flash-ansr 0.4.2__tar.gz

flash_ansr-0.4.2/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2024 Paul Saegert
+
+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.

flash_ansr-0.4.2/PKG-INFO

@@ -0,0 +1,152 @@
+Metadata-Version: 2.4
+Name: flash_ansr
+Version: 0.4.2
+Summary: Flash Amortized Neural Symbolic Regression
+Author: Paul Saegert
+Project-URL: Github, https://github.com/psaegert/flash-ansr
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: absl-py
+Requires-Dist: datasets
+Requires-Dist: drawdata
+Requires-Dist: editdistance
+Requires-Dist: einops
+Requires-Dist: matplotlib
+Requires-Dist: numpy
+Requires-Dist: pyyaml
+Requires-Dist: schedulefree
+Requires-Dist: scikit-learn
+Requires-Dist: scipy
+Requires-Dist: simplipy
+Requires-Dist: torch
+Requires-Dist: torch_optimizer
+Requires-Dist: tqdm
+Requires-Dist: wandb
+Requires-Dist: zss
+Provides-Extra: dev
+Requires-Dist: pre-commit; extra == "dev"
+Requires-Dist: pytest; extra == "dev"
+Requires-Dist: pytest-cov; extra == "dev"
+Requires-Dist: mypy; extra == "dev"
+Requires-Dist: flake8; extra == "dev"
+Requires-Dist: pygount; extra == "dev"
+Requires-Dist: pylint; extra == "dev"
+Requires-Dist: types-setuptools; extra == "dev"
+Requires-Dist: types-tqdm; extra == "dev"
+Requires-Dist: types-toml; extra == "dev"
+Requires-Dist: types-PyYAML; extra == "dev"
+Requires-Dist: radon; extra == "dev"
+Requires-Dist: mkdocs; extra == "dev"
+Requires-Dist: mkdocs-material; extra == "dev"
+Requires-Dist: mkdocs-autorefs; extra == "dev"
+Requires-Dist: mkdocstrings; extra == "dev"
+Requires-Dist: mkdocs-get-deps; extra == "dev"
+Requires-Dist: mkdocs-material-extensions; extra == "dev"
+Requires-Dist: mkdocstrings-python; extra == "dev"
+Dynamic: license-file
+
+<h1 align="center" style="margin-top: 0px;">⚡ANSR:<br>Flash Amortized Neural Symbolic Regression</h1>
+
+<div align="center">
+
+[![PyPI version](https://img.shields.io/pypi/v/flash-ansr.svg)](https://pypi.org/project/flash-ansr/)
+[![PyPI license](https://img.shields.io/pypi/l/flash-ansr.svg)](https://pypi.org/project/flash-ansr/)
+[![Documentation Status](https://readthedocs.org/projects/flash-ansr/badge/?version=latest)](https://flash-ansr.readthedocs.io/en/latest/?badge=latest)
+
+</div>
+
+<div align="center">
+
+[![pytest](https://github.com/psaegert/flash-ansr/actions/workflows/pytest.yml/badge.svg)](https://github.com/psaegert/flash-ansr/actions/workflows/pytest.yml)
+[![quality checks](https://github.com/psaegert/flash-ansr/actions/workflows/pre-commit.yml/badge.svg)](https://github.com/psaegert/flash-ansr/actions/workflows/pre-commit.yml)
+[![CodeQL Advanced](https://github.com/psaegert/flash-ansr/actions/workflows/codeql.yaml/badge.svg)](https://github.com/psaegert/flash-ansr/actions/workflows/codeql.yaml)
+
+</div>
+
+# Papers
+- WIP
+
+
+# Usage
+
+```sh
+pip install flash-ansr
+```
+
+
+```python
+import torch
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+# Import flash_ansr
+from flash_ansr import (
+  FlashANSR,
+  SoftmaxSamplingConfig,
+  install_model,
+  get_path,
+)
+
+# Specify the model
+# Here: https://huggingface.co/psaegert/flash-ansr-v23.0-120M
+MODEL = "psaegert/flash-ansr-v23.0-120M"
+
+# Download the latest snapshot of the model
+# By default, the model is downloaded to the directory `./models/` in the package root
+install_model(MODEL)
+
+# Load the model
+model = FlashANSR.load(
+  directory=get_path('models', MODEL),
+  generation_config=SoftmaxSamplingConfig(choices=32),  # or BeamSearchConfig / MCTSGenerationConfig
+  n_restarts=8,
+).to(device)
+
+# Define data
+X = ...
+y = ...
+
+# Fit the model to the data
+model.fit(X, y, verbose=True)
+
+# Show the best expression
+print(model.get_expression())
+
+# Predict with the best expression
+y_pred = model.predict(X)
+```
+
+# Overview
+
+### Training
+
+<img src="./assets/images/flash-ansr-training.png" width="300">
+
+> **⚡ANSR Training on Fully Procedurally Generated Data** Inspired by NeSymReS ([Biggio et al. 2021](https://arxiv.org/abs/2106.06427))
+
+### Architecture
+
+<img src="./assets/images/flash-ansr.png">
+
+> **FlashANSR Architecture.** The model consists of an upgraded version of the SetTransformer ([Lee et al. 2019](https://arxiv.org/abs/1810.00825)) as an encoder, and a Pre-Norm Transformer decoder ([Vaswani et al. 2017](https://arxiv.org/abs/1706.03762)) as a generative model over symbolic expressions.
+
+### Results
+Coming soon
+<!-- <img src="./assets/images/test_time_compute_fastsrb.svg">
+
+> **Test Time Compute scaling.** ⚡ANSR, NeSymReS ([Biggio et al. 2021](https://arxiv.org/abs/2106.06427)), PySR ([Cranmer 2023](https://arxiv.org/abs/2305.01582)), and E2E ([Kamienny et al. 2022](https://arxiv.org/abs/2204.10532)) are evaluated on the FastSRB benchmark with 10 datasets per equation, $n_{support}=512$, noise level 0.0.\
+> AMD 9950X (16C32T), RTX 4090 (24GB). -->
+
+
+
+# Citation
+```bibtex
+@software{flash-ansr2024,
+    author = {Paul Saegert},
+    title = {Flash Amortized Neural Symbolic Regression},
+    year = 2024,
+    publisher = {GitHub},
+    version = {0.4.2},
+    url = {https://github.com/psaegert/flash-ansr}
+}
+```

