Merge branch 'master' of https://github.com/ITMO-NSS-team/BAMT

.gitignore

@@ -12,3 +12,5 @@ tutorials/bamt
 .DS_Store
 /example_socio.ipynb
 /temp.ipynb
+/bamt_house_price_example_feature_analysis.ipynb
+/bamt_house_price_example_data_sampling.ipynb

.readthedocs.yml

@@ -1,17 +1,39 @@
-# .readthedocs.yml
-# Read the Docs configuration file
+# Read the Docs configuration file for Sphinx projects
 # See https://docs.readthedocs.io/en/stable/config-file/v2.html for details
 
 # Required
 version: 2
 
-# Build documentation in the docs/ directory with Sphinx
+# Set the OS, Python version and other tools you might need
+build:
+  os: ubuntu-22.04
+  tools:
+    python: "3.11"
+    # You can also specify other tool versions:
+    # nodejs: "20"
+    # rust: "1.70"
+    # golang: "1.20"
+
+# Build documentation in the "docs/" directory with Sphinx
 sphinx:
   configuration: docs/source/conf.py
-  fail_on_warning: true
+  # You can configure Sphinx to use a different builder, for instance use the dirhtml builder for simpler URLs
+  # builder: "dirhtml"
+  # Fail on all warnings to avoid broken references
+  # fail_on_warning: true
+
+# Optionally build your docs in additional formats such as PDF and ePub
+# formats:
+#    - pdf
+#    - epub
 
+# Optional but recommended, declare the Python requirements required
+# to build your documentation
+# See https://docs.readthedocs.io/en/stable/guides/reproducible-builds.html
+# python:
+#    install:
+#    - requirements: docs/requirements.txt
 
 python:
-  version: "3.8"
   install:
     - requirements: other_requirements/readthedocs.txt

README.rst

@@ -84,22 +84,22 @@ Then the necessary classes are imported from the library:
 
 .. code-block:: python
 
-   import bamt.networks as Nets
+   from bamt.networks.hybrid_bn import HybridBN
 
 Next, a network instance is created and training (structure and parameters) is performed:
 
 .. code-block:: python
 
-   bn = Nets.HybridBN(has_logit=False, use_mixture=True)
-   bn.add_edges(discretized_data, scoring_function=('K2',K2Score))
+   bn = HybridBN(has_logit=False, use_mixture=True)
+   bn.add_edges(preprocessed_data)
    bn.fit_parameters(data)
 
 
 
 Examples & Tutorials
 ^^^^^^^^^^^^^^^^^^^^^^
 
-More examples can be found in `tutorials <https://github.com/ITMO-NSS-team/BAMT/tree/master/tutorials>`__  and `Documentation <https://bamt.readthedocs.io/en/latest/examples/learn_save.html>`__.
+More examples can be found in `Documentation <https://bamt.readthedocs.io/en/latest/examples/learn_save.html>`__.
 
 Publications about BAMT
 ^^^^^^^^^^^^^^^^^^^^^^^

bamt/utils/composite_utils/MLUtils.py

@@ -1,5 +1,7 @@
 import json
 from random import choice
+
+import pkg_resources
 from typing import Union
 
 from catboost import CatBoostClassifier, CatBoostRegressor
@@ -33,6 +35,11 @@
     LGBMRegressor = None
     LGBMClassifier = None
 
+lgbm_params = "lgbm_params.json"
+models_repo = "models_repo.json"
+lgbm_params_path = pkg_resources.resource_filename(__name__, lgbm_params)
+models_repo_path = pkg_resources.resource_filename(__name__, models_repo)
+
 
 class MlModels:
     def __init__(self):
@@ -94,7 +101,7 @@ def __init__(self):
             self.operations_by_types["lgbm"] = "LGBMClassifier"
 
         if LGBMClassifier and LGBMRegressor is not None:
-            with open("bamt/utils/composite_utils/lgbm_params.json") as file:
+            with open(lgbm_params_path) as file:
                 self.lgbm_dict = json.load(file)
 
     def get_model_by_children_type(self, node: Union[GraphNode, CompositeNode]):
@@ -104,7 +111,7 @@ def get_model_by_children_type(self, node: Union[GraphNode, CompositeNode]):
         else:
             type_model = "class"
         forbidden_tags = ["non-default", "expensive"]
-        with open("bamt/utils/composite_utils/models_repo.json", "r") as f:
+        with open(models_repo_path, "r") as f:
             models_json = json.load(f)
             models = models_json["operations"]
             if LGBMClassifier and LGBMRegressor is not None:

docs/source/examples/add_ml_models_to_nodes.rst

@@ -7,7 +7,7 @@ First, lets import BAMT modules and required machine learning modules.
 
