Strip torch (#246)

* have users install torch separately
Co-authored-by: Karla Saur <[email protected]>

.github/workflows/pythonapp.yml

@@ -27,6 +27,18 @@ jobs:
       uses: actions/setup-python@v2
       with:
         python-version: ${{ matrix.python-version }}
+    - name: Install pytorch for mac
+      if: ${{ matrix.os == 'macos-latest' }}
+      run: |
+        pip install torch
+    - name: Install pytorch 1.5.1+cpu if python 3.5 (not mac)
+      if: ${{ matrix.python-version == '3.5' && matrix.os != 'macos-latest' }}
+      run: |
+        pip install torch==1.5.1+cpu -f https://download.pytorch.org/whl/torch_stable.html
+    - name: Install pytorch 1.6.0+cpu if python > 3.5 (not mac)
+      if:  ${{ matrix.python-version != '3.5' && matrix.os != 'macos-latest' }}
+      run: |
+        pip install torch==1.6.0+cpu -f https://download.pytorch.org/whl/torch_stable.html
     - name: Install basic dependencies
       run: |
         python -m pip install --upgrade pip
@@ -67,4 +79,4 @@ jobs:
       uses: codecov/codecov-action@v1
       with:
         file: ./coverage.xml
-        flags: unittests
+        flags: unittests

CONTRIBUTING.md

@@ -30,22 +30,23 @@ For more information see the [Code of Conduct FAQ](https://opensource.microsoft.
 contact [[email protected]](mailto:[email protected]) with any additional questions or comments.
 
 ## Developing
-The simplest setup is:
+The simplest setup (assuming you have PyTorch already installed) is:
 ```
 mkdir hummingbird
 cd hummingbird
 git clone https://github.com/microsoft/hummingbird.git .
 pip install -e .[docs,tests,extra]
 ```
-On Windows, the last line above must also contain `-f https://download.pytorch.org/whl/torch_stable.html`. (This is required because PyTorch version for Windows is not up to date.)
+
+If you do not have pytorch installed, please [install it](https://pytorch.org/get-started/locally/) first.
 
 ### Docker
 We provide a simple [Dockerfile](https://github.com/microsoft/hummingbird/blob/master/Dockerfile) that you can customize to your preferred development environment.
 ```
 docker build git://github.com/microsoft/hummingbird -t hb-jupy
 docker run -it hb-dev
 ```
-### Codespases
+### Codespaces
 For a light-weight, web-based experience, we provide the configuration ([.devcontainer](https://github.com/microsoft/hummingbird/tree/master/.devcontainer)) for [Codespaces](https://online.visualstudio.com/environments).  More information on this setup can be found [here]( https://docs.microsoft.com/en-us/visualstudio/online/reference/configuring).
 
 ### Tools

README.md

@@ -9,24 +9,22 @@
 ## Introduction
 *Hummingbird* is a library for compiling trained traditional ML models into tensor computations. *Hummingbird* allows users to seamlessly leverage neural network frameworks (such as [PyTorch](https://pytorch.org/)) to accelerate traditional ML models. Thanks to *Hummingbird*, users can benefit from: (1) all the current and future optimizations implemented in neural network frameworks; (2) native hardware acceleration; (3) having a unique platform to support for both traditional and neural network models; and have all of this (4) without having to re-engineer their models.
 
-Currently, you can use *Hummingbird* to convert your trained traditional ML models into [PyTorch](https://pytorch.org/), [TorchScript](https://pytorch.org/docs/stable/jit.html), and [ONNX](https://onnx.ai/). *Hummingbird* [supports](https://github.com/microsoft/hummingbird/wiki/Supported-Operators) a variety of tree-based classifiers and regressors.  These models include
+Currently, you can use *Hummingbird* to convert your trained traditional ML models into [PyTorch](https://pytorch.org/), [TorchScript](https://pytorch.org/docs/stable/jit.html), and [ONNX](https://onnx.ai/). *Hummingbird* [supports](https://github.com/microsoft/hummingbird/wiki/Supported-Operators) a variety of ML models and featurizers.  These models include
 [scikit-learn](https://scikit-learn.org/stable/) Decision Trees and Random Forest, and also [LightGBM](https://github.com/Microsoft/LightGBM) and [XGBoost](https://github.com/dmlc/xgboost) Classifiers/Regressors. Support for other neural network backends (e.g., [TVM](https://docs.tvm.ai/)) and models is on our [roadmap](https://github.com/microsoft/hummingbird/wiki/Roadmap-for-Upcoming-Features-and-Support).
 
