feature/svm (#79)

* fix naive bayes index, add svm support

* add `svm` chain

* add `svm chain` to pymilo router in `pymilo_func.py`

* add svm models' support to pymilo param

* add svm chain to `test_pymilo.py`

* add LinearSVC test

* add LinearSVR test

* add NuSVC test

* add NuSVR test

* add OneClassSVM test

* add SVC test

* add SVR test

* add svm models' test runner

* add exported svm model's folder to gitignore

* run tests + remove unused imports + remove trailing whitespaces

* remove `dual` parameter

* support for `numpy.intc` serialization added

* transporting `numpy.intc` data structure added

* `CHANGELOG.md` updated

* add relative import

* deep serialize LaberBinarizer ndarray

* deep deserialize_ndarray function implemented

* deep serialize_ndarray function implemented

* refactor pure primitive data type or numpy primitive data type deserialization into a function

* add deep ndarray serializer to serialize_tuple

* add deep ndarray serializer to serialize_dict

* add deep ndarray serializer to serialize function

* refactor `deserialize` function to have deep ndarray deserialization

* refactor primitive type deserialization

* refactor primitive type deserialization

* add deep ndarray deserializer to `get_deserialized_dict` function

* add & update docstring

* strict the condition

* add pymilo bypass

* handle complex dtype str

* add pymilo bypass

* change not(a in b) to a not in b

* apply `autopep8.sh`

* refactor `check_str_in_iterable` function

* remove unused import

* refactor `isinstance(X,dict) and Y in X` with `check check_str_in_iterable(Y,X)`

* update check_str_in_iterable function

* apply `autopep8.sh`

* update function name

* `CHANGELOG.md` updated

* user `enumerate(x)` instead of `range(len(x))`

* use `isinstance` instead of `type`

* `CHANGELOG.md` updated

* remove unused import

* update IDs

* raise error & warning instead of print

* refactor `get_homogeneous_type` function

* Using a predefined string imported from `pymilo_praram.py` instead.

* remove trailing whitespaces

Author: AHReccese

.gitignore

@@ -104,4 +104,5 @@ gen
 /tests/exported_decision_trees
 /tests/exported_clusterings
 /tests/exported_naive_bayes
+/tests/exported_svms
 /.VSCodeCounter

CHANGELOG.md

@@ -6,7 +6,31 @@ and this project adheres to [Semantic Versioning](http://semver.org/spec/v2.0.0.
 
 ## [Unreleased]
 ### Added
+- `deserialize_primitive_type` function in `GeneralDataStructureTransporter`
+- `is_deserialized_ndarray` function in `GeneralDataStructureTransporter`
+- `deep_deserialize_ndarray` function in `GeneralDataStructureTransporter`
+- `deep_serialize_ndarray`  function in `GeneralDataStructureTransporter`
+- `SVR` model
+- `SVC` model
+- `One Class SVM` model
+- `NuSVR` model
+- `NuSVC` model
+- `Linear SVR` model
+- `Linear SVC` model
+- SVM models test runner
+- SVM chain
 ### Changed
+- `TreeTransporter` updated
+- `get_homogeneous_type` function in `util.py` updated
+- `GeneralDataStructureTransporter` updated to use deep ndarray serializer & deserializer
+- `check_str_in_iterable` updated
+- `Label Binarizer` Transporter updated
+- `Function` Transporter updated
+- `CFNode` Transporter updated
+- `Bisecting Tree` Transporter updated
+- Tests config modified
+- SVM params initialized in `pymilo_param`
+- SVM support added to `pymilo_func.py`
 - `SUPPORTED_MODELS.md` updated
 - `README.md` updated
 ## [0.5] - 2024-01-31
@@ -19,7 +43,7 @@ and this project adheres to [Semantic Versioning](http://semver.org/spec/v2.0.0.
 - `Bernoulli Naive Bayes` model declared as `BernoulliNB` model
 - `Categorical Naive Bayes` model declared as `CategoricalNB` model
 - Naive Bayes models test runner
-- Naive Bayes chain 
+- Naive Bayes chain
 ### Changed
 - `Transport` function of `AbstractTransporter` updated
 - fix the order of `CFNode` fields serialization in `CFNodeTransporter`

