5345",
+ "node_Slice_5346",
+ "node_Slice_5348",
+ "node_Concat_5350",
+ "node_Concat_5353",
+ "node_Concat_5495",
+ "node_Concat_5499",
+ "node_Slice_5505",
+ "node_Slice_5506",
+ "node_Slice_5508",
+ "node_Concat_5510",
+ "node_Concat_5513",
+ "node_Slice_5633",
+ "node_Slice_5634",
+ "node_Slice_5636",
+ "node_Concat_5638",
+ "node_Concat_5641",
+ "node_Concat_5783",
+ "node_Concat_5787",
+ "node_Slice_5793",
+ "node_Slice_5794",
+ "node_Slice_5796",
+ "node_Concat_5798",
+ "node_Concat_5801",
+ "node_Slice_5921",
+ "node_Slice_5922",
+ "node_Slice_5924",
+ "node_Concat_5926",
+ "node_Concat_5929",
+ "node_Concat_6071",
+ "node_Concat_6075",
+ "node_Slice_6081",
+ "node_Slice_6082",
+ "node_Slice_6084",
+ "node_Concat_6086",
+ "node_Concat_6089",
+ "node_Slice_6209",
+ "node_Slice_6210",
+ "node_Slice_6212",
+ "node_Concat_6214",
+ "node_Concat_6217",
+ "node_Concat_6359",
+ "node_Concat_6363",
+ "node_Slice_6369",
+ "node_Slice_6370",
+ "node_Slice_6372",
+ "node_Concat_6374",
+ "node_Concat_6377",
+ "node_Slice_6497",
+ "node_Slice_6498",
+ "node_Slice_6500",
+ "node_Concat_6502",
+ "node_Concat_6505",
+ "node_Concat_6647",
+ "node_Concat_6651",
+ "node_Slice_6657",
+ "node_Slice_6658",
+ "node_Slice_6660",
+ "node_Concat_6662",
+ "node_Concat_6665",
+ "node_Slice_6785",
+ "node_Slice_6786",
+ "node_Slice_6788",
+ "node_Concat_6790",
+ "node_Concat_6793",
+ "node_Concat_6935",
+ "node_Concat_6939",
+ "node_Slice_6945",
+ "node_Slice_6946",
+ "node_Slice_6948",
+ "node_Concat_6950",
+ "node_Concat_6953",
+ "node_Slice_7073",
+ "node_Slice_7074",
+ "node_Slice_7076",
+ "node_Concat_7078",
+ "node_Concat_7081",
+ "node_Concat_7223",
+ "node_Concat_7227",
+ "node_Slice_7233",
+ "node_Slice_7234",
+ "node_Slice_7236",
+ "node_Concat_7238",
+ "node_Concat_7241",
+ "node_Slice_7361",
+ "node_Slice_7362",
+ "node_Slice_7364",
+ "node_Concat_7366",
+ "node_Concat_7369",
+ "node_Concat_7511",
+ "node_Concat_7515",
+ "node_Slice_7521",
+ "node_Slice_7522",
+ "node_Slice_7524",
+ "node_Concat_7526",
+ "node_Concat_7529",
+ "node_Slice_7649",
+ "node_Slice_7650",
+ "node_Slice_7652",
+ "node_Concat_7654",
+ "node_Concat_7657",
+ "node_Concat_7799",
+ "node_Concat_7803",
+ "node_Slice_7809",
+ "node_Slice_7810",
+ "node_Slice_7812",
+ "node_Concat_7814",
+ "node_Concat_7817",
+ "node_Slice_7937",
+ "node_Slice_7938",
+ "node_Slice_7940",
+ "node_Concat_7942",
+ "node_Concat_7945",
+ "node_Concat_8087",
+ "node_Concat_8091",
+ "node_Slice_8097",
+ "node_Slice_8098",
+ "node_Slice_8100",
+ "node_Concat_8102",
+ "node_Concat_8105",
+ "node_Slice_8225",
+ "node_Slice_8226",
+ "node_Slice_8228",
+ "node_Concat_8230",
+ "node_Concat_8233",
+ "node_Concat_8375",
+ "node_Concat_8379",
+ "node_Slice_8385",
+ "node_Slice_8386",
+ "node_Slice_8388",
+ "node_Concat_8390",
+ "node_Concat_8393",
+ "node_Slice_8513",
+ "node_Slice_8514",
+ "node_Slice_8516",
+ "node_Concat_8518",
+ "node_Concat_8521",
+ "node_Concat_8663",
+ "node_Concat_8667",
+ "node_Slice_8673",
+ "node_Slice_8674",
+ "node_Slice_8676",
+ "node_Concat_8678",
+ "node_Concat_8681",
+ "node_Slice_8801",
+ "node_Slice_8802",
+ "node_Slice_8804",
+ "node_Concat_8806",
+ "node_Concat_8809",
+ "node_Concat_8951",
+ "node_Concat_8955",
+ "node_Slice_8961",
+ "node_Slice_8962",
+ "node_Slice_8964",
+ "node_Concat_8966",
+ "node_Concat_8969",
+ "node_Slice_9089",
+ "node_Slice_9090",
+ "node_Slice_9092",
+ "node_Concat_9094",
+ "node_Concat_9097",
+ "node_Concat_9239",
+ "node_Concat_9243",
+ "node_Slice_9249",
+ "node_Slice_9250",
+ "node_Slice_9252",
+ "node_Concat_9254",
+ "node_Concat_9257",
+ "node_Slice_9377",
+ "node_Slice_9378",
+ "node_Slice_9380",
+ "node_Concat_9382",
+ "node_Concat_9385",
+ "node_Concat_9527",
+ "node_Concat_9531",
+ "node_Slice_9537",
+ "node_Slice_9538",
+ "node_Slice_9540",
+ "node_Concat_9542",
+ "node_Concat_9545",
+ "node_Concat_9713",
+ "node_Concat_9717",
+ "node_Slice_9723",
+ "node_Slice_9724",
+ "node_Slice_9726",
+ "node_Concat_9728",
+ "node_Concat_9731"
+ ],
+ "int64Arithmetic": [
+ "node_add_15330 (Add)",
+ "node_mul_12676 (Mul)",
+ "node_add_15332 (Add)"
+ ],
+ "ops": {
+ "Mul": 732,
+ "Reshape": 586,
+ "Add": 503,
+ "Cast": 391,
+ "Transpose": 371,
+ "Slice": 350,
+ "Pow": 253,
+ "Reciprocal": 253,
+ "ReduceMean": 253,
+ "Sqrt": 253,
+ "Concat": 245,
+ "MatMul": 228,
+ "Gemm": 220,
+ "Split": 143,
+ "Expand": 96,
+ "Shape": 96,
+ "Gelu": 72,
+ "Unsqueeze": 72,
+ "Softmax": 65,
+ "Sub": 12,
+ "Gather": 6,
+ "Tanh": 6,
+ "Range": 5,
+ "Where": 5,
+ "ScatterND": 4,
+ "Squeeze": 4,
+ "Clip": 3,
+ "ReduceSum": 3,
+ "Cos": 2,
+ "Log": 2,
+ "Sin": 2,
+ "Div": 1,
+ "IsNaN": 1,
+ "Not": 1
+ },
+ "inputs": [
+ {
+ "name": "x",
+ "type": "float32",
+ "shape": [
+ "R",
+ "C"
+ ]
+ },
+ {
+ "name": "y_context",
+ "type": "int32",
+ "shape": [
+ "Rc"
+ ]
+ },
+ {
+ "name": "categorical_mask",
+ "type": "bool",
+ "shape": [
+ "C"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "logits",
+ "type": "float32",
+ "shape": [
+ "Rq",
+ 10
+ ]
+ }
+ ]
+ },
+ {
+ "size": "large",
+ "task": "regressor",
+ "variant": "f16w",
+ "file": "large-regressor.f16w.onnx",
+ "bytes": 432193733,
+ "sha256": "6b144ae2177a19b16e0db9232d66f7be242cee476d4f44cca54dad89915d62d0",
+ "largestInitializerBytes": 12582912,
+ "weightBytesFp32": 862832684,
+ "nodes": 5239,
+ "cpuNodes": [
+ "node_sym_size_int_166",
+ "node_sym_size_int_167",
+ "node_sym_size_int_168",
+ "node_add_15336",
+ "node_Concat_271",
+ "node_Gather_284",
+ "node_Range_285",
+ "node_Unsqueeze_286",
+ "node_Slice_288",
+ "node_Unsqueeze_289",
+ "node_Gather_310",
+ "node_Range_311",
+ "node_Slice_315",
+ "node_Unsqueeze_316",
+ "node_Concat_326",
+ "node_Concat_359",
+ "node_Concat_389",
+ "node_Concat_408",
+ "node_Slice_432",
+ "node_Slice_433",
+ "node_Slice_435",
+ "node_Concat_437",
+ "node_Concat_440",
+ "node_Concat_487",
+ "node_Concat_519",
+ "node_Slice_555",
+ "node_Slice_556",
+ "node_Slice_558",
+ "node_Concat_560",
+ "node_Concat_563",
+ "node_Concat_582",
+ "node_Reshape_607",
+ "node_Concat_610",
+ "node_Concat_677",
+ "node_Slice_713",
+ "node_Slice_714",
+ "node_Slice_716",
+ "node_Concat_718",
+ "node_Concat_721",
+ "node_Concat_740",
+ "node_Concat_753",
+ "node_Slice_851",
+ "node_Slice_852",
+ "node_Slice_854",
+ "node_Concat_856",
+ "node_Concat_859",
+ "node_Slice_974",
+ "node_Slice_975",
+ "node_Slice_977",
+ "node_Concat_979",
+ "node_Concat_982",
+ "node_Slice_1131",
+ "node_Slice_1132",
+ "node_Slice_1134",
+ "node_Concat_1136",
+ "node_Concat_1139",
+ "node_Slice_1269",
+ "node_Slice_1270",
+ "node_Slice_1272",
+ "node_Concat_1274",
+ "node_Concat_1277",
+ "node_Slice_1392",
+ "node_Slice_1393",
+ "node_Slice_1395",
+ "node_Concat_1397",
+ "node_Concat_1400",
+ "node_Slice_1549",
+ "node_Slice_1550",
+ "node_Slice_1552",
+ "node_Concat_1554",
+ "node_Concat_1557",
+ "node_Slice_1687",
+ "node_Slice_1688",
+ "node_Slice_1690",
+ "node_Concat_1692",
+ "node_Concat_1695",
+ "node_Slice_1810",
+ "node_Slice_1811",
+ "node_Slice_1813",
+ "node_Concat_1815",
+ "node_Concat_1818",
+ "node_Slice_1967",
+ "node_Slice_1968",
+ "node_Slice_1970",
+ "node_Concat_1972",
+ "node_Concat_1975",
+ "node_Slice_2105",
+ "node_Slice_2106",
+ "node_Slice_2108",
+ "node_Concat_2110",
+ "node_Concat_2113",
+ "node_Slice_2228",
+ "node_Slice_2229",
+ "node_Slice_2231",
+ "node_Concat_2233",
+ "node_Concat_2236",
+ "node_Slice_2385",
+ "node_Slice_2386",
+ "node_Slice_2388",
+ "node_Concat_2390",
+ "node_Concat_2393",
+ "node_Slice_2523",
+ "node_Slice_2524",
+ "node_Slice_2526",
+ "node_Concat_2528",
+ "node_Concat_2531",
+ "node_Slice_2646",
+ "node_Slice_2647",
+ "node_Slice_2649",
+ "node_Concat_2651",
+ "node_Concat_2654",
+ "node_Slice_2813",
+ "node_Slice_2814",
+ "node_Slice_2816",
+ "node_Concat_2818",
+ "node_Concat_2821",
+ "node_Gather_2873",
+ "node_Range_2874",
+ "node_Slice_2877",
+ "node_Unsqueeze_2878",
+ "node_Gather_2916",
+ "node_Range_2917",
+ "node_Slice_2921",
+ "node_Unsqueeze_2922",
+ "node_Slice_3035",
+ "node_Slice_3036",
+ "node_Slice_3038",
+ "node_Concat_3040",
+ "node_Concat_3043",
+ "node_mul_12678",
+ "node_add_15338",
+ "node_Reshape_3140",
+ "node_Concat_3143",
+ "node_Concat_3191",
+ "node_Concat_3195",
+ "node_Slice_3201",
+ "node_Slice_3202",
+ "node_Slice_3204",
+ "node_Concat_3206",
+ "node_Concat_3209",
+ "node_Slice_3329",
+ "node_Slice_3330",
+ "node_Slice_3332",
+ "node_Concat_3334",
+ "node_Concat_3337",
+ "node_Concat_3479",
+ "node_Concat_3483",
+ "node_Slice_3489",
+ "node_Slice_3490",
+ "node_Slice_3492",
+ "node_Concat_3494",
+ "node_Concat_3497",
+ "node_Slice_3617",
+ "node_Slice_3618",
+ "node_Slice_3620",
+ "node_Concat_3622",
+ "node_Concat_3625",
+ "node_Concat_3767",
+ "node_Concat_3771",
+ "node_Slice_3777",
+ "node_Slice_3778",
+ "node_Slice_3780",
+ "node_Concat_3782",
+ "node_Concat_3785",
+ "node_Slice_3905",
+ "node_Slice_3906",
+ "node_Slice_3908",
+ "node_Concat_3910",
+ "node_Concat_3913",
+ "node_Concat_4055",
+ "node_Concat_4059",
+ "node_Slice_4065",
+ "node_Slice_4066",
+ "node_Slice_4068",
+ "node_Concat_4070",
+ "node_Concat_4073",
+ "node_Slice_4193",
+ "node_Slice_4194",
+ "node_Slice_4196",
+ "node_Concat_4198",
+ "node_Concat_4201",
+ "node_Concat_4343",
+ "node_Concat_4347",
+ "node_Slice_4353",
+ "node_Slice_4354",
+ "node_Slice_4356",
+ "node_Concat_4358",
+ "node_Concat_4361",
+ "node_Slice_4481",
+ "node_Slice_4482",
+ "node_Slice_4484",
+ "node_Concat_4486",
+ "node_Concat_4489",
+ "node_Concat_4631",
+ "node_Concat_4635",
+ "node_Slice_4641",
+ "node_Slice_4642",
+ "node_Slice_4644",
+ "node_Concat_4646",
+ "node_Concat_4649",
+ "node_Slice_4769",
+ "node_Slice_4770",
+ "node_Slice_4772",
+ "node_Concat_4774",
+ "node_Concat_4777",
+ "node_Concat_4919",
+ "node_Concat_4923",
+ "node_Slice_4929",
+ "node_Slice_4930",
+ "node_Slice_4932",
+ "node_Concat_4934",
+ "node_Concat_4937",
+ "node_Slice_5057",
+ "node_Slice_5058",
+ "node_Slice_5060",
+ "node_Concat_5062",
+ "node_Concat_5065",
+ "node_Concat_5207",
+ "node_Concat_5211",
+ "node_Slice_5217",
+ "node_Slice_5218",
+ "node_Slice_5220",
+ "node_Concat_5222",
+ "node_Concat_5225",
+ "node_Slice_5345",
+ "node_Slice_5346",
+ "node_Slice_5348",
+ "node_Concat_5350",
+ "node_Concat_5353",
+ "node_Concat_5495",
+ "node_Concat_5499",
+ "node_Slice_5505",
+ "node_Slice_5506",
+ "node_Slice_5508",
+ "node_Concat_5510",
+ "node_Concat_5513",
+ "node_Slice_5633",
+ "node_Slice_5634",
+ "node_Slice_5636",
+ "node_Concat_5638",
+ "node_Concat_5641",
+ "node_Concat_5783",
+ "node_Concat_5787",
+ "node_Slice_5793",
+ "node_Slice_5794",
+ "node_Slice_5796",
+ "node_Concat_5798",
+ "node_Concat_5801",
+ "node_Slice_5921",
+ "node_Slice_5922",
+ "node_Slice_5924",
+ "node_Concat_5926",
+ "node_Concat_5929",
+ "node_Concat_6071",
+ "node_Concat_6075",
+ "node_Slice_6081",
+ "node_Slice_6082",
+ "node_Slice_6084",
+ "node_Concat_6086",
+ "node_Concat_6089",
+ "node_Slice_6209",
+ "node_Slice_6210",
+ "node_Slice_6212",
+ "node_Concat_6214",
+ "node_Concat_6217",
+ "node_Concat_6359",
+ "node_Concat_6363",
+ "node_Slice_6369",
+ "node_Slice_6370",
+ "node_Slice_6372",
+ "node_Concat_6374",
+ "node_Concat_6377",
+ "node_Slice_6497",
+ "node_Slice_6498",
+ "node_Slice_6500",
+ "node_Concat_6502",
+ "node_Concat_6505",
+ "node_Concat_6647",
+ "node_Concat_6651",
+ "node_Slice_6657",
+ "node_Slice_6658",
+ "node_Slice_6660",
+ "node_Concat_6662",
+ "node_Concat_6665",
+ "node_Slice_6785",
+ "node_Slice_6786",
+ "node_Slice_6788",
+ "node_Concat_6790",
+ "node_Concat_6793",
+ "node_Concat_6935",
+ "node_Concat_6939",
+ "node_Slice_6945",
+ "node_Slice_6946",
+ "node_Slice_6948",
+ "node_Concat_6950",
+ "node_Concat_6953",
+ "node_Slice_7073",
+ "node_Slice_7074",
+ "node_Slice_7076",
+ "node_Concat_7078",
+ "node_Concat_7081",
+ "node_Concat_7223",
+ "node_Concat_7227",
+ "node_Slice_7233",
+ "node_Slice_7234",
+ "node_Slice_7236",
+ "node_Concat_7238",
+ "node_Concat_7241",
+ "node_Slice_7361",
+ "node_Slice_7362",
+ "node_Slice_7364",
+ "node_Concat_7366",
+ "node_Concat_7369",
+ "node_Concat_7511",
+ "node_Concat_7515",
+ "node_Slice_7521",
+ "node_Slice_7522",
+ "node_Slice_7524",
+ "node_Concat_7526",
+ "node_Concat_7529",
+ "node_Slice_7649",
+ "node_Slice_7650",
+ "node_Slice_7652",
+ "node_Concat_7654",
+ "node_Concat_7657",
+ "node_Concat_7799",
+ "node_Concat_7803",
+ "node_Slice_7809",
+ "node_Slice_7810",
+ "node_Slice_7812",
+ "node_Concat_7814",
+ "node_Concat_7817",
+ "node_Slice_7937",
+ "node_Slice_7938",
+ "node_Slice_7940",
+ "node_Concat_7942",
+ "node_Concat_7945",
+ "node_Concat_8087",
+ "node_Concat_8091",
+ "node_Slice_8097",
+ "node_Slice_8098",
+ "node_Slice_8100",
+ "node_Concat_8102",
+ "node_Concat_8105",
+ "node_Slice_8225",
+ "node_Slice_8226",
+ "node_Slice_8228",
+ "node_Concat_8230",
+ "node_Concat_8233",
+ "node_Concat_8375",
+ "node_Concat_8379",
+ "node_Slice_8385",
+ "node_Slice_8386",
+ "node_Slice_8388",
+ "node_Concat_8390",
+ "node_Concat_8393",
+ "node_Slice_8513",
+ "node_Slice_8514",
+ "node_Slice_8516",
+ "node_Concat_8518",
+ "node_Concat_8521",
+ "node_Concat_8663",
+ "node_Concat_8667",
+ "node_Slice_8673",
+ "node_Slice_8674",
+ "node_Slice_8676",
+ "node_Concat_8678",
+ "node_Concat_8681",
+ "node_Slice_8801",
+ "node_Slice_8802",
+ "node_Slice_8804",
+ "node_Concat_8806",
+ "node_Concat_8809",
+ "node_Concat_8951",
+ "node_Concat_8955",
+ "node_Slice_8961",
+ "node_Slice_8962",
+ "node_Slice_8964",
+ "node_Concat_8966",
+ "node_Concat_8969",
+ "node_Slice_9089",
+ "node_Slice_9090",
+ "node_Slice_9092",
+ "node_Concat_9094",
+ "node_Concat_9097",
+ "node_Concat_9239",
+ "node_Concat_9243",
+ "node_Slice_9249",
+ "node_Slice_9250",
+ "node_Slice_9252",
+ "node_Concat_9254",
+ "node_Concat_9257",
+ "node_Slice_9377",
+ "node_Slice_9378",
+ "node_Slice_9380",
+ "node_Concat_9382",
+ "node_Concat_9385",
+ "node_Concat_9527",
+ "node_Concat_9531",
+ "node_Slice_9537",
+ "node_Slice_9538",
+ "node_Slice_9540",
+ "node_Concat_9542",
+ "node_Concat_9545",
+ "node_Concat_9713",
+ "node_Concat_9717",
+ "node_Slice_9723",
+ "node_Slice_9724",
+ "node_Slice_9726",
+ "node_Concat_9728",
+ "node_Concat_9731"
+ ],
+ "int64Arithmetic": [
+ "node_add_15336 (Add)",
+ "node_mul_12678 (Mul)",
+ "node_add_15338 (Add)"
+ ],
+ "ops": {
+ "Mul": 732,
+ "Reshape": 586,
+ "Add": 503,
+ "Cast": 390,
+ "Transpose": 371,
+ "Slice": 350,
+ "Pow": 253,
+ "Reciprocal": 253,
+ "ReduceMean": 253,
+ "Sqrt": 253,
+ "Concat": 245,
+ "MatMul": 228,
+ "Gemm": 222,
+ "Split": 143,
+ "Expand": 96,
+ "Shape": 96,
+ "Unsqueeze": 73,
+ "Gelu": 72,
+ "Softmax": 65,
+ "Sub": 12,
+ "Tanh": 6,
+ "Range": 5,
+ "Where": 5,
+ "Gather": 4,
+ "ScatterND": 4,
+ "Squeeze": 4,
+ "Clip": 3,
+ "ReduceSum": 3,
+ "Cos": 2,
+ "Log": 2,
+ "Sin": 2,
+ "Div": 1,
+ "IsNaN": 1,
+ "Not": 1
+ },
+ "inputs": [
+ {
+ "name": "x",
+ "type": "float32",
+ "shape": [
+ "R",
+ "C"
+ ]
+ },
+ {
+ "name": "y_context",
+ "type": "float32",
+ "shape": [
+ "Rc"
+ ]
+ },
+ {
+ "name": "categorical_mask",
+ "type": "bool",
+ "shape": [
+ "C"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "logits",
+ "type": "float32",
+ "shape": [
+ "Rq",
+ 999
+ ]
+ }
+ ]
+ }
+ ],
+ "parity": {
+ "small-classifier": {
+ "randomShapes": [
+ [
+ 70,
+ 50,
+ 7
+ ],
+ [
+ 30,
+ 29,
+ 1
+ ],
+ [
+ 12,
+ 8,
+ 2
+ ],
+ [
+ 300,