flash_ansr-0.4.2/README.md

@@ -0,0 +1,104 @@
+<h1 align="center" style="margin-top: 0px;">⚡ANSR:<br>Flash Amortized Neural Symbolic Regression</h1>
+
+<div align="center">
+
+[![PyPI version](https://img.shields.io/pypi/v/flash-ansr.svg)](https://pypi.org/project/flash-ansr/)
+[![PyPI license](https://img.shields.io/pypi/l/flash-ansr.svg)](https://pypi.org/project/flash-ansr/)
+[![Documentation Status](https://readthedocs.org/projects/flash-ansr/badge/?version=latest)](https://flash-ansr.readthedocs.io/en/latest/?badge=latest)
+
+</div>
+
+<div align="center">
+
+[![pytest](https://github.com/psaegert/flash-ansr/actions/workflows/pytest.yml/badge.svg)](https://github.com/psaegert/flash-ansr/actions/workflows/pytest.yml)
+[![quality checks](https://github.com/psaegert/flash-ansr/actions/workflows/pre-commit.yml/badge.svg)](https://github.com/psaegert/flash-ansr/actions/workflows/pre-commit.yml)
+[![CodeQL Advanced](https://github.com/psaegert/flash-ansr/actions/workflows/codeql.yaml/badge.svg)](https://github.com/psaegert/flash-ansr/actions/workflows/codeql.yaml)
+
+</div>
+
+# Papers
+- WIP
+
+
+# Usage
+
+```sh
+pip install flash-ansr
+```
+
+
+```python
+import torch
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+# Import flash_ansr
+from flash_ansr import (
+  FlashANSR,
+  SoftmaxSamplingConfig,
+  install_model,
+  get_path,
+)
+
+# Specify the model
+# Here: https://huggingface.co/psaegert/flash-ansr-v23.0-120M
+MODEL = "psaegert/flash-ansr-v23.0-120M"
+
+# Download the latest snapshot of the model
+# By default, the model is downloaded to the directory `./models/` in the package root
+install_model(MODEL)
+
+# Load the model
+model = FlashANSR.load(
+  directory=get_path('models', MODEL),
+  generation_config=SoftmaxSamplingConfig(choices=32),  # or BeamSearchConfig / MCTSGenerationConfig
+  n_restarts=8,
+).to(device)
+
+# Define data
+X = ...
+y = ...
+
+# Fit the model to the data
+model.fit(X, y, verbose=True)
+
+# Show the best expression
+print(model.get_expression())
+
+# Predict with the best expression
+y_pred = model.predict(X)
+```
+
+# Overview
+
+### Training
+
+<img src="./assets/images/flash-ansr-training.png" width="300">
+
+> **⚡ANSR Training on Fully Procedurally Generated Data** Inspired by NeSymReS ([Biggio et al. 2021](https://arxiv.org/abs/2106.06427))
+
+### Architecture
+
+<img src="./assets/images/flash-ansr.png">
+
+> **FlashANSR Architecture.** The model consists of an upgraded version of the SetTransformer ([Lee et al. 2019](https://arxiv.org/abs/1810.00825)) as an encoder, and a Pre-Norm Transformer decoder ([Vaswani et al. 2017](https://arxiv.org/abs/1706.03762)) as a generative model over symbolic expressions.
+
+### Results
+Coming soon
+<!-- <img src="./assets/images/test_time_compute_fastsrb.svg">
+
+> **Test Time Compute scaling.** ⚡ANSR, NeSymReS ([Biggio et al. 2021](https://arxiv.org/abs/2106.06427)), PySR ([Cranmer 2023](https://arxiv.org/abs/2305.01582)), and E2E ([Kamienny et al. 2022](https://arxiv.org/abs/2204.10532)) are evaluated on the FastSRB benchmark with 10 datasets per equation, $n_{support}=512$, noise level 0.0.\
+> AMD 9950X (16C32T), RTX 4090 (24GB). -->
+
+
+
+# Citation
+```bibtex
+@software{flash-ansr2024,
+    author = {Paul Saegert},
+    title = {Flash Amortized Neural Symbolic Regression},
+    year = 2024,
+    publisher = {GitHub},
+    version = {0.4.2},
+    url = {https://github.com/psaegert/flash-ansr}
+}
+```

