sharada 0.1.0.dev0__tar.gz

sharada-0.1.0.dev0/.github/workflows/ci.yml

@@ -0,0 +1,27 @@
+name: CI
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+  workflow_call:
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    strategy:
+      fail-fast: false
+      matrix:
+        python-version: ["3.10", "3.11", "3.12", "3.13"]
+    steps:
+      - uses: actions/checkout@v7
+      - uses: astral-sh/setup-uv@v7
+        with:
+          python-version: ${{ matrix.python-version }}
+      # The architecture tests run a real encoder, so its weights are worth keeping between runs.
+      - uses: actions/cache@v4
+        with:
+          path: ~/.cache/huggingface
+          key: hf-${{ hashFiles('pyproject.toml') }}
+      - run: uv sync
+      - run: uv run pytest -q

sharada-0.1.0.dev0/.github/workflows/release.yml

@@ -0,0 +1,69 @@
+# Publishing, with no tokens (PyPI "trusted publishing"):
+#   - Run this workflow by hand (Actions -> Release -> Run workflow) to publish to TestPyPI.
+#   - Publish a GitHub release with tag vX.Y.Z to publish to PyPI.
+# One-time setup on pypi.org and test.pypi.org: add a trusted publisher with
+# owner LenaBarretta, repository sharada, workflow release.yml and environment
+# `pypi` (on pypi.org) or `testpypi` (on test.pypi.org).
+name: Release
+
+on:
+  release:
+    types: [published]
+  workflow_dispatch:
+
+jobs:
+  test:
+    uses: ./.github/workflows/ci.yml
+
+  build:
+    needs: test
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v7
+      - uses: astral-sh/setup-uv@v7
+      - run: uv build
+      - name: Check that the release tag matches the package version
+        if: github.event_name == 'release'
+        run: |
+          version="${GITHUB_REF_NAME#v}"
+          test -f "dist/sharada-${version}-py3-none-any.whl" || {
+            echo "Tag $GITHUB_REF_NAME does not match the version in sharada/__init__.py"
+            ls dist
+            exit 1
+          }
+      - uses: actions/upload-artifact@v7
+        with:
+          name: dist
+          path: dist/
+
+  testpypi:
+    if: github.event_name == 'workflow_dispatch'
+    needs: build
+    runs-on: ubuntu-latest
+    environment: testpypi
+    permissions:
+      id-token: write
+    steps:
+      - uses: actions/download-artifact@v8
+        with:
+          name: dist
+          path: dist/
+      - uses: pypa/gh-action-pypi-publish@release/v1
+        with:
+          repository-url: https://test.pypi.org/legacy/
+          # A version can be uploaded only once; reruns just re-check the setup.
+          skip-existing: true
+
+  pypi:
+    if: github.event_name == 'release'
+    needs: build
+    runs-on: ubuntu-latest
+    environment: pypi
+    permissions:
+      id-token: write
+    steps:
+      - uses: actions/download-artifact@v8
+        with:
+          name: dist
+          path: dist/
+      - uses: pypa/gh-action-pypi-publish@release/v1

sharada-0.1.0.dev0/.gitignore

@@ -0,0 +1,23 @@
+__pycache__/
+*.py[cod]
+*.egg-info/
+build/
+dist/
+.venv/
+.pytest_cache/
+.ipynb_checkpoints/
+
+# weights, checkpoints, runs
+*.pt
+*.bin
+*.safetensors
+checkpoints/
+runs/
+out/
+wandb/
+
+# datasets cache
+data/
+.cache/
+
+.DS_Store

sharada-0.1.0.dev0/LICENSE

@@ -0,0 +1,201 @@
+                                 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.