+ 200,
+ 40
+ ],
+ [
+ 260,
+ 256,
+ 3
+ ]
+ ],
+ "fp32": {
+ "randomMaxAbsDiff": 7.724761962890625e-05,
+ "sdmRecipeMaxAbsDiff": 5.1975250244140625e-05
+ },
+ "f16w": {
+ "randomMaxAbsDiff": 0.01811504364013672,
+ "sdmRecipeMaxAbsDiff": 0.019417762756347656
+ },
+ "sdmRecipe": {
+ "members": 4,
+ "shape": [
+ 640,
+ 12
+ ],
+ "contextRows": 540,
+ "categoricalColumns": 2,
+ "missingCells": 372,
+ "largestAbsOutput": 14.519216537475586
+ }
+ },
+ "small-regressor": {
+ "randomShapes": [
+ [
+ 70,
+ 50,
+ 7
+ ],
+ [
+ 30,
+ 29,
+ 1
+ ],
+ [
+ 12,
+ 8,
+ 2
+ ],
+ [
+ 300,
+ 200,
+ 40
+ ],
+ [
+ 260,
+ 256,
+ 3
+ ]
+ ],
+ "fp32": {
+ "randomMaxAbsDiff": 3.0517578125e-05,
+ "sdmRecipeMaxAbsDiff": 2.193450927734375e-05
+ },
+ "f16w": {
+ "randomMaxAbsDiff": 0.017749309539794922,
+ "sdmRecipeMaxAbsDiff": 0.006098270416259766
+ },
+ "sdmRecipe": {
+ "members": 4,
+ "shape": [
+ 640,
+ 12
+ ],
+ "contextRows": 540,
+ "categoricalColumns": 2,
+ "missingCells": 372,
+ "largestAbsOutput": 4.582178115844727
+ }
+ },
+ "medium-classifier": {
+ "randomShapes": [
+ [
+ 70,
+ 50,
+ 7
+ ],
+ [
+ 30,
+ 29,
+ 1
+ ],
+ [
+ 12,
+ 8,
+ 2
+ ],
+ [
+ 300,
+ 200,
+ 40
+ ],
+ [
+ 260,
+ 256,
+ 3
+ ]
+ ],
+ "fp32": {
+ "randomMaxAbsDiff": 9.5367431640625e-05,
+ "sdmRecipeMaxAbsDiff": 4.1961669921875e-05
+ },
+ "f16w": {
+ "randomMaxAbsDiff": 0.018278121948242188,
+ "sdmRecipeMaxAbsDiff": 0.015291213989257812
+ },
+ "sdmRecipe": {
+ "members": 4,
+ "shape": [
+ 640,
+ 12
+ ],
+ "contextRows": 540,
+ "categoricalColumns": 2,
+ "missingCells": 372,
+ "largestAbsOutput": 30.128097534179688
+ }
+ },
+ "medium-regressor": {
+ "randomShapes": [
+ [
+ 70,
+ 50,
+ 7
+ ],
+ [
+ 30,
+ 29,
+ 1
+ ],
+ [
+ 12,
+ 8,
+ 2
+ ],
+ [
+ 300,
+ 200,
+ 40
+ ],
+ [
+ 260,
+ 256,
+ 3
+ ]
+ ],
+ "fp32": {
+ "randomMaxAbsDiff": 1.2159347534179688e-05,
+ "sdmRecipeMaxAbsDiff": 9.119510650634766e-06
+ },
+ "f16w": {
+ "randomMaxAbsDiff": 0.006520748138427734,
+ "sdmRecipeMaxAbsDiff": 0.0013577938079833984
+ },
+ "sdmRecipe": {
+ "members": 4,
+ "shape": [
+ 640,
+ 12
+ ],
+ "contextRows": 540,
+ "categoricalColumns": 2,
+ "missingCells": 372,
+ "largestAbsOutput": 3.672037363052368
+ }
+ },
+ "large-classifier": {
+ "randomShapes": [
+ [
+ 70,
+ 50,
+ 7
+ ],
+ [
+ 30,
+ 29,
+ 1
+ ],
+ [
+ 12,
+ 8,
+ 2
+ ],
+ [
+ 300,
+ 200,
+ 40
+ ],
+ [
+ 260,
+ 256,
+ 3
+ ]
+ ],
+ "fp32": {
+ "randomMaxAbsDiff": 6.29425048828125e-05,
+ "sdmRecipeMaxAbsDiff": 2.4124979972839355e-05
+ },
+ "f16w": {
+ "randomMaxAbsDiff": 0.024433612823486328,
+ "sdmRecipeMaxAbsDiff": 0.010754108428955078
+ },
+ "sdmRecipe": {
+ "members": 4,
+ "shape": [
+ 640,
+ 12
+ ],
+ "contextRows": 540,
+ "categoricalColumns": 2,
+ "missingCells": 372,
+ "largestAbsOutput": 8.254240989685059
+ }
+ },
+ "large-regressor": {
+ "randomShapes": [
+ [
+ 70,
+ 50,
+ 7
+ ],
+ [
+ 30,
+ 29,
+ 1
+ ],
+ [
+ 12,
+ 8,
+ 2
+ ],
+ [
+ 300,
+ 200,
+ 40
+ ],
+ [
+ 260,
+ 256,
+ 3
+ ]
+ ],
+ "fp32": {
+ "randomMaxAbsDiff": 0.00012874603271484375,
+ "sdmRecipeMaxAbsDiff": 6.866455078125e-05
+ },
+ "f16w": {
+ "randomMaxAbsDiff": 0.0042247772216796875,
+ "sdmRecipeMaxAbsDiff": 0.002736330032348633
+ },
+ "sdmRecipe": {
+ "members": 4,
+ "shape": [
+ 640,
+ 12
+ ],
+ "contextRows": 540,
+ "categoricalColumns": 2,
+ "missingCells": 372,
+ "largestAbsOutput": 4.520740985870361
+ }
+ }
+ }
+}
diff --git a/kumo/medium-classifier.f16w.onnx b/kumo/medium-classifier.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..9d7be5d9272b70c3e5126873bffd2971304a369a
--- /dev/null
+++ b/kumo/medium-classifier.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:20217f82c506f01852c6b9da36f5d6adfb7424a441db66bce7b210b420ea04f1
+size 123899156
diff --git a/kumo/medium-regressor.f16w.onnx b/kumo/medium-regressor.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..eb8022914a3ad66fff0c675089276327f332e28d
--- /dev/null
+++ b/kumo/medium-regressor.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e778f2da3b522690016020edaf7dfea3683b7cfbecce67990864a2cd2d75a5e9
+size 125916284
diff --git a/kumo/small-classifier.f16w.onnx b/kumo/small-classifier.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..81661846fff3f26c97ccf5b43b18eda959ff0671
--- /dev/null
+++ b/kumo/small-classifier.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e51425be1d80433b2709c2cda2eda8dd6d2f3190aae7afd9e4f58b8db5fe97d
+size 55301219
diff --git a/kumo/small-regressor.f16w.onnx b/kumo/small-regressor.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..46d36df3c7453c7234ba6f80b75a4d5d54daad99
--- /dev/null
+++ b/kumo/small-regressor.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e97b5abbc5bdf71106d2768656c8d91d981e1359251a137983b711baddb745e8
+size 57320387
diff --git a/rfm/LICENSE b/rfm/LICENSE
new file mode 100644
index 0000000000000000000000000000000000000000..161a8f01c0e68899bf4175a5c641c5fa0e9b11db
--- /dev/null
+++ b/rfm/LICENSE
@@ -0,0 +1,52 @@
+Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+
+
+OpenMDW License Agreement, version 1.1 (OpenMDW-1.1)
+
+By exercising rights granted to you under this agreement, you accept and agree
+to its terms.
+
+As used in this agreement, "Model Materials" means the materials provided to
+you under this agreement, consisting of: (1) one or more machine learning
+models (including architecture and parameters); and (2) all related artifacts
+(including associated data, documentation and software) that are provided to
+you hereunder.
+
+Subject to your compliance with this agreement, permission is hereby granted,
+free of charge, to deal in the Model Materials without restriction, including
+under all copyright, patent, database, and trade secret rights included or
+embodied therein.
+
+If you distribute any portion of the Model Materials, you shall retain in your
+distribution (1) a copy of this agreement, and (2) all copyright notices and
+other notices of origin included in the Model Materials that are applicable to
+your distribution.
+
+If you file, maintain, or voluntarily participate in a lawsuit against any
+person or entity asserting that the Model Materials directly or indirectly
+infringe any patent or copyright, then all rights and grants made to you
+hereunder are terminated, unless that lawsuit was in response to a
+corresponding lawsuit first brought against you.
+
+This agreement does not impose any restrictions or obligations with respect to
+any use, modification, or sharing of any outputs generated by using the Model
+Materials.
+
+THE MODEL MATERIALS ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
+OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE, TITLE, NONINFRINGEMENT, ACCURACY, OR THE
+ABSENCE OF LATENT OR OTHER DEFECTS OR ERRORS, WHETHER OR NOT DISCOVERABLE, ALL
+TO THE GREATEST EXTENT PERMISSIBLE UNDER APPLICABLE LAW.
+
+YOU ARE SOLELY RESPONSIBLE FOR (1) CLEARING RIGHTS OF OTHER PERSONS THAT MAY
+APPLY TO THE MODEL MATERIALS OR ANY USE THEREOF, INCLUDING WITHOUT LIMITATION
+ANY PERSON'S COPYRIGHTS OR OTHER RIGHTS INCLUDED OR EMBODIED IN THE MODEL
+MATERIALS; (2) OBTAINING ANY NECESSARY CONSENTS, PERMISSIONS OR OTHER RIGHTS
+REQUIRED FOR ANY USE OF THE MODEL MATERIALS; OR (3) PERFORMING ANY DUE
+DILIGENCE OR UNDERTAKING ANY OTHER INVESTIGATIONS INTO THE MODEL MATERIALS OR
+ANYTHING INCORPORATED OR EMBODIED THEREIN.
+
+IN NO EVENT SHALL THE PROVIDERS OF THE MODEL MATERIALS BE LIABLE FOR ANY CLAIM,
+DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
+OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE MODEL MATERIALS, THE
+USE THEREOF OR OTHER DEALINGS THEREIN.
\ No newline at end of file
diff --git a/rfm/LICENSE.structured-data-models b/rfm/LICENSE.structured-data-models
new file mode 100644
index 0000000000000000000000000000000000000000..001a3281169d6ade30cd1a44544029254ba490a8
--- /dev/null
+++ b/rfm/LICENSE.structured-data-models
@@ -0,0 +1,204 @@
+Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+
+
+ Apache License
+ Version 2.0, January 2004
+ http://www.apache.org/licenses/
+
+ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+ 1. Definitions.
+
+ "License" shall mean the terms and conditions for use, reproduction,
+ and distribution as defined by Sections 1 through 9 of this document.
+
+ "Licensor" shall mean the copyright owner or entity authorized by
+ the copyright owner that is granting the License.
+
+ "Legal Entity" shall mean the union of the acting entity and all
+ other entities that control, are controlled by, or are under common
+ control with that entity. For the purposes of this definition,
+ "control" means (i) the power, direct or indirect, to cause the
+ direction or management of such entity, whether by contract or
+ otherwise, or (ii) ownership of fifty percent (50%) or more of the
+ outstanding shares, or (iii) beneficial ownership of such entity.
+
+ "You" (or "Your") shall mean an individual or Legal Entity
+ exercising permissions granted by this License.
+
+ "Source" form shall mean the preferred form for making modifications,
+ including but not limited to software source code, documentation
+ source, and configuration files.
+
+ "Object" form shall mean any form resulting from mechanical
+ transformation or translation of a Source form, including but
+ not limited to compiled object code, generated documentation,
+ and conversions to other media types.
+
+ "Work" shall mean the work of authorship, whether in Source or
+ Object form, made available under the License, as indicated by a
+ copyright notice that is included in or attached to the work
+ (an example is provided in the Appendix below).
+
+ "Derivative Works" shall mean any work, whether in Source or Object
+ form, that is based on (or derived from) the Work and for which the
+ editorial revisions, annotations, elaborations, or other modifications
+ represent, as a whole, an original work of authorship. For the purposes
+ of this License, Derivative Works shall not include works that remain
+ separable from, or merely link (or bind by name) to the interfaces of,
+ the Work and Derivative Works thereof.
+
+ "Contribution" shall mean any work of authorship, including
+ the original version of the Work and any modifications or additions
+ to that Work or Derivative Works thereof, that is intentionally
+ submitted to Licensor for inclusion in the Work by the copyright owner
+ or by an individual or Legal Entity authorized to submit on behalf of
+ the copyright owner. For the purposes of this definition, "submitted"
+ means any form of electronic, verbal, or written communication sent
+ to the Licensor or its representatives, including but not limited to
+ communication on electronic mailing lists, source code control systems,
+ and issue tracking systems that are managed by, or on behalf of, the
+ Licensor for the purpose of discussing and improving the Work, but
+ excluding communication that is conspicuously marked or otherwise
+ designated in writing by the copyright owner as "Not a Contribution."
+
+ "Contributor" shall mean Licensor and any individual or Legal Entity
+ on behalf of whom a Contribution has been received by Licensor and
+ subsequently incorporated within the Work.
+
+ 2. Grant of Copyright License. Subject to the terms and conditions of
+ this License, each Contributor hereby grants to You a perpetual,
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+ copyright license to reproduce, prepare Derivative Works of,
+ publicly display, publicly perform, sublicense, and distribute the
+ Work and such Derivative Works in Source or Object form.
+
+ 3. Grant of Patent License. Subject to the terms and conditions of
+ this License, each Contributor hereby grants to You a perpetual,
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+ (except as stated in this section) patent license to make, have made,
+ use, offer to sell, sell, import, and otherwise transfer the Work,
+ where such license applies only to those patent claims licensable
+ by such Contributor that are necessarily infringed by their
+ Contribution(s) alone or by combination of their Contribution(s)
+ with the Work to which such Contribution(s) was submitted. If You
+ institute patent litigation against any entity (including a
+ cross-claim or counterclaim in a lawsuit) alleging that the Work
+ or a Contribution incorporated within the Work constitutes direct
+ or contributory patent infringement, then any patent licenses
+ granted to You under this License for that Work shall terminate
+ as of the date such litigation is filed.
+
+ 4. Redistribution. You may reproduce and distribute copies of the
+ Work or Derivative Works thereof in any medium, with or without
+ modifications, and in Source or Object form, provided that You
+ meet the following conditions:
+
+ (a) You must give any other recipients of the Work or
+ Derivative Works a copy of this License; and
+
+ (b) You must cause any modified files to carry prominent notices
+ stating that You changed the files; and
+
+ (c) You must retain, in the Source form of any Derivative Works
+ that You distribute, all copyright, patent, trademark, and
+ attribution notices from the Source form of the Work,
+ excluding those notices that do not pertain to any part of
+ the Derivative Works; and
+
+ (d) If the Work includes a "NOTICE" text file as part of its
+ distribution, then any Derivative Works that You distribute must
+ include a readable copy of the attribution notices contained
+ within such NOTICE file, excluding those notices that do not
+ pertain to any part of the Derivative Works, in at least one
+ of the following places: within a NOTICE text file distributed
+ as part of the Derivative Works; within the Source form or
+ documentation, if provided along with the Derivative Works; or,
+ within a display generated by the Derivative Works, if and
+ wherever such third-party notices normally appear. The contents
+ of the NOTICE file are for informational purposes only and
+ do not modify the License. You may add Your own attribution
+ notices within Derivative Works that You distribute, alongside
+ or as an addendum to the NOTICE text from the Work, provided
+ that such additional attribution notices cannot be construed
+ as modifying the License.
+
+ You may add Your own copyright statement to Your modifications and
+ may provide additional or different license terms and conditions
+ for use, reproduction, or distribution of Your modifications, or
+ for any such Derivative Works as a whole, provided Your use,
+ reproduction, and distribution of the Work otherwise complies with
+ the conditions stated in this License.
+
+ 5. Submission of Contributions. Unless You explicitly state otherwise,
+ any Contribution intentionally submitted for inclusion in the Work
+ by You to the Licensor shall be under the terms and conditions of
+ this License, without any additional terms or conditions.
+ Notwithstanding the above, nothing herein shall supersede or modify
+ the terms of any separate license agreement you may have executed
+ with Licensor regarding such Contributions.
+
+ 6. Trademarks. This License does not grant permission to use the trade
+ names, trademarks, service marks, or product names of the Licensor,
+ except as required for reasonable and customary use in describing the
+ origin of the Work and reproducing the content of the NOTICE file.