+Hummingbird also provides a convenient uniform "inference" API following the Sklearn API. This allows swapping Sklearn models with Hummingbird-generated ones without having to change the inference code.
+
 ## Installation
 
 Hummingbird was tested on Python >= 3.5 on Linux, Windows and MacOS machines.  It is recommended to use a virtual environment (See: [python3 venv doc](https://docs.python.org/3/tutorial/venv.html) or [Using Python environments in VS Code](https://code.visualstudio.com/docs/python/environments).)
 
+Hummingbird requires PyTorch >= 1.4.0. Please go [here](https://pytorch.org/) for instructions on how to install PyTorch based on your platform and hardware.
 
-Install the Hummingbird package in Linux or Mac
+Once PyTorch is installed, you can get Hummingbird from pip with:
 ```
 pip install hummingbird-ml
 ```
 
-To install the package in Windows instead:
-```
-pip install hummingbird-ml -f https://download.pytorch.org/whl/torch_stable.html
-```
-
 If you require the optional dependencies lightgbm and xgboost, you can use:
 ```
 pip install hummingbird-ml[extra]

doc/html/hummingbird/ml/convert.html

@@ -75,7 +75,7 @@ <h1 class="title">Module <code>hummingbird.ml.convert</code></h1>
     &#34;&#34;&#34;
     Function used to check whether the specified backend and configuration pair is supported or not.
     &#34;&#34;&#34;
-    assert torch_installed()
+    assert torch_installed(), &#34;To use Hummingbird you need to install torch.&#34;
     import onnx
     import torch
 

doc/html/hummingbird/ml/operator_converters/onnx/onnx_operator.html

@@ -327,18 +327,10 @@ <h3>Ancestors</h3>
 <h3>Methods</h3>
 <dl>
 <dt id="hummingbird.ml.operator_converters.onnx.onnx_operator.Cast.forward"><code class="name flex">
-<span>def <span class="ident">forward</span></span>(<span>self, x)</span>
+<span>def <span class="ident">forward</span></span>(<span>self, x) -> Callable[..., Any]</span>
 </code></dt>
 <dd>
-<div class="desc"><p>Defines the computation performed at every call.</p>
-<p>Should be overridden by all subclasses.</p>
-<div class="admonition note">
-<p class="admonition-title">Note</p>
-<p>Although the recipe for forward pass needs to be defined within
-this function, one should call the :class:<code>Module</code> instance afterwards
-instead of this since the former takes care of running the
-registered hooks while the latter silently ignores them.</p>
-</div></div>
+<div class="desc"></div>
 <details class="source">
 <summary>
 <span>Expand source code</span>
@@ -377,18 +369,10 @@ <h3>Ancestors</h3>
 <h3>Methods</h3>
 <dl>
 <dt id="hummingbird.ml.operator_converters.onnx.onnx_operator.Concat.forward"><code class="name flex">
-<span>def <span class="ident">forward</span></span>(<span>self, *x)</span>
+<span>def <span class="ident">forward</span></span>(<span>self, *x) -> Callable[..., Any]</span>
 </code></dt>
 <dd>
-<div class="desc"><p>Defines the computation performed at every call.</p>
-<p>Should be overridden by all subclasses.</p>
-<div class="admonition note">
-<p class="admonition-title">Note</p>
-<p>Although the recipe for forward pass needs to be defined within
-this function, one should call the :class:<code>Module</code> instance afterwards
-instead of this since the former takes care of running the
-registered hooks while the latter silently ignores them.</p>
-</div></div>
+<div class="desc"></div>
 <details class="source">
 <summary>