sharada-0.1.0.dev0/PKG-INFO

@@ -0,0 +1,223 @@
+Metadata-Version: 2.5
+Name: sharada
+Version: 0.1.0.dev0
+Summary: Typed decisions from text in one forward pass — options in the request, calibrated probabilities, easy to fine-tune.
+Project-URL: Homepage, https://github.com/LenaBarretta/sharada
+Project-URL: Article, https://lenatriestounderstand.com/notes/llm/024-rlcr/
+Author: Lena Barretta
+License-Expression: Apache-2.0
+License-File: LICENSE
+Keywords: calibration,decision-making,text-classification,uncertainty,zero-shot-classification
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: Science/Research
+Classifier: Programming Language :: Python :: 3
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Requires-Python: >=3.10
+Requires-Dist: huggingface-hub>=0.26
+Requires-Dist: safetensors>=0.4
+Requires-Dist: torch>=2.2
+Requires-Dist: transformers==5.12.1
+Provides-Extra: train
+Requires-Dist: datasets>=4; extra == 'train'
+Description-Content-Type: text/markdown
+
+<p align="center">
+  <img src="docs/logo.png" alt="Sharada" width="200">
+</p>
+
+<h1 align="center">Sharada</h1>
+
+<p align="center">
+  <em>Typed decisions about text in one forward pass — options in the request, calibrated probabilities out.</em>
+</p>
+
+<p align="center">
+  <a href="https://github.com/LenaBarretta/sharada/blob/main/LICENSE"><img src="https://img.shields.io/badge/license-Apache%202.0-blue.svg" alt="Apache 2.0"></a>
+  <a href="https://huggingface.co/LenaBarretta/sharada-base"><img src="https://img.shields.io/badge/%F0%9F%A4%97-sharada--base-yellow" alt="sharada-base on the Hub"></a>
+  <a href="https://lenatriestounderstand.com/notes/llm/024-rlcr/"><img src="https://img.shields.io/badge/write--up-lenatriestounderstand-ff69b4" alt="the write-up"></a>
+</p>
+
+A small encoder that makes **typed decisions about text in one forward pass**: the options come with the
+request, and what comes back is a probability for each of them. Nothing is generated, nothing is parsed,
+and the same model answers a question it has never seen because the labels are part of the input rather
+than part of the weights.
+
+```python
+from sharada import DecisionModel
+
+model = DecisionModel.from_pretrained("LenaBarretta/sharada-base")
+
+d = model.decide(
+    text="My card still hasn't arrived and I ordered it two weeks ago.",
+    question="Which team should handle this?",
+    options=["billing", "card delivery", "technical support", "account closure"],
+)
+
+d.answer          # 'card delivery'
+d.confidence      # 0.86
+d.probabilities   # {'billing': 0.07, 'card delivery': 0.86, ...}
+```
+
+Three kinds of question are typed, so the model knows what the options mean:
+
+| kind | the options are | example |
+| --- | --- | --- |
+| `choice` | unordered labels | which team, which topic, which intent |
+| `scale` | ordered steps | severity 1–5, how positive, how urgent |
+| `binary` | yes or no | is this spam, does this need a human |
+
+```python
+from sharada import Request
+
+model.decide_many([
+    Request(text=review, question="How positive is this review?",
+            options=["very negative", "negative", "neutral", "positive", "very positive"],
+            kind="scale", task="review-tone"),
+    Request(text=review, question="Does this mention a refund?",
+            options=["no", "yes"], kind="binary", task="refund-flag"),
+])
+```
+
+## Install
+
+```bash
+pip install git+https://github.com/LenaBarretta/sharada
+```
+
+Python 3.10+, `torch` and `transformers`; CPU is enough to run it.
+
+## Models
+
+| model | encoder | parameters | what it is for |
+| --- | --- | --- | --- |
+| `LenaBarretta/sharada-base` | ModernBERT-base | 150M | the default; fine-tune this one |
+| `LenaBarretta/sharada-large` | ModernBERT-large | 400M | a few points better, ~2.5× the time |
+
+Accuracy and calibration per label set are in each model card, measured on label sets the model was
+**not** trained on as well as on the ones it was.
+
+## Three things the architecture guarantees
+
+The request is laid out as one sequence — text, question, then every option as a parallel branch — and a
+mask decides who may read whom. Both are in [`layout.py`](sharada/layout.py) and
+[`masking.py`](sharada/masking.py), and they buy three properties that hold by construction, not because
+training got them approximately right:
+
+1. **The order of the options cannot matter.** Every option branch starts at the same position id, and
+   the encoder's positions are rotary, so no option is earlier or later than another. Reshuffle them and
+   the probabilities follow their options exactly.
+2. **An option's score does not depend on which other options are offered.** An option reads the text,
+   the question and itself, nothing else. Drop two options from a list of four and the remaining two keep
+   their scores to the last bit — so the probabilities are a renormalisation, and a long list of options
+   does not make each one noisier.