+
+ 7. Disclaimer of Warranty. Unless required by applicable law or
+ agreed to in writing, Licensor provides the Work (and each
+ Contributor provides its Contributions) on an "AS IS" BASIS,
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+ implied, including, without limitation, any warranties or conditions
+ of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+ PARTICULAR PURPOSE. You are solely responsible for determining the
+ appropriateness of using or redistributing the Work and assume any
+ risks associated with Your exercise of permissions under this License.
+
+ 8. Limitation of Liability. In no event and under no legal theory,
+ whether in tort (including negligence), contract, or otherwise,
+ unless required by applicable law (such as deliberate and grossly
+ negligent acts) or agreed to in writing, shall any Contributor be
+ liable to You for damages, including any direct, indirect, special,
+ incidental, or consequential damages of any character arising as a
+ result of this License or out of the use or inability to use the
+ Work (including but not limited to damages for loss of goodwill,
+ work stoppage, computer failure or malfunction, or any and all
+ other commercial damages or losses), even if such Contributor
+ has been advised of the possibility of such damages.
+
+ 9. Accepting Warranty or Additional Liability. While redistributing
+ the Work or Derivative Works thereof, You may choose to offer,
+ and charge a fee for, acceptance of support, warranty, indemnity,
+ or other liability obligations and/or rights consistent with this
+ License. However, in accepting such obligations, You may act only
+ on Your own behalf and on Your sole responsibility, not on behalf
+ of any other Contributor, and only if You agree to indemnify,
+ defend, and hold each Contributor harmless for any liability
+ incurred by, or claims asserted against, such Contributor by reason
+ of your accepting any such warranty or additional liability.
+
+ END OF TERMS AND CONDITIONS
+
+ APPENDIX: How to apply the Apache License to your work.
+
+ To apply the Apache License to your work, attach the following
+ boilerplate notice, with the fields enclosed by brackets "[]"
+ replaced with your own identifying information. (Don't include
+ the brackets!) The text should be enclosed in the appropriate
+ comment syntax for the file format. We also recommend that a
+ file or class name and description of purpose be included on the
+ same "printed page" as the copyright notice for easier
+ identification within third-party archives.
+
+ Copyright [yyyy] [name of copyright owner]
+
+ Licensed under the Apache License, Version 2.0 (the "License");
+ you may not use this file except in compliance with the License.
+ You may obtain a copy of the License at
+
+ http://www.apache.org/licenses/LICENSE-2.0
+
+ Unless required by applicable law or agreed to in writing, software
+ distributed under the License is distributed on an "AS IS" BASIS,
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ See the License for the specific language governing permissions and
+ limitations under the License.
diff --git a/rfm/README.md b/rfm/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..203e1c846a6a99deddfe0e8c766c09ebb9ac993e
--- /dev/null
+++ b/rfm/README.md
@@ -0,0 +1,108 @@
+---
+license: openmdw-1.1
+library_name: onnx
+base_model: nvidia/Kumo-Relational
+base_model_relation: quantized
+pipeline_tag: tabular-classification
+tags:
+ - onnx
+ - onnxruntime-web
+ - webgpu
+ - relational-foundation-model
+ - structured-data-models
+ - tabular-classification
+ - tabular-regression
+ - in-context-learning
+---
+
+# KumoRelational for onnxruntime-web
+
+An unofficial ONNX conversion of NVIDIA's [Kumo Relational](https://huggingface.co/nvidia/Kumo-Relational)
+(revision `v1.0.1`), a foundation model for classification and regression
+across related tables, made for Tabular Data Workbench, which runs it in the
+browser with onnxruntime-web on WebGPU or WebAssembly. The weights are
+NVIDIA's: unchanged in the `fp32` variant, and with the larger ones rounded to
+float16 in the `f16w` variant. NVIDIA has not reviewed or endorsed this
+conversion.
+
+## Files
+
+Each head (`classifier`, `regressor`) is cut into five graphs:
+
+| file | inputs | output |
+| --- | --- | --- |
+| `-col..onnx` | `X` (columns, rows, 3), `y` (train), `keys` (keys) | `E` (columns, rows, 128) |
+| `-row..onnx` | `E` (columns, rows, 128) | `Z` (rows, 512) |
+| `-hop..onnx` | `x` (nodes, 512), `src`, `dst`, `edge_type` (edges), `noise` (types, 512), `deg` (nodes, 1) | `h` (nodes, 512) |
+| `-out..onnx` | `x` (readouts, 512) | `z` (readouts, 512) |
+| `-icl..onnx` | `Z` (tokens, 512), `y` (train) | `out` (queries, 10 logits or 999 quantiles) |
+
+`col` and `row` embed each table's rows, `hop` is one round of relational
+message passing (run once per hop), `out` is the readout and `icl` is the
+in-context block over the task's context and query rows. The graph neural
+network's segment reductions are written as ScatterND, and the graphs contain
+no int64 shape arithmetic.
+
+Each graph comes in two variants (``):
+
+- `fp32`: the checkpoint's weights as they are (244 MB for both heads).
+- `f16w`: every float32 weight of 1,024 or more elements stored as float16
+ and cast back to float32 on load (123 MB). Compute stays float32.
+
+`manifest.json` lists every file with its size and sha256, the graphs'
+inputs and outputs, each head's empty-row embedding and the parity figures
+below. The time-aware neighbour sampler, the task tables, the encoders and
+the ensemble stay outside the graphs; the app implements them in TypeScript.
+
+onnxruntime 1.30's CPU ScatterND with a reduction is not deterministic with
+more than one intra-op thread when indices repeat, which they do in the hop
+graph. Run it single-threaded where results must be reproducible.
+
+## How they were made
+
+Source: `nvidia/Kumo-Relational` at `v1.0.1` (`classifier.pt`,
+`regressor.pt`), loaded through NVIDIA's
+[structured-data-models](https://github.com/NVIDIA/structured-data-models)
+at commit `2b1696f4fdc22004cc757c44703a69cf326c0a11`.
+
+```sh
+node models/rfm/dump_shop.mjs # once, from the repository root: the sample tables
+cd models/rfm
+uv run export.py # writes public/models/rfm/
+```
+
+`export.py` (in `models/rfm/` of Tabular Data Workbench) exports the parts
+defined in `exportable.py` with `torch.onnx` (dynamo, opset 18, PyTorch 2.14
+on CPU) and removes the exporter's node metadata and doc strings. It checks
+each graph against PyTorch, then checks the five graphs chained against
+structured-data-models' own forward pass on inputs captured from a real run
+over the app's sample shop tables.
+
+## Parity
+
+Chained graphs (onnxruntime 1.30, CPU, one thread) against
+structured-data-models' forward pass, maximum absolute difference:
+
+| head | task | fp32 | f16w | largest output |
+| --- | --- | --- | --- | --- |
+| classifier | churn (logits) | 8.6e-3 | 1.2e-2 | 10.6 |
+| regressor | spend (quantiles) | 1.6e-3 | 6.5e-3 | 4.9 |
+
+Each graph alone, on random inputs, matches PyTorch to 1.1e-5 (fp32).
+
+## Licence
+
+The weights are released by NVIDIA under the OpenMDW License Agreement 1.1
+(`LICENSE`, Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES). It allows
+use, modification and redistribution provided that the agreement and the
+notices of origin travel with the weights, and it ends the rights of anyone
+who sues claiming the model infringes a patent or copyright.
+
+Kumo Relational fine-tunes the TabICLv2 weights (Soda team, Inria),
+licensed under BSD-3-Clause; `THIRD-PARTY-NOTICES.txt` is NVIDIA's notice
+with that licence's full text. The graphs were exported from
+structured-data-models' model code, licensed under Apache-2.0
+(`LICENSE.structured-data-models`).
+
+Keep `LICENSE`, `THIRD-PARTY-NOTICES.txt` and
+`LICENSE.structured-data-models` with the graphs when redistributing them.
diff --git a/rfm/THIRD-PARTY-NOTICES.txt b/rfm/THIRD-PARTY-NOTICES.txt
new file mode 100644
index 0000000000000000000000000000000000000000..28215aee453d02bec3fc22ac534197b7665bc7f2
--- /dev/null
+++ b/rfm/THIRD-PARTY-NOTICES.txt
@@ -0,0 +1,38 @@
+THIRD-PARTY MODEL NOTICES
+
+Component: TabICLv2 pretrained model weights
+Model source: https://huggingface.co/jingang/TabICL
+License declaration: https://huggingface.co/jingang/TabICL/blob/main/README.md
+License: BSD 3-Clause License (BSD-3-Clause)
+
+Kumo Relational finetunes pretrained TabICLv2 weights.
+Kumo Relational model weights are released under OpenMDW-1.1; see LICENSE.
+
+BSD 3-Clause License
+
+Copyright (c) 2025, Soda team @ Inria
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+1. Redistributions of source code must retain the above copyright notice, this
+ list of conditions and the following disclaimer.
+
+2. Redistributions in binary form must reproduce the above copyright notice,
+ this list of conditions and the following disclaimer in the documentation
+ and/or other materials provided with the distribution.
+
+3. Neither the name of the copyright holder nor the names of its
+ contributors may be used to endorse or promote products derived from
+ this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
\ No newline at end of file
diff --git a/rfm/classifier-col.f16w.onnx b/rfm/classifier-col.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..ed01d4bfaa7c263f4cf7067afa087467fc3ad1bb
--- /dev/null
+++ b/rfm/classifier-col.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35a0cde45113e557ec25ff7300d94f8c1114f725517cc36654e004e8280faa51
+size 1881044
diff --git a/rfm/classifier-col.fp32.onnx b/rfm/classifier-col.fp32.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..910cca97a026773290b02c24e14d8bb7d95df02b
--- /dev/null
+++ b/rfm/classifier-col.fp32.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4258b08a30fa4eec00b5804f6c575b4574584ad6e1a311f9b211030a3533ea6f
+size 3613403
diff --git a/rfm/classifier-hop.f16w.onnx b/rfm/classifier-hop.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..423152b2faa348a6aebfd42172b990028830df49
--- /dev/null
+++ b/rfm/classifier-hop.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a9bb35380493127f4a76adf44a19a33a9a69d1b58343f38ecdd4f77792d9b4f
+size 4209838
diff --git a/rfm/classifier-hop.fp32.onnx b/rfm/classifier-hop.fp32.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..24e2c840a8027e18a090135ecea593a30edd39f6
--- /dev/null
+++ b/rfm/classifier-hop.fp32.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a52f544ef90eb6de61c517e7e61c7fe63a6e3286a0a0a20356eae2e169616333
+size 8403886
diff --git a/rfm/classifier-icl.f16w.onnx b/rfm/classifier-icl.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..db27e175740969ce1f70b698c93f625a6ca5d52b
--- /dev/null
+++ b/rfm/classifier-icl.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e161ab3388e4fb622513c5c8a6b14f73cdba4575b4c16dfb1d9a1b518379f27f
+size 52943138
diff --git a/rfm/classifier-icl.fp32.onnx b/rfm/classifier-icl.fp32.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..09a918625b966bfc872b1a244f8e13de22becef2
--- /dev/null
+++ b/rfm/classifier-icl.fp32.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a0f29920552b068feb889bbc4aa58ebf3ce8514b8c29c071a6a7baabd6a5b1b
+size 105355516
diff --git a/rfm/classifier-out.f16w.onnx b/rfm/classifier-out.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..4868d8880c115f0b8bbeb4fa8c143538ea1ce5d1
--- /dev/null
+++ b/rfm/classifier-out.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7acf1c5378cb5e2edf67b8040a440914e0cd7d8a2081737b2e863ac8fc566b5a
+size 531196
diff --git a/rfm/classifier-out.fp32.onnx b/rfm/classifier-out.fp32.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..675748be55a20b4efb2f8fefb4bf598fedf15af3
--- /dev/null
+++ b/rfm/classifier-out.fp32.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5bbde304395d3eb511173a82fbc8ac3ffdf11c9d933c5ebc7df7ab439d0e38bf
+size 1055415
diff --git a/rfm/classifier-row.f16w.onnx b/rfm/classifier-row.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..8255bbeede3971c4417de4a7f034c83946bc681e
--- /dev/null
+++ b/rfm/classifier-row.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87b5b9cd288b251bee4b84a056c123941407174a7d0c887a1e386bf326aad3f5
+size 854772
diff --git a/rfm/classifier-row.fp32.onnx b/rfm/classifier-row.fp32.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..7e3038b2f864a7b232de2ba4164c3130c28e1a9b
--- /dev/null
+++ b/rfm/classifier-row.fp32.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:58f6cb77884920c4e92813cb785fa8a95567a94814a2f5d7ae8a4827c214e328
+size 1640460
diff --git a/rfm/manifest.json b/rfm/manifest.json
new file mode 100644
index 0000000000000000000000000000000000000000..398e1fe23d34ff4be4ac78f411357d844652d973
--- /dev/null
+++ b/rfm/manifest.json
@@ -0,0 +1,1840 @@
+{
+ "model": "KumoRelational",
+ "source": {
+ "package": "structured-data-models @ 2b1696f4fdc2",
+ "repo": "nvidia/Kumo-Relational",
+ "revision": "v1.0.1",
+ "checkpoints": {
+ "classifier": "classifier.pt",
+ "regressor": "regressor.pt"
+ },
+ "license": "OpenMDW-1.1 (weights; fine-tuned from TabICLv2, BSD-3-Clause)",
+ "notices": [
+ "LICENSE",
+ "THIRD-PARTY-NOTICES.txt"
+ ]
+ },
+ "opset": 18,
+ "exporter": "torch.onnx (dynamo) 2.14.0+cpu",
+ "limits": {
+ "maxClasses": 10,
+ "numQuantiles": 999,
+ "featureGroupSize": 3,
+ "cellDim": 128,
+ "rowDim": 512,
+ "maxKeys": 20000,
+ "temperature": 0.9
+ },
+ "tasks": {
+ "classifier": {
+ "numClasses": 10,
+ "outDim": 10,
+ "emptyRowEmbedding": [
+ -0.6515481472015381,
+ -0.6578977108001709,
+ -0.6777721643447876,
+ -0.7822152376174927,
+ 1.7381236553192139,
+ 4.868432521820068,
+ 0.021998055279254913,
+ 0.2011304348707199,
+ -1.1834341287612915,
+ -0.6197394728660583,
+ -0.26040998101234436,
+ 1.8957850933074951,
+ 0.1128038763999939,
+ -0.5565066337585449,
+ 0.2857104241847992,
+ -0.1305471807718277,
+ -0.02089848183095455,
+ 0.6901000142097473,
+ -0.633758544921875,
+ -0.5899574160575867,
+ 0.17074818909168243,
+ 0.9734656810760498,
+ -0.3416438698768616,
+ 0.3050732910633087,
+ 1.0541795492172241,
+ 0.3212589621543884,
+ 0.22909268736839294,
+ 0.03650117665529251,
+ 0.036828238517045975,
+ -0.7325596213340759,
+ 2.1401281356811523,
+ 2.0198001861572266,
+ 1.0548911094665527,
+ 1.1351673603057861,
+ -0.6504899263381958,
+ 1.2082951068878174,
+ 0.3168076276779175,
+ -0.908993661403656,
+ -0.028187081217765808,
+ -0.08583246916532516,
+ 0.23075464367866516,
+ -0.1810135841369629,
+ 0.934986412525177,
+ 0.2889697551727295,
+ -0.9346817135810852,
+ -0.9134857654571533,
+ -0.2210700660943985,
+ -0.192105233669281,
+ -0.7671796679496765,
+ 0.07601941376924515,
+ 0.008838735520839691,
+ -0.20181573927402496,
+ -0.330454021692276,
+ 0.19383491575717926,
+ -1.8057305812835693,
+ -0.28069818019866943,
+ -1.304654836654663,
+ 0.5029457211494446,
+ -1.1884835958480835,
+ 0.7161130309104919,
+ -0.3595302104949951,
+ 0.2729124426841736,
+ -0.7834827899932861,
+ -0.6621738076210022,
+ 0.06356482207775116,
+ -0.19944192469120026,
+ -1.109471321105957,
+ -0.19683974981307983,
+ -0.46237966418266296,
+ 0.7138513922691345,
+ -0.1656678318977356,
+ 1.0790364742279053,
+ -0.19757182896137238,
+ 0.009998749010264874,
+ 0.3477487862110138,
+ 0.6488944888114929,
+ 0.6338379383087158,
+ -0.4470149576663971,
+ 0.13880422711372375,
+ -0.7759476900100708,
+ -1.757300615310669,
+ -0.1753847748041153,
+ -0.42946237325668335,
+ -0.2217670977115631,
+ -0.8044957518577576,
+ 1.1557780504226685,
+ -0.36158156394958496,
+ 1.0754748582839966,
+ 0.6351344585418701,
+ -0.18819473683834076,
+ 0.1983778178691864,
+ -0.2253461331129074,
+ -0.5574930310249329,
+ -0.4255942702293396,
+ 0.19450682401657104,
+ -0.39619654417037964,
+ 1.1550514698028564,
+ 0.9769047498703003,
+ -0.7456809282302856,
+ -1.3812838792800903,
+ 0.07662203907966614,
+ -1.4201765060424805,
+ 0.3417740762233734,
+ -0.3149818181991577,
+ 0.18378210067749023,
+ -0.6355498433113098,
+ 0.02634390816092491,
+ -0.0017376020550727844,
+ -0.8521059155464172,
+ -0.23024842143058777,
+ -0.1522301733493805,
+ -0.7140592932701111,
+ 0.03488187864422798,
+ 0.9833220839500427,
+ -0.6239383816719055,
+ 0.6861448287963867,
+ 0.3349078893661499,
+ -0.3160772919654846,
+ -0.7955625057220459,
+ -0.7610108852386475,
+ -0.5100499391555786,
+ 0.034780535846948624,
+ -0.12492454051971436,
+ -0.06336754560470581,
+ -0.6480013728141785,
+ -0.5075410604476929,
+ 0.9258944988250732,
+ -0.23691321909427643,
+ 0.422885924577713,
+ 1.1779035329818726,
+ -0.29671525955200195,
+ 0.45856454968452454,
+ 2.2199974060058594,
+ 6.049717426300049,
+ 0.4247852861881256,
+ 0.03326593339443207,
+ 0.18774375319480896,
+ -3.0617928504943848,
+ -0.3538617491722107,
+ 0.8910201787948608,
+ 0.06939065456390381,
+ -0.45427799224853516,
+ 0.1145935133099556,
+ -1.8876785039901733,
+ -0.6034613251686096,
+ 0.30548396706581116,
+ -0.1077260747551918,
+ -1.8353358507156372,
+ 0.6320286393165588,
+ 0.342405766248703,
+ -0.9466231465339661,
+ -0.2223590910434723,