 <span>Expand source code</span>
@@ -429,18 +413,10 @@ <h3>Ancestors</h3>
 <h3>Methods</h3>
 <dl>
 <dt id="hummingbird.ml.operator_converters.onnx.onnx_operator.Reshape.forward"><code class="name flex">
-<span>def <span class="ident">forward</span></span>(<span>self, x)</span>
+<span>def <span class="ident">forward</span></span>(<span>self, x) -> Callable[..., Any]</span>
 </code></dt>
 <dd>
-<div class="desc"><p>Defines the computation performed at every call.</p>
-<p>Should be overridden by all subclasses.</p>
-<div class="admonition note">
-<p class="admonition-title">Note</p>
-<p>Although the recipe for forward pass needs to be defined within
-this function, one should call the :class:<code>Module</code> instance afterwards
-instead of this since the former takes care of running the
-registered hooks while the latter silently ignores them.</p>
-</div></div>
+<div class="desc"></div>
 <details class="source">
 <summary>
 <span>Expand source code</span>

doc/html/hummingbird/ml/operator_converters/sklearn/skl_sv.html

@@ -304,18 +304,10 @@ <h3>Ancestors</h3>
 <h3>Methods</h3>
 <dl>
 <dt id="hummingbird.ml.operator_converters.sklearn.skl_sv.SVC.forward"><code class="name flex">
-<span>def <span class="ident">forward</span></span>(<span>self, x)</span>
+<span>def <span class="ident">forward</span></span>(<span>self, x) -> Callable[..., Any]</span>
 </code></dt>
 <dd>
-<div class="desc"><p>Defines the computation performed at every call.</p>
-<p>Should be overridden by all subclasses.</p>
-<div class="admonition note">
-<p class="admonition-title">Note</p>
-<p>Although the recipe for forward pass needs to be defined within
-this function, one should call the :class:<code>Module</code> instance afterwards
-instead of this since the former takes care of running the
-registered hooks while the latter silently ignores them.</p>
-</div></div>
+<div class="desc"></div>
 <details class="source">
 <summary>
 <span>Expand source code</span>

hummingbird/ml/convert.py

@@ -44,7 +44,7 @@ def _supported_backend_check_config(model, backend, extra_config):
     """
     Function used to check whether the specified backend and configuration pair is supported or not.
     """
-    assert torch_installed()
+    assert torch_installed(), "To use Hummingbird you need to install torch."
     import onnx
     import torch
 

setup.py

@@ -21,14 +21,7 @@
     if start_pos >= 0:
         long_description = long_description[start_pos:]
 
-install_requires = ["numpy>=1.15", "onnxconverter-common>=1.6.0", "scikit-learn==0.22.1"]
-if sys.platform == "darwin":
-    install_requires.append("torch")
-else:
-    if sys.version_info[1] == 5:
-        install_requires.append("torch==1.5.1+cpu")
-    else:
-        install_requires.append("torch==1.6.0+cpu")
+install_requires = ["numpy>=1.15", "onnxconverter-common>=1.6.0", "scikit-learn>=0.22.1", "torch>=1.4.*"]
 setup(
     name="hummingbird-ml",
     version=version_str,

tests/test_onnxml_lightgbm_converter.py

@@ -78,6 +78,7 @@ def _test_classifier(self, X, model, rtol=1e-06, atol=1e-06, extra_config={}):
     @unittest.skipIf(
         not (onnx_ml_tools_installed() and onnx_runtime_installed()), reason="ONNXML test require ONNX, ORT and ONNXMLTOOLS"
     )
+    @unittest.skipIf(not lightgbm_installed(), reason="LightGBM test requires LightGBM installed")
     def test_lightgbm_pytorch(self):
         warnings.filterwarnings("ignore")