SUPPORTED_MODELS.md

@@ -10,6 +10,7 @@
 	* [Decision Trees](#scikit-learn-trees)
 	* [Clustering Models](#scikit-learn-clustering)
 	* [Naive Bayes](#scikit-learn-naivebayes)
+	* [Support Vector Machine](#scikit-learn-svm)
 
 <h2 id="scikit-learn">Scikit-Learn</h2> 
 <h3 id="scikit-learn-linear">Linear Models</h3>
@@ -396,3 +397,51 @@
         <td>>=0.5</td>
 	</tr>
 </table>
+
+<h3 id="scikit-learn-svm">Support Vector Machine</h3> 
+
+📚 <a href="https://scikit-learn.org/stable/modules/svm.html" target="_blank"><b>Models Document</b></a>
+
+
+<table>
+	<tr align="center">
+		<th>ID</th>
+		<th>Model Name</th>
+        <th>PyMilo Version</th>
+	</tr>
+	<tr align="center">
+		<td>1</td>
+		<td><b>Linear SVC</b></td>
+        <td>>=0.6</td>
+	</tr>
+	<tr align="center">
+		<td>2</td>
+		<td><b>Linear SVR</b></td>
+        <td>>=0.6</td>
+	</tr>
+	<tr align="center">
+		<td>3</td>
+		<td><b>NuSVC</b></td>
+        <td>>=0.6</td>
+	</tr>
+	<tr align="center">
+		<td>4</td>
+		<td><b>NuSVR</b></td>
+        <td>>=0.6</td>
+	</tr>
+	<tr align="center">
+		<td>5</td>
+		<td><b>One Class SVM</b></td>
+        <td>>=0.6</td>
+	</tr>
+	<tr align="center">
+		<td>6</td>
+		<td><b>SVC</b></td>
+        <td>>=0.6</td>
+	</tr>
+	<tr align="center">
+		<td>7</td>
+		<td><b>SVR</b></td>
+        <td>>=0.6</td>
+	</tr>
+</table>

pymilo/chains/svm_chain.py

@@ -0,0 +1,134 @@
+# -*- coding: utf-8 -*-
+"""PyMilo chain for svm models."""
+from ..transporters.transporter import Command
+
+from ..transporters.general_data_structure_transporter import GeneralDataStructureTransporter
+from ..transporters.randomstate_transporter import RandomStateTransporter
+
+from ..pymilo_param import SKLEARN_SVM_TABLE
+from ..exceptions.serialize_exception import PymiloSerializationException, SerilaizatoinErrorTypes
+from ..exceptions.deserialize_exception import PymiloDeserializationException, DeSerilaizatoinErrorTypes
+from traceback import format_exc
+
+SVM_CHAIN = {
+    "GeneralDataStructureTransporter": GeneralDataStructureTransporter(),
+    "RandomStateTransporter": RandomStateTransporter(),
+}
+
+
+def is_svm(model):
+    """
+    Check if the input model is a sklearn's svm model.
+
+    :param model: is a string name of a svm or a sklearn object of it
+    :type model: any object
+    :return: check result as bool
+    """
+    if isinstance(model, str):
+        return model in SKLEARN_SVM_TABLE
+    else:
+        return type(model) in SKLEARN_SVM_TABLE.values()
+
+
+def transport_svm(request, command):
+    """
+    Return the transported (Serialized or Deserialized) model.
+
+    :param request: given svm to be transported
+    :type request: any object
+    :param command: command to specify whether the request should be serialized or deserialized
+    :type command: transporter.Command
+    :return: the transported request as a json string or sklearn svm model
+    """
+    _validate_input(request, command)
+
+    if command == Command.SERIALIZE:
+        try:
+            return serialize_svm(request)
+        except Exception as e:
+            raise PymiloSerializationException(
+                {
+                    'error_type': SerilaizatoinErrorTypes.VALID_MODEL_INVALID_INTERNAL_STRUCTURE,
+                    'error': {
+                        'Exception': repr(e),
+                        'Traceback': format_exc(),
+                    },
+                    'object': request,
+                })
+
+    elif command == Command.DESERIALZIE:
+        try:
+            return deserialize_svm(request)
+        except Exception as e:
+            raise PymiloDeserializationException(
+                {
+                    'error_type': SerilaizatoinErrorTypes.VALID_MODEL_INVALID_INTERNAL_STRUCTURE,
+                    'error': {
+                        'Exception': repr(e),
+                        'Traceback': format_exc()},
+                    'object': request})
+
+
+def serialize_svm(svm_object):
+    """
+    Return the serialized json string of the given svm model.
+
+    :param svm_object: given model to be get serialized
+    :type svm_object: any sklearn svm model
+    :return: the serialized json string of the given svm
+    """
+    for transporter in SVM_CHAIN:
+        SVM_CHAIN[transporter].transport(
+            svm_object, Command.SERIALIZE)
+    return svm_object.__dict__
+
+
+def deserialize_svm(svm):
+    """
+    Return the associated sklearn svm model of the given svm.
+
+    :param svm: given json string of a svm model to get deserialized to associated sklearn svm model
+    :type svm: obj
+    :return: associated sklearn svm model
+    """
+    raw_model = SKLEARN_SVM_TABLE[svm.type]()
+    data = svm.data
+
+    for transporter in SVM_CHAIN:
+        SVM_CHAIN[transporter].transport(
+            svm, Command.DESERIALZIE)
+    for item in data:
+        setattr(raw_model, item, data[item])
+    return raw_model
+
+
+def _validate_input(model, command):
+    """
+    Check if the provided inputs are valid in relation to each other.
+
+    :param model: a sklearn svm model or a json string of it, serialized through the pymilo export.
+    :type model: obj
+    :param command: command to specify whether the request should be serialized or deserialized
+    :type command: transporter.Command
+    :return: None
+    """
+    if command == Command.SERIALIZE:
+        if is_svm(model):
+            return
+        else:
+            raise PymiloSerializationException(
+                {
+                    'error_type': SerilaizatoinErrorTypes.INVALID_MODEL,
+                    'object': model
+                }
+            )
+    elif command == Command.DESERIALZIE:
+        if is_svm(model.type):
+            return
+        else:
+            raise PymiloDeserializationException(
+                {
+                    'error_type': DeSerilaizatoinErrorTypes.INVALID_MODEL,
+                    'object': model
+                }
+            )