+ 1.633811116218567,
+ 1.1988320350646973,
+ 0.5507926344871521,
+ 0.022187625989317894,
+ 0.08708589524030685,
+ -0.46132567524909973,
+ 0.9039878845214844,
+ 1.2047632932662964,
+ 1.6597583293914795,
+ 0.3509847819805145,
+ -0.7708027362823486,
+ -0.00012385949958115816,
+ 0.02622387185692787,
+ 0.23687326908111572,
+ -0.23483432829380035,
+ 0.785348653793335,
+ 1.387162685394287,
+ 0.05733492970466614,
+ 0.2834599018096924,
+ -1.5549767017364502,
+ -1.1869844198226929,
+ -0.5001720786094666,
+ 0.1129777655005455,
+ 0.2399885356426239,
+ -0.36499422788619995,
+ 0.11703646183013916,
+ -0.9573560953140259,
+ 1.7035828828811646,
+ -0.47999221086502075,
+ -0.6850872039794922,
+ -1.9630496501922607,
+ -0.26216503977775574,
+ -1.236363172531128,
+ 1.2131667137145996,
+ -1.0335935354232788,
+ 0.49745985865592957,
+ -0.36879947781562805,
+ -1.4335675239562988,
+ -1.2505924701690674,
+ -0.09342067688703537,
+ -0.1031719297170639,
+ -0.845089316368103,
+ -0.25144851207733154,
+ 0.37272748351097107,
+ 0.28259360790252686,
+ 0.1442316174507141,
+ 1.0918049812316895,
+ -0.4222988486289978,
+ -0.5521408915519714,
+ -0.08937276899814606,
+ -0.6735305190086365,
+ 0.09348229318857193,
+ 0.8985345363616943,
+ 0.4040052592754364,
+ -0.9244634509086609,
+ -0.8207292556762695,
+ -0.7995850443840027,
+ -0.7133394479751587,
+ 1.4457272291183472,
+ 0.6235929727554321,
+ -1.1042652130126953,
+ -0.18853268027305603,
+ -0.030584143474698067,
+ 0.7769511342048645,
+ 0.21775323152542114,
+ 0.6901780962944031,
+ 0.0605451837182045,
+ -0.2597784101963043,
+ 0.011389640159904957,
+ -0.2782304883003235,
+ 0.5337550044059753,
+ 0.13002461194992065,
+ 2.858579635620117,
+ 2.2165541648864746,
+ -1.6957606077194214,
+ -1.3484234809875488,
+ 0.5086627006530762,
+ 0.5374454259872437,
+ 0.600132405757904,
+ -0.35753437876701355,
+ 0.9092206358909607,
+ 0.5283166170120239,
+ 0.2509515583515167,
+ -0.17753618955612183,
+ 0.6733176112174988,
+ 0.045449499040842056,
+ -0.8949448466300964,
+ -0.16959534585475922,
+ -0.09252668917179108,
+ 1.1087617874145508,
+ -0.3724087178707123,
+ 0.3322495222091675,
+ 0.48003607988357544,
+ -0.3687056601047516,
+ -1.3073554039001465,
+ -0.31638821959495544,
+ -0.5134118795394897,
+ -0.9235721230506897,
+ -0.36659881472587585,
+ -0.1856650561094284,
+ -0.3007179796695709,
+ -0.2346922606229782,
+ 0.11110550165176392,
+ 0.5007892847061157,
+ -1.2198755741119385,
+ -0.3743126690387726,
+ -0.3759503960609436,
+ 0.2762299180030823,
+ 2.34713077545166,
+ 6.7869110107421875,
+ -0.003611281979829073,
+ 0.24288053810596466,
+ -0.8049203157424927,
+ -0.8765348792076111,
+ -0.23434144258499146,
+ 2.1146841049194336,
+ 0.20689937472343445,
+ -0.07038551568984985,
+ 0.377560019493103,
+ -1.0866844654083252,
+ -0.1386021077632904,
+ 0.6897559762001038,
+ -0.4297366142272949,
+ -0.39869794249534607,
+ 0.05390496179461479,
+ 0.7515197396278381,
+ -1.0570094585418701,
+ 0.6468269228935242,
+ 1.0746899843215942,
+ 0.7809686660766602,
+ 0.11603648215532303,
+ 0.1724310964345932,
+ 0.10877268761396408,
+ -0.5221180319786072,
+ 1.6492156982421875,
+ 2.0212700366973877,
+ 1.0316636562347412,
+ 1.2041032314300537,
+ -0.4907228350639343,
+ 0.9765589237213135,
+ 0.3453936278820038,
+ 0.27868375182151794,
+ -0.03254920244216919,
+ -0.6575308442115784,
+ 0.10713095963001251,
+ -0.04376266896724701,
+ 0.23342663049697876,
+ -0.7458587288856506,
+ -0.6449275016784668,
+ -0.856620192527771,
+ 0.1639704704284668,
+ 0.01428339071571827,
+ 0.697925329208374,
+ 0.060491833835840225,
+ -0.32361650466918945,
+ -1.05996572971344,
+ -0.12626299262046814,
+ 1.0041906833648682,
+ -1.414699673652649,
+ -0.13821031153202057,
+ -1.269945740699768,
+ 0.1878547966480255,
+ -0.8674177527427673,
+ 0.46431005001068115,
+ 0.2327304184436798,
+ -1.3188105821609497,
+ -0.9720017910003662,
+ -0.5720484256744385,
+ 0.18390484154224396,
+ 0.16114766895771027,
+ -0.8153122663497925,
+ -0.3771165907382965,
+ -0.061796486377716064,
+ 0.17708726227283478,
+ 0.5890390872955322,
+ 1.07102632522583,
+ -0.3739709258079529,
+ 0.06789779663085938,
+ 0.3396647572517395,
+ 1.523109793663025,
+ -0.7677808403968811,
+ -0.6499733328819275,
+ -0.043903522193431854,
+ -0.9914679527282715,
+ -1.8342633247375488,
+ -0.1429421305656433,
+ -0.07747496664524078,
+ -0.6471755504608154,
+ 0.018194684758782387,
+ 0.8335069417953491,
+ 0.42902958393096924,
+ 0.7429405450820923,
+ -0.07746612280607224,
+ 0.6400280594825745,
+ 0.06367930769920349,
+ -0.2671132981777191,
+ -0.025443751364946365,
+ -0.7863265872001648,
+ -0.12874262034893036,
+ -0.5645725131034851,
+ 1.3786741495132446,
+ 0.6260976195335388,
+ -0.37118515372276306,
+ -1.3806756734848022,
+ -0.8428974747657776,
+ -0.8127872347831726,
+ 0.24264702200889587,
+ 1.0861274003982544,
+ -0.6611175537109375,
+ -1.5513105392456055,
+ 0.1873193085193634,
+ -0.1576937735080719,
+ -0.537710964679718,
+ 0.33131834864616394,
+ 0.05729297176003456,
+ -0.5702476501464844,
+ -0.6196622252464294,
+ 0.7584403157234192,
+ -0.48518112301826477,
+ 0.10447536408901215,
+ 1.2382197380065918,
+ -0.19881097972393036,
+ -0.11184627562761307,
+ -1.3794587850570679,
+ -0.6244295239448547,
+ -1.1302515268325806,
+ -0.7000928521156311,
+ -0.41212040185928345,
+ -0.7566840052604675,
+ -0.6411684155464172,
+ 1.1274703741073608,
+ 0.6366463303565979,
+ -1.3027212619781494,
+ 0.6816779971122742,
+ 0.1727447509765625,
+ 0.25206440687179565,
+ 1.8186497688293457,
+ 3.5384328365325928,
+ -0.09391462057828903,
+ -0.31659889221191406,
+ -0.4867439568042755,
+ -0.6523049473762512,
+ -0.03733422979712486,
+ 2.59932804107666,
+ 0.2793954014778137,
+ -0.16736052930355072,
+ 0.2004922330379486,
+ -1.0194627046585083,
+ -0.2152218073606491,
+ 0.916074275970459,
+ -0.7214213013648987,
+ -0.2644707262516022,
+ -0.5747984051704407,
+ 1.2473877668380737,
+ -0.9385489821434021,
+ 0.41783803701400757,
+ 1.4941927194595337,
+ 2.101767063140869,
+ 0.36512520909309387,
+ 0.1193167120218277,
+ 0.17989879846572876,
+ -0.5801778435707092,
+ 2.2088305950164795,
+ 1.5660384893417358,
+ 1.1682071685791016,
+ 0.563066303730011,
+ -0.6116885542869568,
+ 1.3193466663360596,
+ 0.267682820558548,
+ -0.62977534532547,
+ 0.010850176215171814,
+ -0.10765872895717621,
+ 0.39721477031707764,
+ -0.02617522142827511,
+ -0.33186206221580505,
+ -1.3612327575683594,
+ -0.3969060480594635,
+ -1.0420327186584473,
+ 0.19964535534381866,
+ -0.05370185524225235,
+ 1.9150503873825073,
+ 0.10939188301563263,
+ -0.812909722328186,
+ -0.2523416578769684,
+ 0.13053910434246063,
+ 0.927209198474884,
+ -2.232102155685425,
+ -0.5840155482292175,
+ -0.7618265748023987,
+ -0.18549108505249023,
+ -0.7721548080444336,
+ 0.30403372645378113,
+ 0.380097895860672,
+ -0.3163313865661621,
+ -1.6085025072097778,
+ -0.24368107318878174,
+ 0.13427546620368958,
+ 0.19827839732170105,
+ -0.4600118398666382,
+ -0.3990802466869354,
+ -0.14053016901016235,
+ -0.23337261378765106,
+ 0.2538866102695465,
+ 1.2537442445755005,
+ -0.7838678956031799,
+ -0.10137762874364853,
+ 0.47031790018081665,
+ 0.5875551700592041,
+ 0.2403402030467987,
+ -0.2355261594057083,
+ -0.6959521770477295,
+ -1.1201555728912354,
+ -1.9424922466278076,
+ 0.20083189010620117,
+ 0.813184380531311,
+ -0.3464331030845642,
+ 0.23560313880443573,
+ 0.5285450220108032,
+ 0.09845944494009018,
+ 0.4913387894630432,
+ -0.5413938164710999,
+ 2.265829086303711,
+ -0.003198265330865979,
+ -0.24163955450057983,
+ 0.35089775919914246,
+ -1.5483036041259766,
+ -0.12350653856992722,
+ -0.8283056020736694,
+ 1.4019381999969482,
+ 0.42219260334968567,
+ 0.4508354067802429,
+ -1.8289210796356201,
+ -0.9122130870819092,
+ 0.8255137205123901,
+ 0.7213162779808044,
+ 0.8427438139915466,
+ -0.7342632412910461,
+ -1.7567100524902344,
+ 0.39213860034942627,
+ -0.15453149378299713,
+ -0.4189983904361725,
+ 0.579968273639679,
+ -0.7543401122093201,
+ 0.13695837557315826,
+ -1.0355637073516846,
+ 0.13065554201602936,
+ -0.2054397016763687,
+ 0.0388934388756752,
+ -0.12198536843061447,
+ -0.4686793386936188,
+ -0.2934783399105072,
+ -1.164990782737732,
+ -0.4009651243686676,
+ -0.8292542099952698,
+ -0.2866267263889313,
+ -0.2246156483888626,
+ -0.3512592613697052,
+ -0.5403620600700378,
+ 0.8036950826644897,
+ 0.47769179940223694
+ ],
+ "parity": {
+ "fp32": {
+ "maxAbsDiff": 0.008566856384277344,
+ "maxAbsReference": 10.59189224243164
+ },
+ "f16w": {
+ "maxAbsDiff": 0.011764049530029297,
+ "maxAbsReference": 10.59189224243164
+ }
+ }
+ },
+ "regressor": {
+ "numClasses": 0,
+ "outDim": 999,
+ "emptyRowEmbedding": [
+ -0.4694979786872864,
+ -0.5284363031387329,
+ 0.7942131161689758,
+ -0.1740059107542038,
+ 0.0819745659828186,
+ -1.3124985694885254,
+ -1.2073546648025513,
+ 0.043875839561223984,
+ 0.4463379681110382,
+ -0.06377585232257843,
+ -0.6623969078063965,
+ 1.4375805854797363,
+ 0.1599443554878235,
+ 0.06443210691213608,
+ -0.07009238004684448,
+ 0.1406291127204895,
+ -0.10016635805368423,
+ -0.11019332706928253,
+ 0.8808096051216125,
+ 1.9223638772964478,
+ 0.1885247379541397,
+ -0.4930430054664612,
+ 0.14843468368053436,
+ 0.05192318931221962,
+ 0.052645809948444366,
+ 0.11989498138427734,
+ 0.7977968454360962,
+ -0.15793216228485107,
+ 0.10221352428197861,
+ 0.9230976700782776,
+ -0.2983953356742859,
+ -0.3075169622898102,
+ -2.147857904434204,
+ -0.14144450426101685,
+ 0.8169675469398499,
+ -0.7120842933654785,
+ 0.0256796982139349,
+ -1.142195463180542,
+ -0.11487342417240143,
+ 2.272392511367798,
+ 0.06951839476823807,
+ -0.1910751760005951,
+ -1.095170259475708,
+ 0.8089185953140259,
+ -0.7023834586143494,
+ -0.08879584074020386,
+ 0.17248710989952087,
+ -0.7444038987159729,
+ 0.0691181942820549,
+ 1.8920302391052246,
+ -0.3236050307750702,
+ -0.3157457709312439,
+ 0.8135353326797485,
+ -0.7584391236305237,
+ -2.9412953853607178,
+ 0.04221900925040245,
+ 0.2583717405796051,
+ -0.15304900705814362,
+ 0.11633336544036865,
+ -2.524087905883789,
+ 0.2954463064670563,
+ -1.1413064002990723,
+ -0.8557645082473755,
+ -1.2421867847442627,
+ 0.6307764053344727,
+ 0.022662712261080742,
+ -0.22013036906719208,
+ -0.015391030348837376,
+ -0.1214335709810257,
+ 0.45584020018577576,
+ -1.1608262062072754,
+ 0.0529545322060585,
+ 0.1444423943758011,
+ 0.0859779417514801,
+ -0.11915931850671768,
+ 0.2944789528846741,
+ -0.12866798043251038,
+ 0.9841223359107971,
+ 0.054290398955345154,
+ 0.47171881794929504,
+ 0.34118175506591797,
+ -2.7080078125,
+ 0.04803191125392914,
+ -2.8414955139160156,
+ -1.3152353763580322,
+ 2.2643065452575684,
+ -0.002209432190284133,
+ 1.4412553310394287,
+ -1.8307772874832153,
+ -0.09562166780233383,
+ 0.22316887974739075,
+ 0.14821569621562958,
+ 0.13283200562000275,
+ 0.04302593693137169,
+ 0.631517767906189,
+ -0.22365802526474,
+ 2.861109733581543,
+ -1.69280207157135,
+ 0.19328664243221283,
+ 0.10534053295850754,
+ 0.0377529077231884,
+ -0.06221825256943703,
+ -0.15376128256320953,
+ 2.131559133529663,
+ 0.43501541018486023,
+ 0.10493450611829758,
+ 0.7153786420822144,
+ 0.2233937680721283,
+ -0.04615440592169762,
+ 0.15523748099803925,
+ 0.9501121640205383,
+ 0.06562787294387817,
+ -4.547725200653076,
+ -0.3559969663619995,
+ -0.25796622037887573,
+ -0.13153403997421265,
+ -1.1350716352462769,
+ 0.42120644450187683,
+ 0.3186573386192322,
+ 0.10512907058000565,
+ -0.10680767893791199,
+ -0.06810915470123291,
+ 0.5287041068077087,
+ -2.7750155925750732,
+ 0.006905125454068184,
+ -2.2520101070404053,
+ 0.15578068792819977,
+ -3.854083776473999,
+ -0.42131710052490234,
+ 0.20317406952381134,
+ 0.983536958694458,
+ 1.3437145948410034,
+ 0.00420773308724165,
+ -1.9905420541763306,
+ -0.5379229187965393,
+ 0.0691051259636879,
+ 0.6821646094322205,
+ -0.4331040382385254,
+ 0.03798132389783859,
+ 1.9691011905670166,
+ 1.4221292734146118,
+ 0.04968034848570824,
+ -0.17964068055152893,
+ 0.15767499804496765,
+ -0.04969444498419762,
+ -0.37859687209129333,
+ 1.0235233306884766,
+ 1.4445098638534546,
+ 0.07226166874170303,
+ -0.3599167466163635,
+ 0.13257861137390137,
+ -0.42844775319099426,
+ 0.22744068503379822,
+ -0.25191816687583923,
+ 0.7403735518455505,
+ -0.14049169421195984,
+ 0.07010772824287415,
+ 0.06584154069423676,
+ -0.9126791954040527,
+ -0.8162152767181396,
+ -0.954106330871582,
+ -0.5332455039024353,
+ 1.339992880821228,
+ 0.05303521081805229,
+ 0.01032345648854971,
+ -1.28511381149292,
+ 0.07129594683647156,
+ 0.6310217976570129,
+ 0.1765824407339096,
+ -0.13819557428359985,
+ -0.850491464138031,
+ 0.9061627984046936,
+ -0.3883548080921173,
+ -0.08780547231435776,
+ 0.023276332765817642,
+ 0.2883932590484619,
+ 0.21321770548820496,
+ 2.3854525089263916,
+ 0.009892505593597889,
+ -0.18126742541790009,
+ 0.6580517292022705,
+ -0.4252259135246277,
+ -0.8437998294830322,
+ 0.09941366314888,
+ -0.7027503848075867,
+ -0.5152944326400757,
+ 0.08010787516832352,
+ -3.069200038909912,
+ 0.10175176709890366,
+ -1.429904580116272,
+ -0.6756195425987244,
+ -1.1351020336151123,
+ 1.1273844242095947,
+ 0.42387914657592773,
+ 0.08681066334247589,
+ -1.101196050643921,
+ -0.09928438067436218,
+ 0.5539436340332031,
+ -1.188781976699829,
+ -0.20027965307235718,
+ 0.36987486481666565,
+ 0.04507005959749222,
+ -0.3869955539703369,
+ 0.3004930317401886,
+ -0.5720422267913818,
+ 0.7905634045600891,
+ -0.2781822383403778,
+ 1.0173399448394775,
+ 0.08428431302309036,
+ -1.9608047008514404,
+ 0.1558610498905182,
+ -2.1886048316955566,
+ -0.8701810240745544,
+ 2.7474851608276367,
+ 0.044570595026016235,
+ 1.9139198064804077,
+ -1.1593388319015503,
+ 0.27288296818733215,
+ -0.03178431838750839,
+ 0.2022245079278946,
+ -0.13363920152187347,
+ 0.0879179835319519,
+ 0.6321004629135132,
+ -0.16256083548069,
+ 2.707683801651001,
+ -0.46909424662590027,
+ -0.20950688421726227,
+ -0.27833977341651917,
+ 0.6684026718139648,
+ 0.04562697932124138,
+ -0.10078641027212143,
+ 1.2334874868392944,
+ 0.3106614351272583,
+ 0.14261791110038757,
+ 0.7899978756904602,
+ 0.7311763763427734,
+ 0.3256126046180725,
+ 0.04568248614668846,
+ 1.373286485671997,
+ -0.08564990013837814,
+ -4.541531085968018,
+ -1.6988294124603271,
+ -0.14241217076778412,
+ -0.4839921295642853,
+ -0.5404940247535706,
+ -1.046378254890442,
+ 0.29746195673942566,
+ -0.1112639531493187,
+ -0.4752063453197479,
+ -0.1271742880344391,
+ 2.1555097103118896,
+ -1.3987003564834595,
+ 0.09240797907114029,
+ -2.9889349937438965,
+ 0.15221719443798065,
+ -1.5289369821548462,
+ -0.5043174624443054,
+ -0.6289622187614441,
+ 0.36569666862487793,
+ 1.7360724210739136,
+ -0.0156303271651268,
+ -1.7020562887191772,
+ -1.234953761100769,
+ -0.04897680506110191,
+ 0.0015682547818869352,
+ -0.19800569117069244,
+ -0.5232179760932922,
+ 0.4615810215473175,
+ 0.38175562024116516,
+ -0.10689545422792435,
+ -0.180118128657341,
+ 0.41886481642723083,
+ -0.16819797456264496,
+ -0.13090743124485016,
+ 0.7626848816871643,
+ 1.5892177820205688,
+ 0.22286584973335266,
+ -0.37462764978408813,
+ 0.03223014995455742,
+ -1.0433772802352905,
+ 0.09789539128541946,
+ 0.0882999524474144,
+ 0.7384302616119385,
+ 0.257904589176178,
+ 0.14308391511440277,
+ 0.3340730369091034,
+ -0.37996524572372437,
+ -0.5544859170913696,
+ -2.0617072582244873,
+ -0.10683762282133102,
+ 0.8719659447669983,
+ -0.0953662171959877,
+ 0.02857489138841629,
+ -1.7508457899093628,
+ -0.001652607461437583,
+ 1.0523642301559448,
+ -1.5780483484268188,
+ -0.06473027914762497,
+ -2.129587411880493,
+ 0.7441123127937317,
+ -0.6347180604934692,
+ -0.09376972168684006,
+ 0.19356223940849304,
+ 1.0599101781845093,
+ 0.010516265407204628,
+ 0.7719600796699524,
+ -0.16044656932353973,
+ -0.23055405914783478,
+ 0.5584052205085754,
+ -0.6527419686317444,
+ -1.7498098611831665,
+ -0.010209263302385807,
+ 0.17469215393066406,
+ -0.2665475904941559,
+ 0.010467955842614174,
+ -2.68434739112854,
+ 0.5477709174156189,
+ -1.2088308334350586,
+ -0.4581184685230255,
+ -1.1672649383544922,
+ 0.6092084646224976,
+ 0.4373133182525635,
+ -0.03000088408589363,
+ -0.3497457206249237,
+ -0.19048893451690674,
+ 0.6278586983680725,
+ 0.6610175967216492,
+ -0.14683549106121063,
+ 0.8250941634178162,
+ 0.07790695130825043,
+ -0.24170733988285065,
+ 0.30327358841896057,
+ -0.15131603181362152,
+ 0.8567246198654175,
+ -0.32184523344039917,
+ 0.8064577579498291,
+ 0.34245675802230835,
+ -0.2098846584558487,
+ -0.08177467435598373,
+ -3.3721365928649902,
+ -0.9989380836486816,
+ 2.252946138381958,
+ -0.16334299743175507,
+ 1.6002048254013062,
+ -0.6729496717453003,
+ 0.10327114164829254,
+ 0.010235420428216457,
+ -0.022049760445952415,
+ 0.005901553202420473,
+ 0.06463601440191269,
+ 0.859628438949585,
+ -0.4298275113105774,
+ 2.4730336666107178,
+ -2.4295125007629395,
+ 0.20150068402290344,
+ 0.21143126487731934,
+ 0.03119017370045185,
+ 0.13222353160381317,
+ -0.14819854497909546,
+ 1.4167088270187378,
+ 0.5117492079734802,
+ 0.3221425712108612,
+ -0.30901554226875305,
+ 1.2348562479019165,
+ 0.08088135719299316,
+ -0.10042165964841843,
+ 0.8788049221038818,
+ 0.05959128215909004,
+ -5.117993354797363,
+ -0.912495493888855,
+ -0.29120442271232605,
+ -0.46083927154541016,
+ -1.2243729829788208,
+ 0.5436981916427612,