flash_ansr-0.4.2/pyproject.toml

@@ -0,0 +1,88 @@
+[project]
+name = "flash_ansr"
+description = "Flash Amortized Neural Symbolic Regression"
+authors = [
+    {name = "Paul Saegert"},
+    ]
+readme = "README.md"
+requires-python = ">=3.10"
+dynamic = ["dependencies"]
+version = "0.4.2"
+urls = { Github = "https://github.com/psaegert/flash-ansr"}
+
+[project.scripts]
+flash_ansr = "flash_ansr.__main__:main"
+
+
+[tool.setuptools.dynamic]
+dependencies = {file = ["requirements.txt"]}
+
+
+[project.optional-dependencies]
+dev = [
+    "pre-commit",
+    "pytest",
+    "pytest-cov",
+    "mypy",
+    "flake8",
+    "pygount",
+    "pylint",
+    "types-setuptools",
+    "types-tqdm",
+    "types-toml",
+    "types-PyYAML",
+    "radon",
+    "mkdocs",
+    "mkdocs-material",
+    "mkdocs-autorefs",
+    "mkdocstrings",
+    "mkdocs-get-deps",
+    "mkdocs-material-extensions",
+    "mkdocstrings-python"
+]
+
+[tool.setuptools]
+include-package-data = true
+
+
+[tool.setuptools.packages.find]
+where = ["src"]
+
+
+[build-system]
+requires = [
+    "setuptools>=68",
+    "wheel",
+    "platformdirs==3.10.0",
+]
+build-backend = "setuptools.build_meta"
+
+
+[tool.flake8]
+ignore = ["E501", "W503"]
+exclude = [
+    ".git",
+	"__pycache__",
+	"documentation",
+	"build",
+	"venv",
+	".venv",
+	"env",
+	".env",
+	"images",
+]
+per-file-ignores = "__init__.py:F401"
+
+
+[tool.mypy]
+no_implicit_optional = false
+disallow_untyped_defs = true
+disallow_incomplete_defs = true
+explicit_package_bases = true
+exclude = "(.venv|tests/|experimental/|src/flash_ansr/compat/nesymres.py)"
+ignore_missing_imports = true
+
+[tool.pytest.ini_options]
+filterwarnings = [
+    "ignore:This process .*use of fork.*deadlocks in the child.:DeprecationWarning",
+]

flash_ansr-0.4.2/requirements.txt

@@ -0,0 +1,17 @@
+absl-py
+datasets
+drawdata
+editdistance
+einops
+matplotlib
+numpy
+pyyaml
+schedulefree
+scikit-learn
+scipy
+simplipy
+torch
+torch_optimizer
+tqdm
+wandb
+zss

flash_ansr-0.4.2/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

flash_ansr-0.4.2/src/flash_ansr/__init__.py

@@ -0,0 +1,29 @@
+from .model import (
+    ModelFactory,
+    FlashANSRModel,
+    SetTransformer,
+    Tokenizer,
+    RotaryEmbedding,
+    IEEE75432PreEncoder,
+    install_model,
+    remove_model,
+)
+from .expressions import SkeletonPool, NoValidSampleFoundError
+from .utils import (
+    GenerationConfig,
+    GenerationConfigBase,
+    BeamSearchConfig,
+    SoftmaxSamplingConfig,
+    MCTSGenerationConfig,
+    create_generation_config,
+    get_path,
+    load_config,
+    save_config,
+    substitute_root_path,
+)
+from .eval import Evaluation
+from .refine import Refiner, ConvergenceError
+from .flash_ansr import FlashANSR
+from .baselines import SkeletonPoolModel, BruteForceModel
+from .data.data import FlashANSRDataset
+from .preprocessing import FlashANSRPreprocessor