 .. code-block:: python
 
-    from bamt.networks.hybrid_bn import HybridBN
+    import bamt.networks as networks
     import bamt.Preprocessors as pp
 
     import pandas as pd
@@ -66,7 +66,7 @@ Next, we initialize Bayesian Network object and add nodes to it.
 
 .. code-block:: python
 
-    bn = HybridBN(has_logit=True, use_mixture=True)
+    bn = networks.HybridBN(has_logit=True, use_mixture=True)
     bn.add_nodes(info)
 
 After adding nodes we can perform structure learning.

docs/source/examples/learn_params_vis.rst

@@ -5,7 +5,7 @@ Used imports:
 
 .. code-block:: python
 
-    from bamt.networks.hybrid_bn import HybridBN
+    import bamt.networks as networks
     import bamt.Preprocessors as pp
 
     import pandas as pd
@@ -44,7 +44,7 @@ Preprocessing data, encode categorical features and discretize numerical feature
     p = pp.Preprocessor([('encoder', encoder), ('discretizer', discretizer)])
     discretized_data, est = p.apply(data)
 
-    bn = HybridBN(has_logit=True, use_mixture=True) # init BN
+    bn = networks.HybridBN(has_logit=True, use_mixture=True) # init BN
     info = p.info
     info
     bn.add_nodes(info)

docs/source/examples/learn_sampling_predict.rst

@@ -5,7 +5,7 @@ Used imports:
 
 .. code-block:: python
 
-    from bamt.networks.hybrid_bn import HybridBN
+    import bamt.networks as networks
     import bamt.Preprocessors as pp
 
     import pandas as pd

docs/source/examples/learn_save.rst

@@ -8,7 +8,7 @@ Used imports:
     from bamt.preprocessors import Preprocessor
     import pandas as pd
     from sklearn import preprocessing as pp
-    from bamt.networks.hybrid_bn import HybridBN
+    from bamt.networks import HybridBN
 
 
 Let's start with data loading and preprocessing: 

docs/source/examples/read_structure_and_params.rst

@@ -8,7 +8,7 @@ Used imports:
     from bamt.preprocessors import Preprocessor
     import pandas as pd
     from sklearn import preprocessing as pp
-    from bamt.networks.hybrid_bn import HybridBN
+    from bamt.networks import HybridBN
     import json
 
 You can read the pre-trained structure and distribution parameters from a file.

docs/source/examples/read_structure_param_learning.rst

@@ -8,7 +8,7 @@ Used imports:
     from bamt.preprocessors import Preprocessor
     import pandas as pd
     from sklearn import preprocessing as pp
-    from bamt.networks.hybrid_bn import HybridBN
+    from bamt.networks import HybridBN
     import json
 
 There are two options for loading a BN structure. The first is to read it directly from a JSON file:

docs/source/index.rst

@@ -57,7 +57,6 @@ generating synthetic data, assessing edges strength e.t.c.
    models/bayesiannetworks/hybrid_bn.rst
    models/bayesiannetworks/sampling_predicting.rst
    models/bayesiannetworks/large_bn_algs.rst
-   models/bayesiannetworks/models_storing.rst
 
 .. toctree::
    :maxdepth: 1
@@ -66,7 +65,6 @@ generating synthetic data, assessing edges strength e.t.c.
 
    bnalgs/bn_learning.rst
    bnalgs/big_bns.rst
-   bnalgs/applied_tasks.rst
 
 .. toctree::
     :maxdepth: 1
@@ -88,6 +86,13 @@ generating synthetic data, assessing edges strength e.t.c.
     examples/add_ml_models_to_nodes.rst
     examples/composite_network_example.rst
 