+ 0.420470654964447,
+ 0.36318808794021606,
+ -0.2511226534843445,
+ -0.0725305899977684,
+ 2.0336923599243164,
+ -1.3193378448486328,
+ 0.030161553993821144,
+ -2.68609619140625,
+ 0.22209231555461884,
+ -2.029188632965088,
+ -0.30868279933929443,
+ 0.05908827483654022,
+ 0.793724775314331,
+ -0.9478907585144043,
+ 0.11153782904148102,
+ -1.657224416732788,
+ -0.5938335061073303,
+ 0.01591452583670616,
+ 1.3182036876678467,
+ -0.1615467667579651,
+ -0.2587709426879883,
+ 1.1507468223571777,
+ 0.4555306136608124,
+ 0.033915795385837555,
+ 0.06200772896409035,
+ 0.17856965959072113,
+ -0.028422415256500244,
+ -0.3590381145477295,
+ 0.7090999484062195,
+ 1.7100611925125122,
+ 0.14070144295692444,
+ -0.5500832200050354,
+ 0.20478008687496185,
+ -0.1213708221912384,
+ 0.20196767151355743,
+ -0.12583468854427338,
+ 0.6796978712081909,
+ -0.30199992656707764,
+ 0.03979099541902542,
+ 0.5898949503898621,
+ 0.03335940092802048,
+ -0.5206530690193176,
+ -0.9246997833251953,
+ -0.08036378026008606,
+ 1.0086718797683716,
+ -0.3107143044471741,
+ -0.15632975101470947,
+ -0.5647521615028381,
+ 0.0388534851372242,
+ 1.5328946113586426,
+ 0.5130590200424194,
+ -0.3926909863948822,
+ 0.46294018626213074,
+ 0.42535480856895447,
+ -0.19435879588127136,
+ -0.1934869885444641,
+ 0.029737945646047592,
+ -0.6507677435874939,
+ 0.15519340336322784,
+ 2.130070924758911,
+ -0.2242933064699173,
+ -0.3942379653453827,
+ 0.8409155011177063,
+ -0.5526861548423767,
+ -2.8079497814178467,
+ 0.13277432322502136,
+ -0.04973578080534935,
+ -0.17795336246490479,
+ 0.09016696363687515,
+ -2.7635128498077393,
+ -0.5937476754188538,
+ -1.7419863939285278,
+ -0.6703135371208191,
+ -0.7442510724067688,
+ 1.0544700622558594,
+ -0.614046037197113,
+ -0.14987458288669586,
+ -0.23138372600078583,
+ -0.06697559356689453,
+ 0.4892526865005493,
+ -1.0696855783462524,
+ -0.09890517592430115,
+ -0.12672454118728638,
+ 0.08710630238056183,
+ -0.14931896328926086,
+ 0.34537580609321594,
+ -0.26897406578063965,
+ 0.9914587736129761,
+ 0.0023401814978569746,
+ 0.3859562575817108,
+ 0.26136720180511475,
+ -3.4191715717315674,
+ 0.12051139771938324,
+ -1.924551248550415,
+ -1.0318630933761597,
+ 2.1765992641448975,
+ -0.176780566573143,
+ 1.68215012550354,
+ -2.0660247802734375,
+ 0.10290340334177017,
+ 0.23090112209320068,
+ 0.21758340299129486,
+ -0.03828301653265953,
+ 0.04786951094865799,
+ 0.6170095801353455,
+ 0.21853074431419373,
+ 1.5742398500442505,
+ -0.31364452838897705,
+ -0.23484764993190765,
+ 0.11317507922649384,
+ 0.3531038761138916,
+ -0.24483861029148102,
+ -0.2011033594608307,
+ 2.19809627532959,
+ 0.40214434266090393,
+ -0.02699330449104309,
+ 1.0313245058059692,
+ -0.1514972448348999,
+ 0.3261636197566986,
+ 0.23002180457115173,
+ 1.2791661024093628,
+ 0.07568946480751038,
+ -3.383969306945801,
+ -1.2306833267211914,
+ -0.13302189111709595,
+ -0.06789056211709976,
+ -0.4665857255458832,
+ -0.09816668927669525,
+ 0.3828674554824829,
+ -0.5048136115074158,
+ -0.34226295351982117,
+ -0.09143981337547302,
+ 0.2877776324748993,
+ -2.8805155754089355,
+ 0.046335142105817795,
+ -1.0330597162246704,
+ 0.10214007645845413,
+ -2.3693912029266357
+ ],
+ "parity": {
+ "fp32": {
+ "maxAbsDiff": 0.0016391277313232422,
+ "maxAbsReference": 4.916708469390869
+ },
+ "f16w": {
+ "maxAbsDiff": 0.006482720375061035,
+ "maxAbsReference": 4.916708469390869
+ }
+ }
+ }
+ },
+ "files": [
+ {
+ "task": "classifier",
+ "stage": "col",
+ "variant": "f16w",
+ "file": "classifier-col.f16w.onnx",
+ "bytes": 1881044,
+ "sha256": "35a0cde45113e557ec25ff7300d94f8c1114f725517cc36654e004e8280faa51",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "X",
+ "shape": [
+ "columns",
+ "rows",
+ 3
+ ]
+ },
+ {
+ "name": "y",
+ "shape": [
+ "train"
+ ]
+ },
+ {
+ "name": "keys",
+ "shape": [
+ "keys"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "E",
+ "shape": [
+ "columns",
+ "rows",
+ 128
+ ]
+ }
+ ]
+ },
+ {
+ "task": "classifier",
+ "stage": "hop",
+ "variant": "f16w",
+ "file": "classifier-hop.f16w.onnx",
+ "bytes": 4209838,
+ "sha256": "5a9bb35380493127f4a76adf44a19a33a9a69d1b58343f38ecdd4f77792d9b4f",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "x",
+ "shape": [
+ "nodes",
+ 512
+ ]
+ },
+ {
+ "name": "src",
+ "shape": [
+ "edges"
+ ]
+ },
+ {
+ "name": "dst",
+ "shape": [
+ "edges"
+ ]
+ },
+ {
+ "name": "edge_type",
+ "shape": [
+ "edges"
+ ]
+ },
+ {
+ "name": "noise",
+ "shape": [
+ "types",
+ 512
+ ]
+ },
+ {
+ "name": "deg",
+ "shape": [
+ "nodes",
+ 1
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "h",
+ "shape": [
+ "nodes",
+ 512
+ ]
+ }
+ ]
+ },
+ {
+ "task": "classifier",
+ "stage": "icl",
+ "variant": "f16w",
+ "file": "classifier-icl.f16w.onnx",
+ "bytes": 52943138,
+ "sha256": "e161ab3388e4fb622513c5c8a6b14f73cdba4575b4c16dfb1d9a1b518379f27f",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "Z",
+ "shape": [
+ "tokens",
+ 512
+ ]
+ },
+ {
+ "name": "y",
+ "shape": [
+ "train"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "out",
+ "shape": [
+ "queries",
+ 10
+ ]
+ }
+ ]
+ },
+ {
+ "task": "classifier",
+ "stage": "out",
+ "variant": "f16w",
+ "file": "classifier-out.f16w.onnx",
+ "bytes": 531196,
+ "sha256": "7acf1c5378cb5e2edf67b8040a440914e0cd7d8a2081737b2e863ac8fc566b5a",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "x",
+ "shape": [
+ "readouts",
+ 512
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "z",
+ "shape": [
+ "readouts",
+ 512
+ ]
+ }
+ ]
+ },
+ {
+ "task": "classifier",
+ "stage": "row",
+ "variant": "f16w",
+ "file": "classifier-row.f16w.onnx",
+ "bytes": 854772,
+ "sha256": "87b5b9cd288b251bee4b84a056c123941407174a7d0c887a1e386bf326aad3f5",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "E",
+ "shape": [
+ "columns",
+ "rows",
+ 128
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "Z",
+ "shape": [
+ "rows",
+ 512
+ ]
+ }
+ ]
+ },
+ {
+ "task": "classifier",
+ "stage": "col",
+ "variant": "fp32",
+ "file": "classifier-col.fp32.onnx",
+ "bytes": 3613403,
+ "sha256": "4258b08a30fa4eec00b5804f6c575b4574584ad6e1a311f9b211030a3533ea6f",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "X",
+ "shape": [
+ "columns",
+ "rows",
+ 3
+ ]
+ },
+ {
+ "name": "y",
+ "shape": [
+ "train"
+ ]
+ },
+ {
+ "name": "keys",
+ "shape": [
+ "keys"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "E",
+ "shape": [
+ "columns",
+ "rows",
+ 128
+ ]
+ }
+ ]
+ },
+ {
+ "task": "classifier",
+ "stage": "hop",
+ "variant": "fp32",
+ "file": "classifier-hop.fp32.onnx",
+ "bytes": 8403886,
+ "sha256": "a52f544ef90eb6de61c517e7e61c7fe63a6e3286a0a0a20356eae2e169616333",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "x",
+ "shape": [
+ "nodes",
+ 512
+ ]
+ },
+ {
+ "name": "src",
+ "shape": [
+ "edges"
+ ]
+ },
+ {
+ "name": "dst",
+ "shape": [
+ "edges"
+ ]
+ },
+ {
+ "name": "edge_type",
+ "shape": [
+ "edges"
+ ]
+ },
+ {
+ "name": "noise",
+ "shape": [
+ "types",
+ 512
+ ]
+ },
+ {
+ "name": "deg",
+ "shape": [
+ "nodes",
+ 1
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "h",
+ "shape": [
+ "nodes",
+ 512
+ ]
+ }
+ ]
+ },
+ {
+ "task": "classifier",
+ "stage": "icl",
+ "variant": "fp32",
+ "file": "classifier-icl.fp32.onnx",
+ "bytes": 105355516,
+ "sha256": "6a0f29920552b068feb889bbc4aa58ebf3ce8514b8c29c071a6a7baabd6a5b1b",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "Z",
+ "shape": [
+ "tokens",
+ 512
+ ]
+ },
+ {
+ "name": "y",
+ "shape": [
+ "train"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "out",
+ "shape": [
+ "queries",
+ 10
+ ]
+ }
+ ]
+ },
+ {
+ "task": "classifier",
+ "stage": "out",
+ "variant": "fp32",
+ "file": "classifier-out.fp32.onnx",
+ "bytes": 1055415,
+ "sha256": "5bbde304395d3eb511173a82fbc8ac3ffdf11c9d933c5ebc7df7ab439d0e38bf",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "x",
+ "shape": [
+ "readouts",
+ 512
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "z",
+ "shape": [
+ "readouts",
+ 512
+ ]
+ }
+ ]
+ },
+ {
+ "task": "classifier",
+ "stage": "row",
+ "variant": "fp32",
+ "file": "classifier-row.fp32.onnx",
+ "bytes": 1640460,
+ "sha256": "58f6cb77884920c4e92813cb785fa8a95567a94814a2f5d7ae8a4827c214e328",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "E",
+ "shape": [
+ "columns",
+ "rows",
+ 128
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "Z",
+ "shape": [
+ "rows",
+ 512
+ ]
+ }
+ ]
+ },
+ {
+ "task": "regressor",
+ "stage": "col",
+ "variant": "f16w",
+ "file": "regressor-col.f16w.onnx",
+ "bytes": 1870902,
+ "sha256": "b82598570e8dca41ecb38371f0b80588d82cd115bddc72324f43e138ef27ed1e",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "X",
+ "shape": [
+ "columns",
+ "rows",
+ 3
+ ]
+ },
+ {
+ "name": "y",
+ "shape": [
+ "train"
+ ]
+ },
+ {
+ "name": "keys",
+ "shape": [
+ "keys"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "E",
+ "shape": [
+ "columns",
+ "rows",
+ 128
+ ]
+ }
+ ]
+ },
+ {
+ "task": "regressor",
+ "stage": "hop",
+ "variant": "f16w",
+ "file": "regressor-hop.f16w.onnx",
+ "bytes": 4209838,
+ "sha256": "d0fe80252bd14032e060beb5d46fc9cb7f5f98458cbd6bfbef90568229c50414",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "x",
+ "shape": [
+ "nodes",
+ 512
+ ]
+ },
+ {
+ "name": "src",
+ "shape": [
+ "edges"
+ ]
+ },
+ {
+ "name": "dst",
+ "shape": [
+ "edges"
+ ]
+ },
+ {
+ "name": "edge_type",
+ "shape": [
+ "edges"
+ ]
+ },
+ {
+ "name": "noise",
+ "shape": [
+ "types",
+ 512
+ ]
+ },
+ {
+ "name": "deg",
+ "shape": [
+ "nodes",
+ 1
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "h",
+ "shape": [
+ "nodes",
+ 512
+ ]
+ }
+ ]
+ },
+ {
+ "task": "regressor",
+ "stage": "icl",
+ "variant": "f16w",
+ "file": "regressor-icl.f16w.onnx",
+ "bytes": 54911832,
+ "sha256": "63b6a9969914ca2f3aa3c4684fb75c71a84f7d2536a9e87674f4771fcef14b9e",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "Z",
+ "shape": [
+ "tokens",
+ 512
+ ]
+ },
+ {
+ "name": "y",
+ "shape": [
+ "train"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "out",
+ "shape": [
+ "queries",
+ 999
+ ]
+ }
+ ]
+ },
+ {
+ "task": "regressor",
+ "stage": "out",
+ "variant": "f16w",
+ "file": "regressor-out.f16w.onnx",
+ "bytes": 531196,
+ "sha256": "4f7f345e5c0632bb8318aee8756cb470a0fc3888db32ee55e3bb398726d0b13f",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "x",
+ "shape": [
+ "readouts",
+ 512
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "z",
+ "shape": [
+ "readouts",
+ 512
+ ]
+ }
+ ]
+ },
+ {
+ "task": "regressor",
+ "stage": "row",
+ "variant": "f16w",
+ "file": "regressor-row.f16w.onnx",
+ "bytes": 850342,
+ "sha256": "18a9cf22082b5c71eb463e20b8a0ff2ca8b2a018afc6492431e869970d924fcc",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "E",
+ "shape": [
+ "columns",
+ "rows",
+ 128
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "Z",
+ "shape": [
+ "rows",
+ 512
+ ]
+ }
+ ]
+ },
+ {
+ "task": "regressor",
+ "stage": "col",
+ "variant": "fp32",
+ "file": "regressor-col.fp32.onnx",
+ "bytes": 3600764,
+ "sha256": "b0feb762e03f66e381ffb12226b62af4e41b08ad2a628baee193b3bb4a1ab997",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "X",
+ "shape": [
+ "columns",
+ "rows",
+ 3
+ ]
+ },
+ {
+ "name": "y",
+ "shape": [
+ "train"
+ ]
+ },
+ {
+ "name": "keys",
+ "shape": [
+ "keys"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "E",
+ "shape": [
+ "columns",
+ "rows",
+ 128
+ ]
+ }
+ ]
+ },
+ {
+ "task": "regressor",
+ "stage": "hop",
+ "variant": "fp32",
+ "file": "regressor-hop.fp32.onnx",
+ "bytes": 8403886,
+ "sha256": "f824b39638fe07d8b9aacd6e310be540c31390071c510daec50fedc4abb52f02",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "x",
+ "shape": [
+ "nodes",
+ 512
+ ]
+ },
+ {
+ "name": "src",
+ "shape": [
+ "edges"
+ ]
+ },
+ {
+ "name": "dst",
+ "shape": [
+ "edges"
+ ]
+ },
+ {
+ "name": "edge_type",
+ "shape": [
+ "edges"
+ ]
+ },
+ {
+ "name": "noise",
+ "shape": [
+ "types",
+ 512
+ ]
+ },
+ {
+ "name": "deg",
+ "shape": [
+ "nodes",
+ 1
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "h",
+ "shape": [
+ "nodes",
+ 512
+ ]
+ }
+ ]
+ },
+ {
+ "task": "regressor",
+ "stage": "icl",
+ "variant": "fp32",
+ "file": "regressor-icl.fp32.onnx",
+ "bytes": 109339507,
+ "sha256": "97c3e0277eec6e9f1f01cd8f0e126c86745b0751f83805124151568112450847",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "Z",
+ "shape": [
+ "tokens",
+ 512
+ ]
+ },
+ {
+ "name": "y",
+ "shape": [
+ "train"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "out",
+ "shape": [
+ "queries",
+ 999
+ ]
+ }
+ ]
+ },
+ {
+ "task": "regressor",
+ "stage": "out",
+ "variant": "fp32",
+ "file": "regressor-out.fp32.onnx",
+ "bytes": 1055415,
+ "sha256": "de9b4ee60e328b8c6cf61924f0e4abb89968e900ed1596e80c3e355a5ad26386",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "x",
+ "shape": [
+ "readouts",
+ 512
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "z",
+ "shape": [
+ "readouts",
+ 512
+ ]
+ }
+ ]
+ },
+ {
+ "task": "regressor",
+ "stage": "row",
+ "variant": "fp32",
+ "file": "regressor-row.fp32.onnx",
+ "bytes": 1636030,
+ "sha256": "5ffa74477d4aa2a33fb2256d8cb053ccb8158e9da330b2317e6e02b113106c11",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "E",
+ "shape": [
+ "columns",
+ "rows",
+ 128
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "Z",
+ "shape": [
+ "rows",
+ 512
+ ]
+ }
+ ]
+ }
+ ]
+}
diff --git a/rfm/regressor-col.f16w.onnx b/rfm/regressor-col.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..93041d15fc455ef33b79bfec43acad999f7e8de2
--- /dev/null
+++ b/rfm/regressor-col.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b82598570e8dca41ecb38371f0b80588d82cd115bddc72324f43e138ef27ed1e
+size 1870902
diff --git a/rfm/regressor-col.fp32.onnx b/rfm/regressor-col.fp32.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..e076b51c25dadc55b0caf2305c253fae3b2301ab
--- /dev/null
+++ b/rfm/regressor-col.fp32.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0feb762e03f66e381ffb12226b62af4e41b08ad2a628baee193b3bb4a1ab997
+size 3600764
diff --git a/rfm/regressor-hop.f16w.onnx b/rfm/regressor-hop.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..e702bcdd96592799b0063ba04ee5d2cd8ef563d2
--- /dev/null
+++ b/rfm/regressor-hop.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0fe80252bd14032e060beb5d46fc9cb7f5f98458cbd6bfbef90568229c50414
+size 4209838
diff --git a/rfm/regressor-hop.fp32.onnx b/rfm/regressor-hop.fp32.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..3b64743636226ff2f987b64ece9bf7a25d881d9d
--- /dev/null
+++ b/rfm/regressor-hop.fp32.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f824b39638fe07d8b9aacd6e310be540c31390071c510daec50fedc4abb52f02
+size 8403886
diff --git a/rfm/regressor-icl.f16w.onnx b/rfm/regressor-icl.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..72a4b05d4705dba4b994a395809f107e58656bc0
--- /dev/null
+++ b/rfm/regressor-icl.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63b6a9969914ca2f3aa3c4684fb75c71a84f7d2536a9e87674f4771fcef14b9e
+size 54911832
diff --git a/rfm/regressor-icl.fp32.onnx b/rfm/regressor-icl.fp32.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..0b8005d8626c835bbccbd000421b75364c6dfe29
--- /dev/null
+++ b/rfm/regressor-icl.fp32.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97c3e0277eec6e9f1f01cd8f0e126c86745b0751f83805124151568112450847
+size 109339507
diff --git a/rfm/regressor-out.f16w.onnx b/rfm/regressor-out.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..aa26f12bb4724e830cba328dd3a165da7047b8bd
--- /dev/null
+++ b/rfm/regressor-out.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f7f345e5c0632bb8318aee8756cb470a0fc3888db32ee55e3bb398726d0b13f
+size 531196
diff --git a/rfm/regressor-out.fp32.onnx b/rfm/regressor-out.fp32.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..d97d5c1ede3b41b6a151bf1c65d098db38a88cf5
--- /dev/null
+++ b/rfm/regressor-out.fp32.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de9b4ee60e328b8c6cf61924f0e4abb89968e900ed1596e80c3e355a5ad26386
+size 1055415
diff --git a/rfm/regressor-row.f16w.onnx b/rfm/regressor-row.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..b7c282813a7155ed0e02064ce408ae8709b59e0a
--- /dev/null
+++ b/rfm/regressor-row.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18a9cf22082b5c71eb463e20b8a0ff2ca8b2a018afc6492431e869970d924fcc
+size 850342
diff --git a/rfm/regressor-row.fp32.onnx b/rfm/regressor-row.fp32.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..e54a3fe7bd28929e3b73737821a11468d3c7764c
--- /dev/null
+++ b/rfm/regressor-row.fp32.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5ffa74477d4aa2a33fb2256d8cb053ccb8158e9da330b2317e6e02b113106c11
+size 1636030
diff --git a/tabicl/LICENSE b/tabicl/LICENSE
new file mode 100644
index 0000000000000000000000000000000000000000..7696701075fac51c45517a103cd629c8bdda58e3
--- /dev/null
+++ b/tabicl/LICENSE
@@ -0,0 +1,28 @@
+BSD 3-Clause License
+
+Copyright (c) 2025, Soda team @ Inria
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+1. Redistributions of source code must retain the above copyright notice, this
+ list of conditions and the following disclaimer.