+3. **The text is read once.** Text tokens read only text tokens, so their states do not depend on the
+   question. Ten questions about one document are ten cheap read-outs over one encoding of it.
+
+These are the tests in [`tests/test_model.py`](tests/test_model.py), checked on an untrained model.
+
+## Fine-tune it on your own labels
+
+This is what the package is built around: a few hundred labelled examples and a few minutes.
+
+```python
+from sharada import DecisionModel, Example
+
+examples = [
+    Example(text="the invoice is wrong again", question="Which team should handle this?",
+            options=["billing", "technical", "sales"], label=0, task="routing"),
+    ...
+]
+
+model = DecisionModel.from_pretrained("LenaBarretta/sharada-base")
+report = model.fit(examples)        # holds out 20%, stops when held-out log loss stops improving
+model.save("my-router")
+
+report        # Report(820 examples, accuracy 0.914, log loss 0.287, calibration error 0.031)
+```
+
+`fit` keeps a part of the examples out, trains on the rest, early-stops, then fits one temperature per
+task on the held-out part and writes a calibration passport. Useful arguments:
+
+| argument | default | |
+| --- | --- | --- |
+| `loss` | `"cross_entropy"` | `"brier"` scores the whole distribution, not just the right option |
+| `freeze_encoder` | `False` | train the read-out only: seconds, and enough for a few hundred examples |
+| `batch_size`, `lr`, `max_epochs`, `patience` | `16`, `2e-5`, `10`, `2` | |
+
+Mixing several tasks in one `fit` is the normal case — give each one its own `task` name and each gets
+its own temperature. See [`examples/finetune_your_own.py`](examples/finetune_your_own.py), which trains
+on a CSV.
+
+## The number next to the answer is supposed to be true
+
+`0.86` should mean right about 86% of the time. That is a property of a distribution, not of a model, so
+it is fitted and it expires:
+
+```python
+from sharada import check_passport, save_passport
+
+passport = model.calibrate(recent_examples)     # one temperature per task
+save_passport(passport, "passport.json")
+
+check_passport(passport, recent_examples, model)
+# ['the calibration expired on 2026-04-01; fit it again on recent answers',
+#  'routing: the options have changed since the calibration']
+```
+
+`check_passport` returns an empty list when it finds nothing wrong — run it in the deployment pipeline
+and fail the build on anything it returns. `evaluate` gives accuracy, log loss, Brier, expected
+calibration error with a bootstrap interval, a reliability curve and a risk–coverage curve.
+
+## What to do with 0.86
+
+A probability is not a decision. `Policy` turns one into the action that costs the least, including
+handing the request to a person:
+
+```python
+from sharada import Policy, escalation_budget
+
+policy = Policy(options=["approve", "reject"],
+                costs={("fraud", "approve"): 10_000, ("clean", "reject"): 100},
+                escalate=30)
+
+policy.act({"fraud": 0.1, "clean": 0.9})     # Action('answer', 'reject', expected_loss=90.0)
+policy.act({"fraud": 0.5, "clean": 0.5})     # Action('escalate', ...)
+
+# a person can look at 5% of the traffic, no more:
+policy = escalation_budget(policy, probabilities, budget=0.05)
+```
+
+## Train the base models yourself
+
+[`training/`](training) has the whole run: a mix of public label sets in
+[`sources.py`](training/sources.py) — intents, topics, sentiment, toxicity, spam, entailment, review
+scores — each one presented with several question phrasings, shuffled options and sampled option
+subsets, so the model learns to read the options rather than their positions. Some label sets are held
+out of training entirely and only measured, which is where the zero-shot numbers come from.
+
+```bash
+python training/run.py --encoder answerdotai/ModernBERT-base --out runs/base
+```
+
+It checkpoints every few hundred steps and resumes from the checkpoint if it finds one, which is what
+makes it survive a Kaggle session; [`training/kaggle.ipynb`](training/kaggle.ipynb) is the notebook
+wrapper. One free T4: about an hour for base, about three for large.
+
+## What it will not do
+
+- **It does not generate.** No free-form answers, no extraction, no reasoning out loud. Options or
+  nothing.
+- **256 tokens of text** by default (`Limits`), with 48 for the question and 12 per option. Longer
+  documents need chunking or a larger limit, and the limit costs quadratic attention.
+- **English.** The encoder is English-only; a multilingual encoder drops in, but the published
+  checkpoints are not trained for it.
+- **It is small.** Where a frontier model knows a fact that is not in the text, it wins. This answers
+  questions about the text in front of it.
+
+## Where it came from
+
+The design, the experiments behind it and what each training signal did are written up in
+[RLCR from Scratch](https://lenatriestounderstand.com/notes/llm/024-rlcr/), with a runnable lab.
+
+Apache 2.0.