+.. toctree::
+    :maxdepth: 1
+    :hidden:
+    :caption: Tutorials
+
+    tutorials/tutorials_gists.rst
+
 
 
 Indices and tables

docs/source/models/bayesiannetworks/base_network.rst

@@ -27,7 +27,7 @@ For Example:
 
 .. code-block:: python
 
-    from bamt.networks.discrete_bn import DiscreteBN
+    import bamt.networks as networks
     import bamt.preprocessors as pp
     import pandas as pd
 
@@ -38,7 +38,7 @@ For Example:
     p = pp.Preprocessor([('encoder', encoder), ('discretizer', discretizer)])
     discretized_data, est = p.apply(asia)
 
-    bn = DiscreteBN()
+    bn = networks.DiscreteBN()
     info = p.info
     info
 

docs/source/models/bayesiannetworks/composite_bn.rst

@@ -13,9 +13,9 @@ To initialize a ``CompositeBN`` object, you can use the following code:
 
 .. code-block:: python
 
-    from  bamt.networks.composite_bn import CompositeBN
+    import bamt.networks as networks
 
-    bn = CompositeBN()
+    bn = networks.CompositeBN()
 
 
 Data Preprocessing

docs/source/models/bayesiannetworks/continuous_bn.rst

@@ -13,9 +13,9 @@ To initialize a ``ContinuousBN`` object, you can use the following code:
 
 .. code-block:: python
 
-    from bamt.networks.continuous_bn import ContinuousBN
+    import bamt.networks as networks
     
-    bn = ContinuousBN(use_mixture=True)
+    bn = networks.ContinuousBN(use_mixture=True)
 
 ContinuousBN has an additional parameter ``use_mixture``. 
 It is used to determine whether to use mixuters of Gaussian distributions  to represent the conditional distribution of continuous variables.

docs/source/models/bayesiannetworks/discrete_bn.rst

@@ -13,9 +13,9 @@ To initialize a ``DiscreteBN`` object, you can use the following code:
 
 .. code-block:: python
 
-    from bamt.networks.discrete_bn import DiscreteBN
+    import bamt.networks as networks
 
-    bn = DiscreteBN()
+    bn = networks.DiscreteBN()
 
 Data Preprocessing
 ~~~~~~~~~~~~~~~~~~

docs/source/models/bayesiannetworks/hybrid_bn.rst

@@ -13,9 +13,9 @@ To initialize a ``HybridBN`` object, you can use the following code:
 
 .. code-block:: python
 
-    from  bamt.networks.hybrid_bn import HybridBN
+    import bamt.networks as networks
     
-    bn = HybridBN(has_logit=True, use_mixture=True)
+    bn = networks.HybridBN(has_logit=True, use_mixture=True)
 
 HybridBN has two additional parameters ``has_logit`` and ``use_mixture``.
 ``has_logit`` is used to determine whether to use logit nodes. Logit nodes use machine learning algorithms to represent variable.

docs/source/models/bayesiannetworks/large_bn_algs.rst

@@ -50,7 +50,7 @@ Then perform structure learning as usual, but use ``ps`` as ``white_list``:
 
 .. code-block:: python
 
-    bn = Nets.ContinuousBN()
+    bn = networks.ContinuousBN()
 
     bn.add_nodes(descriptor=info)
 

docs/source/tutorials/tutorials_gists.rst

@@ -0,0 +1,8 @@
+Full-scale tutorial examples
+============================
+
+In this section full-scale GitHub Gist tutorials for BAMT use cases are listed,
+feel free to leave your comments and questions at the following gists:
+
+* `BAMT house price dataset prediction results improvement with sampling. <https://gist.github.com/jrzkaminski/ca4c9ef9292ef0e8df4da5b81effae1e>`__
+* `BAMT house price dataset linear regression with BAMT. <https://gist.github.com/jrzkaminski/50805963ca02ca8a326e344ea39e592c>`__

tests/sendingRegressors.py

@@ -4,6 +4,7 @@
 from catboost import CatBoostRegressor
 from sklearn import preprocessing as pp
 from sklearn.ensemble import RandomForestRegressor
+
 # from sklearn.linear_model import ElasticNet
 from sklearn.tree import DecisionTreeRegressor