+
+2. Redistributions in binary form must reproduce the above copyright notice,
+ this list of conditions and the following disclaimer in the documentation
+ and/or other materials provided with the distribution.
+
+3. Neither the name of the copyright holder nor the names of its
+ contributors may be used to endorse or promote products derived from
+ this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
diff --git a/tabicl/README.md b/tabicl/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..5d1f7f9d152d7630ae08fb8a236d5559cc59dfdd
--- /dev/null
+++ b/tabicl/README.md
@@ -0,0 +1,118 @@
+---
+license: bsd-3-clause
+library_name: onnx
+base_model: jingang/TabICL
+base_model_relation: quantized
+pipeline_tag: tabular-classification
+tags:
+ - onnx
+ - onnxruntime-web
+ - webgpu
+ - tabicl
+ - tabular-classification
+ - tabular-regression
+ - tabular-foundation-model
+ - in-context-learning
+---
+
+# TabICLv2 for onnxruntime-web
+
+An unofficial ONNX conversion of [TabICLv2](https://huggingface.co/jingang/TabICL)
+by Jingang Qu, David Holzmüller, Gaël Varoquaux and Marine Le Morvan (Soda
+team, Inria), made for Tabular Data Workbench, which runs it in the browser
+with onnxruntime-web on WebGPU or WebAssembly. The weights are those of the
+upstream checkpoints: unchanged in the `fp32` variant, and with the larger
+ones rounded to float16 in the `f16w` variant. The Soda team has not reviewed
+or endorsed this conversion.
+
+## Files
+
+Each task (`classifier`, `regressor`) is cut into three graphs:
+
+| file | inputs | output |
+| --- | --- | --- |
+| `-col..onnx` | `X` (batch, columns, rows, 3) feature-group columns, `y` (batch, digits, train) | `E` (batch, columns, rows, 128) |
+| `-row..onnx` | `E` (batch, columns, rows, 128) | `R` (batch, rows, 512) |
+| `-icl..onnx` | `R` (batch, rows, 512), `y` (batch, train) | `out` (batch, test, 10 logits or 999 quantiles) |
+
+Each graph comes in two variants (``):
+
+- `fp32`: the checkpoint's weights as they are (225 MB for both tasks).
+- `f16w`: every float32 weight of 1,024 or more elements stored as float16
+ and cast back to float32 on load (113 MB). Compute stays float32, so it runs
+ on the WebGPU and WebAssembly providers alike.
+
+`manifest.json` lists every file with its size, sha256, inputs and outputs,
+and the int64 shape nodes (`cpuNodes`, one in each `row` graph) that the app
+pins to the CPU provider on WebGPU. The graphs contain no int64 shape
+arithmetic, which onnxruntime-web's WebGPU provider has been seen to corrupt.
+
+The graphs are the network only. Preprocessing (imputation, the
+normalisation choices, outlier clipping), the ensemble of column and class
+permutations, the mixed-radix digits for more than 10 classes and the
+averaging stay outside them, as in the reference package. The app ports
+those to TypeScript; `models/tabicl/pipeline.py` in Tabular Data Workbench is
+the numpy reference.
+
+## How they were made
+
+Source: `jingang/TabICL` at revision `4dcd344ece2c00be9e831fdd35bed57b5ad83e19`
+(`tabicl-classifier-v2-20260212.ckpt`, `tabicl-regressor-v2-20260212.ckpt`),
+with the `tabicl` 2.2.0 package's model code.
+
+```sh
+cd models/tabicl
+uv run export.py # writes public/models/tabicl/
+uv run verify.py # parity table against tabicl
+```
+
+`export.py` (in `models/tabicl/` of Tabular Data Workbench) splits the
+network in `exportable.py`, exports each part with `torch.onnx` (dynamo,
+opset 18, PyTorch 2.14 on CPU), removes the exporter's node metadata and doc
+strings, and checks the split graphs against tabicl's own forward pass before
+writing the manifest.
+
+## Parity
+
+`verify.py` runs `TabICLClassifier` and `TabICLRegressor` (tabicl 2.2.0, CPU,
+float32, 8 estimators, seed 42), replays the ensemble it drew through the
+ONNX graphs in onnxruntime 1.30 (CPU) and compares the final predictions:
+
+| dataset | fp32 | f16w |
+| --- | --- | --- |
+| breast_cancer (probabilities) | 2.8e-6 | 5.8e-4 |
+| breast_cancer with 5% missing | 1.6e-6 | 3.4e-4 |
+| wine | 8.3e-7 | 2.8e-4 |
+| 14 classes (synthetic) | 1.5e-6 | 7.1e-4 |
+| diabetes (mean, in units of the target's sd) | 2.3e-6 | 4.7e-4 |
+| diabetes (quantiles, in units of the target's sd) | 2.4e-6 | 6.4e-4 |
+
+The figures are maximum absolute differences. Predicted classes agree with
+tabicl on every row of every dataset in both variants.
+
+## Licence
+
+BSD-3-Clause, the licence of TabICL's code and of the v2 checkpoints
+(`license: bsd-3-clause` on `jingang/TabICL`). The licence text with the
+copyright notice (Copyright (c) 2025, Soda team @ Inria) is in `LICENSE`
+beside the graphs and must stay with them. The licence does not allow the
+names of Inria or the Soda team to be used to endorse or promote products
+derived from these files.
+
+## Citation
+
+```bibtex
+@inproceedings{qu2025tabicl,
+ title={Tab{ICL}: {A} Tabular Foundation Model for In-Context Learning on Large Data},
+ author={Qu, Jingang and Holzm{\"u}ller, David and Varoquaux, Ga{\"e}l and Le Morvan, Marine},
+ booktitle={International Conference on Machine Learning},
+ year={2025}
+}
+
+@article{qu2026tabiclv2,
+ title={{TabICLv2}: {A} better, faster, scalable, and open tabular foundation model},
+ author={Qu, Jingang and Holzm{\"u}ller, David and Varoquaux, Ga{\"e}l and Le Morvan, Marine},
+ journal={arXiv preprint arXiv:2602.11139},
+ year={2026}
+}
+```
diff --git a/tabicl/classifier-col.f16w.onnx b/tabicl/classifier-col.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..6429840d779258a0392badc776b6a6f7fa570153
--- /dev/null
+++ b/tabicl/classifier-col.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a94e3998566acf9513e27ff38d5e8f7b9873bd8751d86192fd308ec426e7dd1c
+size 1894277
diff --git a/tabicl/classifier-col.fp32.onnx b/tabicl/classifier-col.fp32.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..3b13d72f1ba253a2bda014ae3b82210851bebd0d
--- /dev/null
+++ b/tabicl/classifier-col.fp32.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb7221f89b5d73f4adc1b337bf0eaba7ab362c2494c33ce1f0f3549a2fa2b020
+size 3626570
diff --git a/tabicl/classifier-icl.f16w.onnx b/tabicl/classifier-icl.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..ec6bbcc472fb2a7b0242e38f77714e79bd78d243
--- /dev/null
+++ b/tabicl/classifier-icl.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:21791e2a7303c4d48611201f42f59f2ad69cf6ec5e6f528785234b889934e1a8
+size 52917443
diff --git a/tabicl/classifier-icl.fp32.onnx b/tabicl/classifier-icl.fp32.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..07c9614bcca2dd7595cd5eb89eb7dc24af842c71
--- /dev/null
+++ b/tabicl/classifier-icl.fp32.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7af95b18d1b45c15ebe26b89f0e066130ef6ee9adc4a3263bc8250f52b372260
+size 105366407
diff --git a/tabicl/classifier-row.f16w.onnx b/tabicl/classifier-row.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..04faf6f37fe525c2e5bc0ae99570f7e02d470b69
--- /dev/null
+++ b/tabicl/classifier-row.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:177bffd02a588f58e25d0ed0269221214863109fc89a21966f1b9c9673bf8601
+size 855980
diff --git a/tabicl/classifier-row.fp32.onnx b/tabicl/classifier-row.fp32.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..297fd79de5b4af5bbd11f24373ec44112ac913e0
--- /dev/null
+++ b/tabicl/classifier-row.fp32.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:316b5f0ce0aa2559731693aae9f126d941e4e1b01b9d9323e35df095b27be397
+size 1641650
diff --git a/tabicl/manifest.json b/tabicl/manifest.json
new file mode 100644
index 0000000000000000000000000000000000000000..96911a81e387080e83e30e7bb26cf288cac8d18b
--- /dev/null
+++ b/tabicl/manifest.json
@@ -0,0 +1,484 @@
+{
+ "model": "TabICLv2",
+ "source": {
+ "package": "tabicl 2.2.0",
+ "repo": "jingang/TabICL",
+ "revision": "4dcd344ece2c00be9e831fdd35bed57b5ad83e19",
+ "checkpoints": {
+ "classifier": "tabicl-classifier-v2-20260212.ckpt",
+ "regressor": "tabicl-regressor-v2-20260212.ckpt"
+ },
+ "license": "BSD-3-Clause"
+ },
+ "opset": 18,
+ "exporter": "torch.onnx (dynamo) 2.14.0+cpu",
+ "limits": {
+ "maxClasses": 10,
+ "numQuantiles": 999,
+ "featureGroupSize": 3,
+ "cellDim": 128,
+ "rowDim": 512
+ },
+ "files": [
+ {
+ "task": "classifier",
+ "stage": "col",
+ "variant": "f16w",
+ "file": "classifier-col.f16w.onnx",
+ "bytes": 1894277,
+ "sha256": "a94e3998566acf9513e27ff38d5e8f7b9873bd8751d86192fd308ec426e7dd1c",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "X",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "columns",
+ "rows",
+ 3
+ ]
+ },
+ {
+ "name": "y",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "digits",
+ "train"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "E",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "columns",
+ "rows",
+ 128
+ ]
+ }
+ ]
+ },
+ {
+ "task": "classifier",
+ "stage": "icl",
+ "variant": "f16w",
+ "file": "classifier-icl.f16w.onnx",
+ "bytes": 52917443,
+ "sha256": "21791e2a7303c4d48611201f42f59f2ad69cf6ec5e6f528785234b889934e1a8",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "R",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "rows",
+ 512
+ ]
+ },
+ {
+ "name": "y",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "train"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "out",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "test",
+ 10
+ ]
+ }
+ ]
+ },
+ {
+ "task": "classifier",
+ "stage": "row",
+ "variant": "f16w",
+ "file": "classifier-row.f16w.onnx",
+ "bytes": 855980,
+ "sha256": "177bffd02a588f58e25d0ed0269221214863109fc89a21966f1b9c9673bf8601",
+ "cpuNodes": [
+ "node_Concat_7"
+ ],
+ "inputs": [
+ {
+ "name": "E",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "columns",
+ "rows",
+ 128
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "R",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "rows",
+ 512
+ ]
+ }
+ ]
+ },
+ {
+ "task": "classifier",
+ "stage": "col",
+ "variant": "fp32",
+ "file": "classifier-col.fp32.onnx",
+ "bytes": 3626570,
+ "sha256": "bb7221f89b5d73f4adc1b337bf0eaba7ab362c2494c33ce1f0f3549a2fa2b020",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "X",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "columns",
+ "rows",
+ 3
+ ]
+ },
+ {
+ "name": "y",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "digits",
+ "train"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "E",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "columns",
+ "rows",
+ 128
+ ]
+ }
+ ]
+ },
+ {
+ "task": "classifier",
+ "stage": "icl",
+ "variant": "fp32",
+ "file": "classifier-icl.fp32.onnx",
+ "bytes": 105366407,
+ "sha256": "7af95b18d1b45c15ebe26b89f0e066130ef6ee9adc4a3263bc8250f52b372260",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "R",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "rows",
+ 512
+ ]
+ },
+ {
+ "name": "y",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "train"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "out",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "test",
+ 10
+ ]
+ }
+ ]
+ },
+ {
+ "task": "classifier",
+ "stage": "row",
+ "variant": "fp32",
+ "file": "classifier-row.fp32.onnx",
+ "bytes": 1641650,
+ "sha256": "316b5f0ce0aa2559731693aae9f126d941e4e1b01b9d9323e35df095b27be397",
+ "cpuNodes": [
+ "node_Concat_7"
+ ],
+ "inputs": [
+ {
+ "name": "E",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "columns",
+ "rows",
+ 128
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "R",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "rows",
+ 512
+ ]
+ }
+ ]
+ },
+ {
+ "task": "regressor",
+ "stage": "col",
+ "variant": "f16w",
+ "file": "regressor-col.f16w.onnx",
+ "bytes": 1883270,
+ "sha256": "d434c1c37b5e0f53e8052869ba93682f42da4f3eb0f1154bda95b40ab6807436",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "X",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "columns",
+ "rows",
+ 3
+ ]
+ },
+ {
+ "name": "y",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "digits",
+ "train"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "E",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "columns",
+ "rows",
+ 128
+ ]
+ }
+ ]
+ },
+ {
+ "task": "regressor",
+ "stage": "icl",
+ "variant": "f16w",
+ "file": "regressor-icl.f16w.onnx",
+ "bytes": 54884074,
+ "sha256": "e2f25c737dee1524687c758487e85c38768984b5ec027ffb318da08fb54c454d",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "R",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "rows",
+ 512
+ ]
+ },
+ {
+ "name": "y",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "train"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "out",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "test",
+ 999
+ ]
+ }
+ ]
+ },
+ {
+ "task": "regressor",
+ "stage": "row",
+ "variant": "f16w",
+ "file": "regressor-row.f16w.onnx",
+ "bytes": 851555,
+ "sha256": "280e72146067baaa2516ffa8483be2c368073fa4171f765a08ac2e8e4c801d8b",
+ "cpuNodes": [
+ "node_Concat_7"
+ ],
+ "inputs": [
+ {
+ "name": "E",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "columns",
+ "rows",
+ 128
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "R",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "rows",
+ 512
+ ]
+ }
+ ]
+ },
+ {
+ "task": "regressor",
+ "stage": "col",
+ "variant": "fp32",
+ "file": "regressor-col.fp32.onnx",
+ "bytes": 3613048,
+ "sha256": "2ea2113646278399b1e58beb57d3e46d4496ee17e9c3871bafd4985dc92f4300",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "X",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "columns",
+ "rows",
+ 3
+ ]
+ },
+ {
+ "name": "y",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "digits",
+ "train"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "E",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "columns",
+ "rows",
+ 128
+ ]
+ }
+ ]
+ },
+ {
+ "task": "regressor",
+ "stage": "icl",
+ "variant": "fp32",
+ "file": "regressor-icl.fp32.onnx",
+ "bytes": 109348315,
+ "sha256": "17bb4ca71d88aea44b3f8886e376c41d382ded9ef71aafc594ae3b736a3b8916",
+ "cpuNodes": [],
+ "inputs": [
+ {
+ "name": "R",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "rows",
+ 512
+ ]
+ },
+ {
+ "name": "y",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "train"
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "out",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "test",
+ 999
+ ]
+ }
+ ]
+ },
+ {
+ "task": "regressor",
+ "stage": "row",
+ "variant": "fp32",
+ "file": "regressor-row.fp32.onnx",
+ "bytes": 1637225,
+ "sha256": "e03db9cdec62860e538afe61c6e1a6f7533a72756066da3660c508efb1f06559",
+ "cpuNodes": [
+ "node_Concat_7"
+ ],
+ "inputs": [
+ {
+ "name": "E",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "columns",
+ "rows",
+ 128
+ ]
+ }
+ ],
+ "outputs": [
+ {
+ "name": "R",
+ "type": "float32",
+ "shape": [
+ "batch",
+ "rows",
+ 512
+ ]
+ }
+ ]
+ }
+ ]
+}
diff --git a/tabicl/regressor-col.f16w.onnx b/tabicl/regressor-col.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..ebaee682a69bbca1e9e79c95911a8cf5d46ff24b
--- /dev/null
+++ b/tabicl/regressor-col.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d434c1c37b5e0f53e8052869ba93682f42da4f3eb0f1154bda95b40ab6807436
+size 1883270
diff --git a/tabicl/regressor-col.fp32.onnx b/tabicl/regressor-col.fp32.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..4c39911c0500e331dcb7c4cf7c54c5a736d98219
--- /dev/null
+++ b/tabicl/regressor-col.fp32.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ea2113646278399b1e58beb57d3e46d4496ee17e9c3871bafd4985dc92f4300
+size 3613048
diff --git a/tabicl/regressor-icl.f16w.onnx b/tabicl/regressor-icl.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..aba02e29e4a349c31682a422eb2475284a74fa93
--- /dev/null
+++ b/tabicl/regressor-icl.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e2f25c737dee1524687c758487e85c38768984b5ec027ffb318da08fb54c454d
+size 54884074
diff --git a/tabicl/regressor-icl.fp32.onnx b/tabicl/regressor-icl.fp32.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..3bac6e3b11fd612319fbc4506d450106d0a163e5
--- /dev/null
+++ b/tabicl/regressor-icl.fp32.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17bb4ca71d88aea44b3f8886e376c41d382ded9ef71aafc594ae3b736a3b8916
+size 109348315
diff --git a/tabicl/regressor-row.f16w.onnx b/tabicl/regressor-row.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..703f0f6f3b8ea75606d16c2eb75e09edda189339
--- /dev/null
+++ b/tabicl/regressor-row.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:280e72146067baaa2516ffa8483be2c368073fa4171f765a08ac2e8e4c801d8b
+size 851555
diff --git a/tabicl/regressor-row.fp32.onnx b/tabicl/regressor-row.fp32.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..849fa6d8310f8ce71e2822638983cc92bc20280c
--- /dev/null
+++ b/tabicl/regressor-row.fp32.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e03db9cdec62860e538afe61c6e1a6f7533a72756066da3660c508efb1f06559
+size 1637225
diff --git a/tsfm/LICENSE b/tsfm/LICENSE
new file mode 100644
index 0000000000000000000000000000000000000000..67db8588217f266eb561f75fae738656325deac9
--- /dev/null
+++ b/tsfm/LICENSE
@@ -0,0 +1,175 @@
+
+ Apache License
+ Version 2.0, January 2004
+ http://www.apache.org/licenses/
+
+ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+ 1. Definitions.