         X = [[0, 1], [1, 1], [2, 0]]
@@ -97,6 +98,7 @@ def test_lightgbm_pytorch(self):
     @unittest.skipIf(
         not (onnx_ml_tools_installed() and onnx_runtime_installed()), reason="ONNXML test require ONNX, ORT and ONNXMLTOOLS"
     )
+    @unittest.skipIf(not lightgbm_installed(), reason="LightGBM test requires LightGBM installed")
     def test_lightgbm_onnx_pytorch(self):
         warnings.filterwarnings("ignore")
         X = [[0, 1], [1, 1], [2, 0]]
@@ -126,6 +128,7 @@ def test_lightgbm_onnx_pytorch(self):
     @unittest.skipIf(
         not (onnx_ml_tools_installed() and onnx_runtime_installed()), reason="ONNXML test require ONNX, ORT and ONNXMLTOOLS"
     )
+    @unittest.skipIf(not lightgbm_installed(), reason="LightGBM test requires LightGBM installed")
     def test_lightgbm_pytorch_extra_config(self):
         warnings.filterwarnings("ignore")
         X = [[0, 1], [1, 1], [2, 0]]
@@ -152,6 +155,7 @@ def test_lightgbm_pytorch_extra_config(self):
     @unittest.skipIf(
         not (onnx_ml_tools_installed() and onnx_runtime_installed()), reason="ONNXML test require ONNX, ORT and ONNXMLTOOLS"
     )
+    @unittest.skipIf(not lightgbm_installed(), reason="LightGBM test requires LightGBM installed")
     def test_lgbm_onnxml_model_regressor(self):
         warnings.filterwarnings("ignore")
         n_features = 28
@@ -170,6 +174,7 @@ def test_lgbm_onnxml_model_regressor(self):
     @unittest.skipIf(
         not (onnx_ml_tools_installed() and onnx_runtime_installed()), reason="ONNXML test require ONNX, ORT and ONNXMLTOOLS"
     )
+    @unittest.skipIf(not lightgbm_installed(), reason="LightGBM test requires LightGBM installed")
     def test_lightgbm_regressor(self):
         warnings.filterwarnings("ignore")
         X = [[0, 1], [1, 1], [2, 0]]
@@ -183,6 +188,7 @@ def test_lightgbm_regressor(self):
     @unittest.skipIf(
         not (onnx_ml_tools_installed() and onnx_runtime_installed()), reason="ONNXML test require ONNX, ORT and ONNXMLTOOLS"
     )
+    @unittest.skipIf(not lightgbm_installed(), reason="LightGBM test requires LightGBM installed")
     def test_lightgbm_regressor1(self):
         warnings.filterwarnings("ignore")
         model = lgb.LGBMRegressor(n_estimators=1, min_child_samples=1)
@@ -196,6 +202,7 @@ def test_lightgbm_regressor1(self):
     @unittest.skipIf(
         not (onnx_ml_tools_installed() and onnx_runtime_installed()), reason="ONNXML test require ONNX, ORT and ONNXMLTOOLS"
     )
+    @unittest.skipIf(not lightgbm_installed(), reason="LightGBM test requires LightGBM installed")
     def test_lightgbm_regressor2(self):
         warnings.filterwarnings("ignore")
         model = lgb.LGBMRegressor(n_estimators=2, max_depth=1, min_child_samples=1)
@@ -209,6 +216,7 @@ def test_lightgbm_regressor2(self):
     @unittest.skipIf(
         not (onnx_ml_tools_installed() and onnx_runtime_installed()), reason="ONNXML test require ONNX, ORT and ONNXMLTOOLS"
     )
+    @unittest.skipIf(not lightgbm_installed(), reason="LightGBM test requires LightGBM installed")
     def test_lightgbm_booster_regressor(self):
         warnings.filterwarnings("ignore")
         X = [[0, 1], [1, 1], [2, 0]]
@@ -225,6 +233,7 @@ def test_lightgbm_booster_regressor(self):
     @unittest.skipIf(
         not (onnx_ml_tools_installed() and onnx_runtime_installed()), reason="ONNXML test require ONNX, ORT and ONNXMLTOOLS"
     )