pymilo/pymilo_func.py

@@ -8,6 +8,7 @@
 from .chains.decision_tree_chain import transport_decision_tree, is_decision_tree
 from .chains.clustering_chain import transport_clusterer, is_clusterer
 from .chains.naive_bayes_chain import transport_naive_bayes, is_naive_bayes
+from .chains.svm_chain import transport_svm, is_svm
 
 from .transporters.transporter import Command
 
@@ -39,6 +40,8 @@ def get_sklearn_data(model):
         return transport_clusterer(model, Command.SERIALIZE)
     elif is_naive_bayes(model):
         return transport_naive_bayes(model, Command.SERIALIZE)
+    elif is_svm(model):
+        return transport_svm(model, Command.SERIALIZE)
     else:
         return None
 
@@ -61,6 +64,8 @@ def to_sklearn_model(import_obj):
         return transport_clusterer(import_obj, Command.DESERIALZIE)
     elif is_naive_bayes(import_obj.type):
         return transport_naive_bayes(import_obj, Command.DESERIALZIE)
+    elif is_svm(import_obj.type):
+        return transport_svm(import_obj, Command.DESERIALZIE)
     else:
         return None
 

pymilo/pymilo_obj.py

@@ -2,12 +2,13 @@
 """PyMilo modules."""
 from .pymilo_func import get_sklearn_data, get_sklearn_version, to_sklearn_model
 from .utils.util import get_sklearn_type
-from .pymilo_param import PYMILO_VERSION
+from .pymilo_param import PYMILO_VERSION, PYMILO_VERSION_DOES_NOT_EXIST, UNEQUAL_PYMILO_VERSIONS
 import json
 
-from pymilo.exceptions.deserialize_exception import PymiloDeserializationException, DeSerilaizatoinErrorTypes
-from pymilo.exceptions.serialize_exception import PymiloSerializationException, SerilaizatoinErrorTypes
+from .exceptions.deserialize_exception import PymiloDeserializationException, DeSerilaizatoinErrorTypes
+from .exceptions.serialize_exception import PymiloSerializationException, SerilaizatoinErrorTypes
 from traceback import format_exc
+from warnings import warn
 