+
+ "License" shall mean the terms and conditions for use, reproduction,
+ and distribution as defined by Sections 1 through 9 of this document.
+
+ "Licensor" shall mean the copyright owner or entity authorized by
+ the copyright owner that is granting the License.
+
+ "Legal Entity" shall mean the union of the acting entity and all
+ other entities that control, are controlled by, or are under common
+ control with that entity. For the purposes of this definition,
+ "control" means (i) the power, direct or indirect, to cause the
+ direction or management of such entity, whether by contract or
+ otherwise, or (ii) ownership of fifty percent (50%) or more of the
+ outstanding shares, or (iii) beneficial ownership of such entity.
+
+ "You" (or "Your") shall mean an individual or Legal Entity
+ exercising permissions granted by this License.
+
+ "Source" form shall mean the preferred form for making modifications,
+ including but not limited to software source code, documentation
+ source, and configuration files.
+
+ "Object" form shall mean any form resulting from mechanical
+ transformation or translation of a Source form, including but
+ not limited to compiled object code, generated documentation,
+ and conversions to other media types.
+
+ "Work" shall mean the work of authorship, whether in Source or
+ Object form, made available under the License, as indicated by a
+ copyright notice that is included in or attached to the work
+ (an example is provided in the Appendix below).
+
+ "Derivative Works" shall mean any work, whether in Source or Object
+ form, that is based on (or derived from) the Work and for which the
+ editorial revisions, annotations, elaborations, or other modifications
+ represent, as a whole, an original work of authorship. For the purposes
+ of this License, Derivative Works shall not include works that remain
+ separable from, or merely link (or bind by name) to the interfaces of,
+ the Work and Derivative Works thereof.
+
+ "Contribution" shall mean any work of authorship, including
+ the original version of the Work and any modifications or additions
+ to that Work or Derivative Works thereof, that is intentionally
+ submitted to Licensor for inclusion in the Work by the copyright owner
+ or by an individual or Legal Entity authorized to submit on behalf of
+ the copyright owner. For the purposes of this definition, "submitted"
+ means any form of electronic, verbal, or written communication sent
+ to the Licensor or its representatives, including but not limited to
+ communication on electronic mailing lists, source code control systems,
+ and issue tracking systems that are managed by, or on behalf of, the
+ Licensor for the purpose of discussing and improving the Work, but
+ excluding communication that is conspicuously marked or otherwise
+ designated in writing by the copyright owner as "Not a Contribution."
+
+ "Contributor" shall mean Licensor and any individual or Legal Entity
+ on behalf of whom a Contribution has been received by Licensor and
+ subsequently incorporated within the Work.
+
+ 2. Grant of Copyright License. Subject to the terms and conditions of
+ this License, each Contributor hereby grants to You a perpetual,
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+ copyright license to reproduce, prepare Derivative Works of,
+ publicly display, publicly perform, sublicense, and distribute the
+ Work and such Derivative Works in Source or Object form.
+
+ 3. Grant of Patent License. Subject to the terms and conditions of
+ this License, each Contributor hereby grants to You a perpetual,
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+ (except as stated in this section) patent license to make, have made,
+ use, offer to sell, sell, import, and otherwise transfer the Work,
+ where such license applies only to those patent claims licensable
+ by such Contributor that are necessarily infringed by their
+ Contribution(s) alone or by combination of their Contribution(s)
+ with the Work to which such Contribution(s) was submitted. If You
+ institute patent litigation against any entity (including a
+ cross-claim or counterclaim in a lawsuit) alleging that the Work
+ or a Contribution incorporated within the Work constitutes direct
+ or contributory patent infringement, then any patent licenses
+ granted to You under this License for that Work shall terminate
+ as of the date such litigation is filed.
+
+ 4. Redistribution. You may reproduce and distribute copies of the
+ Work or Derivative Works thereof in any medium, with or without
+ modifications, and in Source or Object form, provided that You
+ meet the following conditions:
+
+ (a) You must give any other recipients of the Work or
+ Derivative Works a copy of this License; and
+
+ (b) You must cause any modified files to carry prominent notices
+ stating that You changed the files; and
+
+ (c) You must retain, in the Source form of any Derivative Works
+ that You distribute, all copyright, patent, trademark, and
+ attribution notices from the Source form of the Work,
+ excluding those notices that do not pertain to any part of
+ the Derivative Works; and
+
+ (d) If the Work includes a "NOTICE" text file as part of its
+ distribution, then any Derivative Works that You distribute must
+ include a readable copy of the attribution notices contained
+ within such NOTICE file, excluding those notices that do not
+ pertain to any part of the Derivative Works, in at least one
+ of the following places: within a NOTICE text file distributed
+ as part of the Derivative Works; within the Source form or
+ documentation, if provided along with the Derivative Works; or,
+ within a display generated by the Derivative Works, if and
+ wherever such third-party notices normally appear. The contents
+ of the NOTICE file are for informational purposes only and
+ do not modify the License. You may add Your own attribution
+ notices within Derivative Works that You distribute, alongside
+ or as an addendum to the NOTICE text from the Work, provided
+ that such additional attribution notices cannot be construed
+ as modifying the License.
+
+ You may add Your own copyright statement to Your modifications and
+ may provide additional or different license terms and conditions
+ for use, reproduction, or distribution of Your modifications, or
+ for any such Derivative Works as a whole, provided Your use,
+ reproduction, and distribution of the Work otherwise complies with
+ the conditions stated in this License.
+
+ 5. Submission of Contributions. Unless You explicitly state otherwise,
+ any Contribution intentionally submitted for inclusion in the Work
+ by You to the Licensor shall be under the terms and conditions of
+ this License, without any additional terms or conditions.
+ Notwithstanding the above, nothing herein shall supersede or modify
+ the terms of any separate license agreement you may have executed
+ with Licensor regarding such Contributions.
+
+ 6. Trademarks. This License does not grant permission to use the trade
+ names, trademarks, service marks, or product names of the Licensor,
+ except as required for reasonable and customary use in describing the
+ origin of the Work and reproducing the content of the NOTICE file.
+
+ 7. Disclaimer of Warranty. Unless required by applicable law or
+ agreed to in writing, Licensor provides the Work (and each
+ Contributor provides its Contributions) on an "AS IS" BASIS,
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+ implied, including, without limitation, any warranties or conditions
+ of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+ PARTICULAR PURPOSE. You are solely responsible for determining the
+ appropriateness of using or redistributing the Work and assume any
+ risks associated with Your exercise of permissions under this License.
+
+ 8. Limitation of Liability. In no event and under no legal theory,
+ whether in tort (including negligence), contract, or otherwise,
+ unless required by applicable law (such as deliberate and grossly
+ negligent acts) or agreed to in writing, shall any Contributor be
+ liable to You for damages, including any direct, indirect, special,
+ incidental, or consequential damages of any character arising as a
+ result of this License or out of the use or inability to use the
+ Work (including but not limited to damages for loss of goodwill,
+ work stoppage, computer failure or malfunction, or any and all
+ other commercial damages or losses), even if such Contributor
+ has been advised of the possibility of such damages.
+
+ 9. Accepting Warranty or Additional Liability. While redistributing
+ the Work or Derivative Works thereof, You may choose to offer,
+ and charge a fee for, acceptance of support, warranty, indemnity,
+ or other liability obligations and/or rights consistent with this
+ License. However, in accepting such obligations, You may act only
+ on Your own behalf and on Your sole responsibility, not on behalf
+ of any other Contributor, and only if You agree to indemnify,
+ defend, and hold each Contributor harmless for any liability
+ incurred by, or claims asserted against, such Contributor by reason
+ of your accepting any such warranty or additional liability.
diff --git a/tsfm/LICENSE.structured-data-models b/tsfm/LICENSE.structured-data-models
new file mode 100644
index 0000000000000000000000000000000000000000..001a3281169d6ade30cd1a44544029254ba490a8
--- /dev/null
+++ b/tsfm/LICENSE.structured-data-models
@@ -0,0 +1,204 @@
+Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+
+
+ Apache License
+ Version 2.0, January 2004
+ http://www.apache.org/licenses/
+
+ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+ 1. Definitions.
+
+ "License" shall mean the terms and conditions for use, reproduction,
+ and distribution as defined by Sections 1 through 9 of this document.
+
+ "Licensor" shall mean the copyright owner or entity authorized by
+ the copyright owner that is granting the License.
+
+ "Legal Entity" shall mean the union of the acting entity and all
+ other entities that control, are controlled by, or are under common
+ control with that entity. For the purposes of this definition,
+ "control" means (i) the power, direct or indirect, to cause the
+ direction or management of such entity, whether by contract or
+ otherwise, or (ii) ownership of fifty percent (50%) or more of the
+ outstanding shares, or (iii) beneficial ownership of such entity.
+
+ "You" (or "Your") shall mean an individual or Legal Entity
+ exercising permissions granted by this License.
+
+ "Source" form shall mean the preferred form for making modifications,
+ including but not limited to software source code, documentation
+ source, and configuration files.
+
+ "Object" form shall mean any form resulting from mechanical
+ transformation or translation of a Source form, including but
+ not limited to compiled object code, generated documentation,
+ and conversions to other media types.
+
+ "Work" shall mean the work of authorship, whether in Source or
+ Object form, made available under the License, as indicated by a
+ copyright notice that is included in or attached to the work
+ (an example is provided in the Appendix below).
+
+ "Derivative Works" shall mean any work, whether in Source or Object
+ form, that is based on (or derived from) the Work and for which the
+ editorial revisions, annotations, elaborations, or other modifications
+ represent, as a whole, an original work of authorship. For the purposes
+ of this License, Derivative Works shall not include works that remain
+ separable from, or merely link (or bind by name) to the interfaces of,
+ the Work and Derivative Works thereof.
+
+ "Contribution" shall mean any work of authorship, including
+ the original version of the Work and any modifications or additions
+ to that Work or Derivative Works thereof, that is intentionally
+ submitted to Licensor for inclusion in the Work by the copyright owner
+ or by an individual or Legal Entity authorized to submit on behalf of
+ the copyright owner. For the purposes of this definition, "submitted"
+ means any form of electronic, verbal, or written communication sent
+ to the Licensor or its representatives, including but not limited to
+ communication on electronic mailing lists, source code control systems,
+ and issue tracking systems that are managed by, or on behalf of, the
+ Licensor for the purpose of discussing and improving the Work, but
+ excluding communication that is conspicuously marked or otherwise
+ designated in writing by the copyright owner as "Not a Contribution."
+
+ "Contributor" shall mean Licensor and any individual or Legal Entity
+ on behalf of whom a Contribution has been received by Licensor and
+ subsequently incorporated within the Work.
+
+ 2. Grant of Copyright License. Subject to the terms and conditions of
+ this License, each Contributor hereby grants to You a perpetual,
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+ copyright license to reproduce, prepare Derivative Works of,
+ publicly display, publicly perform, sublicense, and distribute the
+ Work and such Derivative Works in Source or Object form.
+
+ 3. Grant of Patent License. Subject to the terms and conditions of
+ this License, each Contributor hereby grants to You a perpetual,
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+ (except as stated in this section) patent license to make, have made,
+ use, offer to sell, sell, import, and otherwise transfer the Work,
+ where such license applies only to those patent claims licensable
+ by such Contributor that are necessarily infringed by their
+ Contribution(s) alone or by combination of their Contribution(s)
+ with the Work to which such Contribution(s) was submitted. If You
+ institute patent litigation against any entity (including a
+ cross-claim or counterclaim in a lawsuit) alleging that the Work
+ or a Contribution incorporated within the Work constitutes direct
+ or contributory patent infringement, then any patent licenses
+ granted to You under this License for that Work shall terminate
+ as of the date such litigation is filed.
+
+ 4. Redistribution. You may reproduce and distribute copies of the
+ Work or Derivative Works thereof in any medium, with or without
+ modifications, and in Source or Object form, provided that You
+ meet the following conditions:
+
+ (a) You must give any other recipients of the Work or
+ Derivative Works a copy of this License; and
+
+ (b) You must cause any modified files to carry prominent notices
+ stating that You changed the files; and
+
+ (c) You must retain, in the Source form of any Derivative Works
+ that You distribute, all copyright, patent, trademark, and
+ attribution notices from the Source form of the Work,
+ excluding those notices that do not pertain to any part of
+ the Derivative Works; and
+
+ (d) If the Work includes a "NOTICE" text file as part of its
+ distribution, then any Derivative Works that You distribute must
+ include a readable copy of the attribution notices contained
+ within such NOTICE file, excluding those notices that do not
+ pertain to any part of the Derivative Works, in at least one
+ of the following places: within a NOTICE text file distributed
+ as part of the Derivative Works; within the Source form or
+ documentation, if provided along with the Derivative Works; or,
+ within a display generated by the Derivative Works, if and
+ wherever such third-party notices normally appear. The contents
+ of the NOTICE file are for informational purposes only and
+ do not modify the License. You may add Your own attribution
+ notices within Derivative Works that You distribute, alongside
+ or as an addendum to the NOTICE text from the Work, provided
+ that such additional attribution notices cannot be construed
+ as modifying the License.
+
+ You may add Your own copyright statement to Your modifications and
+ may provide additional or different license terms and conditions
+ for use, reproduction, or distribution of Your modifications, or
+ for any such Derivative Works as a whole, provided Your use,
+ reproduction, and distribution of the Work otherwise complies with
+ the conditions stated in this License.
+
+ 5. Submission of Contributions. Unless You explicitly state otherwise,
+ any Contribution intentionally submitted for inclusion in the Work
+ by You to the Licensor shall be under the terms and conditions of
+ this License, without any additional terms or conditions.
+ Notwithstanding the above, nothing herein shall supersede or modify
+ the terms of any separate license agreement you may have executed
+ with Licensor regarding such Contributions.
+
+ 6. Trademarks. This License does not grant permission to use the trade
+ names, trademarks, service marks, or product names of the Licensor,
+ except as required for reasonable and customary use in describing the
+ origin of the Work and reproducing the content of the NOTICE file.
+
+ 7. Disclaimer of Warranty. Unless required by applicable law or
+ agreed to in writing, Licensor provides the Work (and each
+ Contributor provides its Contributions) on an "AS IS" BASIS,
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+ implied, including, without limitation, any warranties or conditions
+ of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+ PARTICULAR PURPOSE. You are solely responsible for determining the
+ appropriateness of using or redistributing the Work and assume any
+ risks associated with Your exercise of permissions under this License.
+
+ 8. Limitation of Liability. In no event and under no legal theory,
+ whether in tort (including negligence), contract, or otherwise,
+ unless required by applicable law (such as deliberate and grossly
+ negligent acts) or agreed to in writing, shall any Contributor be
+ liable to You for damages, including any direct, indirect, special,
+ incidental, or consequential damages of any character arising as a
+ result of this License or out of the use or inability to use the
+ Work (including but not limited to damages for loss of goodwill,
+ work stoppage, computer failure or malfunction, or any and all
+ other commercial damages or losses), even if such Contributor
+ has been advised of the possibility of such damages.
+
+ 9. Accepting Warranty or Additional Liability. While redistributing
+ the Work or Derivative Works thereof, You may choose to offer,
+ and charge a fee for, acceptance of support, warranty, indemnity,
+ or other liability obligations and/or rights consistent with this
+ License. However, in accepting such obligations, You may act only
+ on Your own behalf and on Your sole responsibility, not on behalf
+ of any other Contributor, and only if You agree to indemnify,
+ defend, and hold each Contributor harmless for any liability
+ incurred by, or claims asserted against, such Contributor by reason
+ of your accepting any such warranty or additional liability.
+
+ END OF TERMS AND CONDITIONS
+
+ APPENDIX: How to apply the Apache License to your work.
+
+ To apply the Apache License to your work, attach the following
+ boilerplate notice, with the fields enclosed by brackets "[]"
+ replaced with your own identifying information. (Don't include
+ the brackets!) The text should be enclosed in the appropriate
+ comment syntax for the file format. We also recommend that a
+ file or class name and description of purpose be included on the
+ same "printed page" as the copyright notice for easier
+ identification within third-party archives.
+
+ Copyright [yyyy] [name of copyright owner]
+
+ Licensed under the Apache License, Version 2.0 (the "License");
+ you may not use this file except in compliance with the License.
+ You may obtain a copy of the License at
+
+ http://www.apache.org/licenses/LICENSE-2.0
+
+ Unless required by applicable law or agreed to in writing, software
+ distributed under the License is distributed on an "AS IS" BASIS,
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ See the License for the specific language governing permissions and
+ limitations under the License.
diff --git a/tsfm/NOTICE b/tsfm/NOTICE
new file mode 100644
index 0000000000000000000000000000000000000000..3a9eede6c634f5240081f13416d2e0283db20eaa
--- /dev/null
+++ b/tsfm/NOTICE
@@ -0,0 +1,32 @@
+Time-series foundation models converted to ONNX for Tabular Data Workbench.
+The weights in these files are the upstream weights; the graphs around them
+were re-exported, so the files are modified forms of the originals (see
+"Changes" below).