+    @unittest.skipIf(not lightgbm_installed(), reason="LightGBM test requires LightGBM installed")
     def test_lgbm_onnxml_model_binary(self):
         warnings.filterwarnings("ignore")
         n_features = 28
@@ -243,6 +252,7 @@ def test_lgbm_onnxml_model_binary(self):
     @unittest.skipIf(
         not (onnx_ml_tools_installed() and onnx_runtime_installed()), reason="ONNXML test require ONNX, ORT and ONNXMLTOOLS"
     )
+    @unittest.skipIf(not lightgbm_installed(), reason="LightGBM test requires LightGBM installed")
     def test_lightgbm_classifier(self):
         warnings.filterwarnings("ignore")
         model = lgb.LGBMClassifier(n_estimators=3, min_child_samples=1)
@@ -256,6 +266,7 @@ def test_lightgbm_classifier(self):
     @unittest.skipIf(
         not (onnx_ml_tools_installed() and onnx_runtime_installed()), reason="ONNXML test require ONNX, ORT and ONNXMLTOOLS"
     )
+    @unittest.skipIf(not lightgbm_installed(), reason="LightGBM test requires LightGBM installed")
     def test_lightgbm_classifier_zipmap(self):
         warnings.filterwarnings("ignore")
         X = [[0, 1], [1, 1], [2, 0], [1, 2]]
@@ -269,6 +280,7 @@ def test_lightgbm_classifier_zipmap(self):
     @unittest.skipIf(
         not (onnx_ml_tools_installed() and onnx_runtime_installed()), reason="ONNXML test require ONNX, ORT and ONNXMLTOOLS"
     )
+    @unittest.skipIf(not lightgbm_installed(), reason="LightGBM test requires LightGBM installed")
     def test_lightgbm_booster_classifier(self):
         warnings.filterwarnings("ignore")
         X = [[0, 1], [1, 1], [2, 0], [1, 2]]
@@ -282,6 +294,7 @@ def test_lightgbm_booster_classifier(self):
     @unittest.skipIf(
         not (onnx_ml_tools_installed() and onnx_runtime_installed()), reason="ONNXML test require ONNX, ORT and ONNXMLTOOLS"
     )
+    @unittest.skipIf(not lightgbm_installed(), reason="LightGBM test requires LightGBM installed")
     def test_lightgbm_booster_classifier_zipmap(self):
         warnings.filterwarnings("ignore")
         X = [[0, 1], [1, 1], [2, 0], [1, 2]]
@@ -295,6 +308,7 @@ def test_lightgbm_booster_classifier_zipmap(self):
     @unittest.skipIf(
         not (onnx_ml_tools_installed() and onnx_runtime_installed()), reason="ONNXML test require ONNX, ORT and ONNXMLTOOLS"
     )
+    @unittest.skipIf(not lightgbm_installed(), reason="LightGBM test requires LightGBM installed")
     def test_lgbm_onnxml_model_multi(self):
         warnings.filterwarnings("ignore")
         n_features = 28
@@ -313,6 +327,7 @@ def test_lgbm_onnxml_model_multi(self):
     @unittest.skipIf(
         not (onnx_ml_tools_installed() and onnx_runtime_installed()), reason="ONNXML test require ONNX, ORT and ONNXMLTOOLS"
     )
+    @unittest.skipIf(not lightgbm_installed(), reason="LightGBM test requires LightGBM installed")
     def test_lightgbm_classifier_multi(self):
         warnings.filterwarnings("ignore")
         model = lgb.LGBMClassifier(n_estimators=3, min_child_samples=1)
@@ -326,6 +341,7 @@ def test_lightgbm_classifier_multi(self):
     @unittest.skipIf(
         not (onnx_ml_tools_installed() and onnx_runtime_installed()), reason="ONNXML test require ONNX, ORT and ONNXMLTOOLS"
     )
+    @unittest.skipIf(not lightgbm_installed(), reason="LightGBM test requires LightGBM installed")
     def test_lightgbm_booster_multi_classifier(self):
         warnings.filterwarnings("ignore")
         X = [[0, 1], [1, 1], [2, 0], [1, 2], [-1, 2], [1, -2]]