 
 class Export:
@@ -99,6 +100,10 @@ def __init__(self, file_adr, json_dump=None):
             else:
                 with open(file_adr, 'r') as fp:
                     serialized_model_obj = json.load(fp)
+            if "pymilo_version" not in serialized_model_obj:
+                raise Exception(PYMILO_VERSION_DOES_NOT_EXIST)
+            if not serialized_model_obj["pymilo_version"] == PYMILO_VERSION:
+                warn(UNEQUAL_PYMILO_VERSIONS, category=Warning)
             self.data = serialized_model_obj["data"]
             self.version = serialized_model_obj["sklearn_version"]
             self.type = serialized_model_obj["model_type"]

pymilo/pymilo_param.py

@@ -1,7 +1,20 @@
 # -*- coding: utf-8 -*-
 """Parameters and constants."""
+from sklearn.svm import SVR
+from sklearn.svm import SVC
+from sklearn.svm import OneClassSVM
+from sklearn.svm import NuSVR
+from sklearn.svm import NuSVC
+from sklearn.svm import LinearSVR
+from sklearn.svm import LinearSVC
+from sklearn.naive_bayes import CategoricalNB
+from sklearn.naive_bayes import BernoulliNB
+from sklearn.naive_bayes import ComplementNB
+from sklearn.naive_bayes import MultinomialNB
+from sklearn.naive_bayes import GaussianNB
 from sklearn.preprocessing import LabelBinarizer
 from numpy import uint8
+from numpy import intc
 from numpy import inf
 from numpy import float64
 from numpy import int32
@@ -77,7 +90,8 @@
 
 PYMILO_VERSION = "0.5"
 NOT_SUPPORTED = "NOT_SUPPORTED"
-
+PYMILO_VERSION_DOES_NOT_EXIST = "Corrupted JSON file, `pymilo_version` doesn't exist in this file."
+UNEQUAL_PYMILO_VERSIONS = "warning: Installed Pymilo version differes from pymilo version used to create the JSON file."
 
 glm_support = {
     'GammaRegressor': False,
@@ -110,12 +124,6 @@
 except BaseException:
     pass
 
-from sklearn.naive_bayes import GaussianNB
-from sklearn.naive_bayes import MultinomialNB
-from sklearn.naive_bayes import ComplementNB
-from sklearn.naive_bayes import BernoulliNB
-from sklearn.naive_bayes import CategoricalNB
-
 
 SKLEARN_LINEAR_MODEL_TABLE = {
     "LinearRegression": LinearRegression,
@@ -195,6 +203,15 @@
     "CategoricalNB": CategoricalNB,
 }
 
+SKLEARN_SVM_TABLE = {
+    "LinearSVC": LinearSVC,
+    "LinearSVR": LinearSVR,
+    "NuSVC": NuSVC,
+    "NuSVR": NuSVR,
+    "OneClassSVM": OneClassSVM,
+    "SVC": SVC,
+    "SVR": SVR,
+}
 
 KEYS_NEED_PREPROCESSING_BEFORE_DESERIALIZATION = {
     "_label_binarizer": LabelBinarizer,  # in Ridge Classifier
@@ -207,6 +224,7 @@
 }
 
 NUMPY_TYPE_DICT = {
+    "numpy.intc": intc,
     "numpy.int32": int32,
     "numpy.int64": int64,
     "numpy.float64": float64,
@@ -220,4 +238,5 @@
     "DECISION_TREE": "exported_decision_trees",
     "CLUSTERING": "exported_clusterings",
     "NAIVE_BAYES": "exported_naive_bayes",
+    "SVM": "exported_svms",
 }

pymilo/transporters/bisecting_tree_transporter.py

@@ -4,7 +4,7 @@
 from ..pymilo_param import SKLEARN_CLUSTERING_TABLE, NOT_SUPPORTED
 from .transporter import AbstractTransporter
 from .general_data_structure_transporter import GeneralDataStructureTransporter
-from ..utils.util import is_iterable
+from ..utils.util import check_str_in_iterable
 
 bisecting_tree_support = SKLEARN_CLUSTERING_TABLE["BisectingKMeans"] != NOT_SUPPORTED
 if bisecting_tree_support:
@@ -114,7 +114,6 @@ def is_pymilo_serialized_bisecting_tree(psbt):
     :return: boolean
     """
     return (
-        is_iterable(psbt) and
-        "pymilo_model_type" in psbt and
+        check_str_in_iterable("pymilo_model_type", psbt) and
         psbt["pymilo_model_type"] == "_BisectingTree"
     )