+
+Chronos-2 and Chronos-2 small
+ Source: amazon/chronos-2 (revision 29ec3766d36d6f73f0696f85560a422f50e8498c)
+ autogluon/chronos-2-small (revision ddec01313e50b6bc58ebaa92ede81bc24a3d9f9a)
+ https://github.com/amazon-science/chronos-forecasting
+ Copyright Amazon.com, Inc. or its affiliates. All Rights Reserved.
+ Licensed under the Apache License, Version 2.0 (LICENSE).
+
+Kumo-Anomaly (NV-Tesseract-AD)
+ Source: nvidia/Kumo-Anomaly (revision 226c5003b0582adc1cac433123f47324bfae39f2)
+ https://github.com/NVIDIA/structured-data-models
+ Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+ Licensed under the Apache License, Version 2.0 (LICENSE; NVIDIA's copy with
+ its copyright line is LICENSE.structured-data-models).
+ Kumo-TS includes DPM-Solver, Copyright (c) 2022 Cheng Lu, licensed under the
+ MIT License (THIRD-PARTY-NOTICES.txt).
+
+Changes
+ - chronos-2*.fp32.onnx: the Chronos-2 network re-exported with
+ torch.onnx (dynamo, opset 18). Instance scaling, the arcsinh transform,
+ NaN handling and the masks built from group ids are computed outside the
+ graph, and the graph contains no int64 shape arithmetic. Weights
+ unchanged.
+ - kumo-anomaly.fp32.onnx: Kumo-Anomaly's denoiser network re-exported the
+ same way. The diffusion sampling loop (DPM-Solver++) and the scoring run
+ outside the graph. Weights unchanged.
+ - *.f16w.onnx: the same graphs with every weight of 1,024 or more elements
+ stored as float16 and cast back to float32 on load.
diff --git a/tsfm/README.md b/tsfm/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..86fb1e6c05301702b2f0b9ba1e57f9722390375b
--- /dev/null
+++ b/tsfm/README.md
@@ -0,0 +1,128 @@
+---
+license: apache-2.0
+library_name: onnx
+base_model:
+ - amazon/chronos-2
+ - autogluon/chronos-2-small
+ - nvidia/Kumo-Anomaly
+base_model_relation: quantized
+pipeline_tag: time-series-forecasting
+tags:
+ - onnx
+ - onnxruntime-web
+ - webgpu
+ - time-series-forecasting
+ - anomaly-detection
+ - chronos
+ - time-series-foundation-model
+---
+
+# Chronos-2 and Kumo-Anomaly for onnxruntime-web
+
+Unofficial ONNX conversions of three time-series models, made for Tabular
+Data Workbench, which runs them in the browser with onnxruntime-web on WebGPU
+or WebAssembly:
+
+- [Chronos-2](https://huggingface.co/amazon/chronos-2) (120M parameters) and
+ [Chronos-2 small](https://huggingface.co/autogluon/chronos-2-small) (28M),
+ Amazon's forecasters.
+- [Kumo-Anomaly](https://huggingface.co/nvidia/Kumo-Anomaly)
+ (NV-Tesseract-AD, 2.4M), NVIDIA's diffusion-based anomaly detector.
+
+The weights are the upstream weights: unchanged in the `fp32` variant, and
+with the larger ones rounded to float16 in the `f16w` variant. Amazon and
+NVIDIA have not reviewed or endorsed these conversions. NVIDIA's model card
+describes Kumo-Anomaly as "for research and development only".
+
+## Files
+
+| file | inputs | output |
+| --- | --- | --- |
+| `chronos-2..onnx`, `chronos-2-small..onnx` | `features` (batch, tokens, 48), `token_mask` (batch, tokens), `group_mask` (batch, batch), `reg_mask` (tokens, 1), `rope_cos`, `rope_sin` (tokens, 64) | `quantiles` (batch, tokens, Q × 16) |
+| `kumo-anomaly..onnx` | `x` (batch, 40, 100, 2), `step_onehot` (1, 500), `strategy_onehot` (1, 2), `time` (100) | `eps` (batch, 40, 100) |
+
+Each graph comes in two variants (``), listed here with the sizes of
+Chronos-2, Chronos-2 small and Kumo-Anomaly:
+
+- `fp32`: the checkpoints' weights as they are (478 MB, 112 MB and 10 MB).
+- `f16w`: every float32 weight of 1,024 or more elements stored as float16
+ and cast back to float32 on load (239 MB, 56 MB and 5 MB). Compute stays
+ float32.
+
+The Chronos-2 graphs take the encoder's input already assembled: instance
+scaling, arcsinh, patching, time encodings, the rotary tables and the masks
+built from group ids are computed outside the graph, so one graph serves
+every context length up to 8,192 and every horizon up to 1,024 without int64
+shape arithmetic. Q is 21 quantiles for Chronos-2 and 13 for Chronos-2 small.
+
+The Kumo-Anomaly graph is the denoiser only. The masks, the noise, the
+DPM-Solver++ sampling loop, the median over samples and the scores run
+outside it, one graph call per solver step.
+
+`manifest.json` lists every file with its size and sha256, each model's
+configuration (context length, quantile levels, patch size, diffusion
+schedule) and the parity figures below.
+
+## How they were made
+
+Sources: `amazon/chronos-2` at `29ec3766d36d6f73f0696f85560a422f50e8498c`,
+`autogluon/chronos-2-small` at `ddec01313e50b6bc58ebaa92ede81bc24a3d9f9a`
+(with `chronos-forecasting` 2.3.2), and `nvidia/Kumo-Anomaly` at
+`226c5003b0582adc1cac433123f47324bfae39f2` (with NVIDIA's
+[structured-data-models](https://github.com/NVIDIA/structured-data-models)
+at commit `2b1696f4fdc22004cc757c44703a69cf326c0a11`).
+
+```sh
+cd models/tsfm
+uv run export_chronos2.py # Chronos-2 and Chronos-2 small
+uv run export_kumo.py # Kumo-Anomaly
+```
+
+Both scripts (in `models/tsfm/` of Tabular Data Workbench) export with
+`torch.onnx` (dynamo, opset 18, PyTorch 2.14 on CPU), remove the exporter's
+node metadata and doc strings, and check the graphs against the reference
+packages before writing the manifest.
+
+## Parity
+
+Chronos-2: the whole forecast (numpy preparation, the ONNX graph in
+onnxruntime 1.30 on CPU, decoding) against `Chronos2Pipeline.predict_quantiles`,
+maximum absolute difference over all quantiles, divided by the forecast's
+widest spread between the lowest and highest quantile:
+
+| case | Chronos-2 fp32 | Chronos-2 f16w | small fp32 | small f16w |
+| --- | --- | --- | --- | --- |
+| AirPassengers, 24 months | 1.3e-6 | 2.3e-4 | 1.2e-6 | 3.8e-4 |
+| AirPassengers with gaps, 36 months | 1.0e-6 | 2.6e-4 | 8.0e-7 | 1.7e-4 |
+| daily series, weekly cycle, 28 days | 9.3e-7 | 4.6e-4 | 7.9e-7 | 3.3e-4 |
+| the same with a known covariate | 9.3e-7 | 2.7e-4 | 9.5e-7 | 4.9e-4 |
+
+Kumo-Anomaly: the sampling loop with the fp32 graph against
+`sdm.models.KumoAnomaly` on a series with injected anomalies, with the same
+noise (4 samples, 6 solver steps): anomaly scores agree to 1.0e-6 (scores up
+to 1.06).
+
+## Licence
+
+All three models are licensed under the Apache License, Version 2.0
+(`LICENSE`). `NOTICE` carries the copyright notices (Amazon.com, Inc. or its
+affiliates for Chronos-2; NVIDIA CORPORATION & AFFILIATES for Kumo-Anomaly)
+and states how these files differ from the originals.
+`LICENSE.structured-data-models` is NVIDIA's copy of the licence from the
+code the Kumo-Anomaly graph was exported from. Kumo-Anomaly's sampler derives
+from DPM-Solver (Copyright (c) 2022 Cheng Lu, MIT License), whose licence is
+in `THIRD-PARTY-NOTICES.txt`.
+
+Keep `LICENSE`, `NOTICE`, `THIRD-PARTY-NOTICES.txt` and
+`LICENSE.structured-data-models` with the graphs when redistributing them.
+
+## Citation
+
+```bibtex
+@article{ansari2025chronos2,
+ title = {Chronos-2: From Univariate to Universal Forecasting},
+ author = {Abdul Fatir Ansari and Oleksandr Shchur and Jaris Küken and Andreas Auer and Boran Han and Pedro Mercado and Syama Sundar Rangapuram and Huibin Shen and Lorenzo Stella and Xiyuan Zhang and Mononito Goswami and Shubham Kapoor and Danielle C. Maddix and Pablo Guerron and Tony Hu and Junming Yin and Nick Erickson and Prateek Mutalik Desai and Hao Wang and Huzefa Rangwala and George Karypis and Yuyang Wang and Michael Bohlke-Schneider},
+ year = {2025},
+ url = {https://arxiv.org/abs/2510.15821}
+}
+```
diff --git a/tsfm/THIRD-PARTY-NOTICES.txt b/tsfm/THIRD-PARTY-NOTICES.txt
new file mode 100644
index 0000000000000000000000000000000000000000..167a9864e399a62c5bc8e1f7112c0784fc892bc4
--- /dev/null
+++ b/tsfm/THIRD-PARTY-NOTICES.txt
@@ -0,0 +1,29 @@
+THIRD-PARTY NOTICES
+
+Component: DPM-Solver
+Source: https://github.com/LuChengTHU/dpm-solver
+Used by: Kumo-Anomaly (sdm.models.kumo.timeseries.anomaly). Its sampler derives from
+DPM-Solver; the sampler runs outside the ONNX graph, and its step tables (solver
+orders, alphas, sigmas) are what the client code reproduces.
+
+MIT License
+
+Copyright (c) 2022 Cheng Lu
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
diff --git a/tsfm/chronos-2-small.f16w.onnx b/tsfm/chronos-2-small.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..654556926c734b0fe5a9d2cdde085a86e996c885
--- /dev/null
+++ b/tsfm/chronos-2-small.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:56e2743689b8c841d39ec5935d0fd18aa92dd5e8fc83710e08009791e0eff5b2
+size 55960044
diff --git a/tsfm/chronos-2-small.fp32.onnx b/tsfm/chronos-2-small.fp32.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..5e2290d3dc65faa86981a8fe726da05b9f62b277
--- /dev/null
+++ b/tsfm/chronos-2-small.fp32.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d11c8ebf5caeafa27205ba8a5919135e4a75cc25648f81d109e06d39513e4534
+size 111801066
diff --git a/tsfm/chronos-2.f16w.onnx b/tsfm/chronos-2.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..a6a12e09186089b9cf0974eddec9383987fbf09a
--- /dev/null
+++ b/tsfm/chronos-2.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44fa813bd35869f5765234cfae8d0e4a337b7c25c50c15ecb66944fc46e1f5a3
+size 239161416
diff --git a/tsfm/chronos-2.fp32.onnx b/tsfm/chronos-2.fp32.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..d865ac0e1b18e611cb38c4f9ac442a459ea8b1c7
--- /dev/null
+++ b/tsfm/chronos-2.fp32.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e606c3d80cc3c1323466b864bb7236842fbe86d639df75598ec69389a0a8221
+size 478045331
diff --git a/tsfm/kumo-anomaly.f16w.onnx b/tsfm/kumo-anomaly.f16w.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..cdfc5eca0639fcbf5f2199b0dc1c65821ff47564
--- /dev/null
+++ b/tsfm/kumo-anomaly.f16w.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:351fa5558baed4b91b10b8f516660b02da3727db61c6fca3e366b901c665851f
+size 5327808
diff --git a/tsfm/kumo-anomaly.fp32.onnx b/tsfm/kumo-anomaly.fp32.onnx
new file mode 100644
index 0000000000000000000000000000000000000000..f9bdce34d181e785d057efa259fbeb361f09c973
--- /dev/null
+++ b/tsfm/kumo-anomaly.fp32.onnx
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3627470deafd7d497b6b37704a9adedf422bed7b070076bdbf54fe21ab9d5210
+size 10398173
diff --git a/tsfm/manifest.json b/tsfm/manifest.json
new file mode 100644
index 0000000000000000000000000000000000000000..51ec1904e2fc879531bfbb4cae430583c857c2c1
--- /dev/null
+++ b/tsfm/manifest.json
@@ -0,0 +1,220 @@
+{
+ "opset": 18,
+ "files": [
+ {
+ "model": "chronos-2-small",
+ "variant": "fp32",
+ "file": "chronos-2-small.fp32.onnx",
+ "bytes": 111801066,
+ "sha256": "d11c8ebf5caeafa27205ba8a5919135e4a75cc25648f81d109e06d39513e4534",
+ "cpuNodes": []
+ },
+ {
+ "model": "chronos-2-small",
+ "variant": "f16w",
+ "file": "chronos-2-small.f16w.onnx",
+ "bytes": 55960044,
+ "sha256": "56e2743689b8c841d39ec5935d0fd18aa92dd5e8fc83710e08009791e0eff5b2",
+ "cpuNodes": []
+ },
+ {
+ "model": "chronos-2",
+ "variant": "fp32",
+ "file": "chronos-2.fp32.onnx",
+ "bytes": 478045331,
+ "sha256": "8e606c3d80cc3c1323466b864bb7236842fbe86d639df75598ec69389a0a8221",
+ "cpuNodes": []
+ },
+ {
+ "model": "chronos-2",
+ "variant": "f16w",
+ "file": "chronos-2.f16w.onnx",
+ "bytes": 239161416,
+ "sha256": "44fa813bd35869f5765234cfae8d0e4a337b7c25c50c15ecb66944fc46e1f5a3",
+ "cpuNodes": []
+ },
+ {
+ "model": "kumo-anomaly",
+ "variant": "fp32",
+ "file": "kumo-anomaly.fp32.onnx",
+ "bytes": 10398173,
+ "sha256": "3627470deafd7d497b6b37704a9adedf422bed7b070076bdbf54fe21ab9d5210",
+ "cpuNodes": []
+ },
+ {
+ "model": "kumo-anomaly",
+ "variant": "f16w",
+ "file": "kumo-anomaly.f16w.onnx",
+ "bytes": 5327808,
+ "sha256": "351fa5558baed4b91b10b8f516660b02da3727db61c6fca3e366b901c665851f",
+ "cpuNodes": []
+ }
+ ],
+ "models": {
+ "chronos-2-small": {
+ "family": "chronos2",
+ "source": "autogluon/chronos-2-small@ddec01313e50b6bc58ebaa92ede81bc24a3d9f9a",
+ "parameters": 27934624,
+ "contextLength": 8192,
+ "maxOutputPatches": 64,
+ "patch": 16,
+ "timeScale": 8192.0,
+ "dKv": 64,
+ "quantiles": [
+ 0.01,
+ 0.05,
+ 0.1,
+ 0.2,
+ 0.3,
+ 0.4,
+ 0.5,
+ 0.6,
+ 0.7,
+ 0.8,
+ 0.9,
+ 0.95,
+ 0.99
+ ],
+ "exporter": "torch.onnx (dynamo) 2.14.0+cpu, chronos-forecasting 2.3.2",
+ "checks": {
+ "air_passengers": {
+ "fp32": {
+ "maxAbsDiff": 0.0004534340109785262,
+ "maxAbsDiffOverSpread": 1.2389919830395411e-06
+ },
+ "f16w": {
+ "maxAbsDiff": 0.14048303526010386,
+ "maxAbsDiffOverSpread": 0.0003838647966982188
+ }
+ },
+ "daily_weekly": {
+ "fp32": {
+ "maxAbsDiff": 1.936539992186681e-05,
+ "maxAbsDiffOverSpread": 7.937634073890962e-07
+ },
+ "f16w": {
+ "maxAbsDiff": 0.00816448589266372,
+ "maxAbsDiffOverSpread": 0.00033465201689035035
+ }
+ },
+ "daily_weekly_covariate": {
+ "fp32": {
+ "maxAbsDiff": 1.4207511100039483e-05,
+ "maxAbsDiffOverSpread": 9.504695011556084e-07
+ },
+ "f16w": {
+ "maxAbsDiff": 0.00725193513707012,
+ "maxAbsDiffOverSpread": 0.0004851478294551331
+ }
+ },
+ "air_passengers_gaps": {
+ "fp32": {
+ "maxAbsDiff": 0.0005596565754331095,
+ "maxAbsDiffOverSpread": 8.044586999305215e-07
+ },
+ "f16w": {
+ "maxAbsDiff": 0.12156327028947089,
+ "maxAbsDiffOverSpread": 0.00017473685590255084
+ }
+ }
+ }
+ },
+ "chronos-2": {
+ "family": "chronos2",
+ "source": "amazon/chronos-2@29ec3766d36d6f73f0696f85560a422f50e8498c",
+ "parameters": 119477664,
+ "contextLength": 8192,
+ "maxOutputPatches": 64,
+ "patch": 16,
+ "timeScale": 8192.0,
+ "dKv": 64,
+ "quantiles": [
+ 0.01,
+ 0.05,
+ 0.1,
+ 0.15,
+ 0.2,
+ 0.25,
+ 0.3,
+ 0.35,
+ 0.4,
+ 0.45,
+ 0.5,
+ 0.55,
+ 0.6,
+ 0.65,
+ 0.7,
+ 0.75,
+ 0.8,
+ 0.85,
+ 0.9,
+ 0.95,
+ 0.99
+ ],
+ "exporter": "torch.onnx (dynamo) 2.14.0+cpu, chronos-forecasting 2.3.2",
+ "checks": {
+ "air_passengers": {
+ "fp32": {
+ "maxAbsDiff": 0.00041321987168885244,
+ "maxAbsDiffOverSpread": 1.3253376289255459e-06
+ },
+ "f16w": {
+ "maxAbsDiff": 0.07163125687907268,
+ "maxAbsDiffOverSpread": 0.00022974596976921714
+ }
+ },
+ "daily_weekly": {
+ "fp32": {
+ "maxAbsDiff": 2.2354764610099664e-05,
+ "maxAbsDiffOverSpread": 9.333428173677023e-07
+ },
+ "f16w": {
+ "maxAbsDiff": 0.011134839485819725,
+ "maxAbsDiffOverSpread": 0.00046489518623411645
+ }
+ },
+ "daily_weekly_covariate": {
+ "fp32": {
+ "maxAbsDiff": 1.1173277229659107e-05,
+ "maxAbsDiffOverSpread": 9.343844901163358e-07
+ },
+ "f16w": {
+ "maxAbsDiff": 0.0031910641166916776,
+ "maxAbsDiffOverSpread": 0.00026685821503548776
+ }
+ },
+ "air_passengers_gaps": {
+ "fp32": {
+ "maxAbsDiff": 0.000852749681143905,
+ "maxAbsDiffOverSpread": 1.0165590811508801e-06
+ },
+ "f16w": {
+ "maxAbsDiff": 0.21954769663511797,
+ "maxAbsDiffOverSpread": 0.00026172182727855466
+ }
+ }
+ }
+ },
+ "kumo-anomaly": {
+ "family": "kumo-anomaly",
+ "source": "nvidia/Kumo-Anomaly@226c5003b0582adc1cac433123f47324bfae39f2",
+ "parameters": 2439169,
+ "channels": 40,
+ "window": 100,
+ "trainSteps": 500,
+ "betas": [
+ 0.0001,
+ 0.01
+ ],
+ "exporter": "torch.onnx (dynamo) 2.14.0+cpu",
+ "checks": {
+ "samples4-steps6": {
+ "maxAbsDiff": 1.0004111891115386e-06,
+ "maxScore": 1.0615177154541016,
+ "graphCalls": 12,
+ "cpuSeconds": 16.310610055923462
+ }
+ }
+ }
+ }
+}