to join.
16 |
--------------------------------------------------------------------------------
/docs/autoFS.rst:
--------------------------------------------------------------------------------
1 | =============
2 | autoFS Module
3 | =============
4 |
5 | Description :
6 | - This module is used for features selection:
7 | * Automate the feature selection with several selectors
8 | * Evaluate the outputs from all selector methods, and ranked a final list of the top important features
9 |
10 | .. image:: Parallel_Coordinates_Plot.png
11 | :width: 980
12 |
13 | - Class:
14 | * dynaFS_clf : Focus on classification problems
15 | - fit() - fit and transform method for classifier
16 | * dynaFS_reg : Focus on regression problems
17 | - fit() - fit and transform method for regressor
18 |
19 | - Current available selectors
20 | * clf_fs : Class focusing on classification features selection
21 | - kbest_f : SelectKBest() with f_classif core
22 | - kbest_chi2 - SelectKBest() with chi2 core
23 | - rfe_lr - RFE with LogisticRegression() estimator
24 | - rfe_svm - RFE with SVC() estimator
25 | - rfecv_svm - RFECV with SVC() estimator
26 | - rfe_tree - RFE with DecisionTreeClassifier() estimator
27 | - rfecv_tree - RFECV with DecisionTreeClassifier() estimator
28 | - rfe_rf - RFE with RandomForestClassifier() estimator
29 | - rfecv_rf - RFECV with RandomForestClassifier() estimator
30 |
31 | * reg_fs : Class focusing on regression features selection
32 | - kbest_f : SelectKBest() with f_regression core
33 | - rfe_svm : RFE with SVC() estimator
34 | - rfecv_svm : RFECV with SVC() estimator
35 | - rfe_tree : RFE with DecisionTreeRegressor() estimator
36 | - rfecv_tree : RFECV with DecisionTreeRegressor() estimator
37 | - rfe_rf : RFE with RandomForestRegressor() estimator
38 | - rfecv_rf : RFECV with RandomForestRegressor() estimator
39 |
40 | dynaFS_clf
41 | ---------------------
42 |
43 | .. autoclass:: optimalflow.autoFS.dynaFS_clf
44 | :members:
45 |
46 | dynaFS_reg
47 | ---------------------
48 |
49 | .. autoclass:: optimalflow.autoFS.dynaFS_reg
50 | :members:
51 |
52 | clf_fs
53 | ---------------------
54 |
55 | .. autoclass:: optimalflow.selectorFS.clf_fs
56 |
57 | reg_fs
58 | ---------------------
59 |
60 | .. autoclass:: optimalflow.selectorFS.reg_fs
61 |
--------------------------------------------------------------------------------
/docs/autoFlow.rst:
--------------------------------------------------------------------------------
1 | =============
2 | autoFlow Module
3 | =============
4 |
5 | Description :
6 | - This module is used for logging & model tracking:
7 | * Log and retrieve each module's operation history & intermediate results;
8 | * Using this module as the open interface for further interactive model tracking development.
9 |
10 | Here're samples of each module's log file:
11 | - autoCV module:
12 | https://raw.githubusercontent.com/tonyleidong/OptimalFlow/master/docs/autoCV_log_2020.08.07.17.28.34.log
13 | - autoFS module:
14 | https://raw.githubusercontent.com/tonyleidong/OptimalFlow/master/docs/autoFS_log_2020.07.16.12.25.48.log
15 | - autoPP module:
16 | https://raw.githubusercontent.com/tonyleidong/OptimalFlow/master/docs/autoPP_log_2020.08.07.17.28.34.log
17 | - autoPipe module:
18 | https://raw.githubusercontent.com/tonyleidong/OptimalFlow/master/docs/autoPipe_log_2020.08.07.17.28.34.log
19 |
20 |
--------------------------------------------------------------------------------
/docs/autoPP.rst:
--------------------------------------------------------------------------------
1 | =============
2 | autoPP Module
3 | =============
4 |
5 | Description :
6 | - This module is used for data preprocessing operation:
7 | * Impute with missing value
8 | * Winsorize with outlier
9 | * Scaling using popular scaler approaches
10 | * Encoding category features using popular encoder approaches
11 | * Generated all combination datasets for further modeling and evaluation
12 | * Sparsity calculation as the critera for output datasets filtering
13 | * Custom parameters initial settings, add/remove winsorization, scaling, or encoding strategies.
14 |
15 | - Class:
16 | * dynaPreprocessing : Focus on classification/regression prprocessing problems
17 | - fit() - fit & transform method for preprocessing
18 |
19 | - Current available strategies:
20 | * Scaling : Numeric features scaling, default settings
21 | (NOTE: When you select 'None', might cause overfitting with too high R-Squared Score in Regression Problem)
22 | - "None" : None approach involve in scaling step
23 | - "standard" : StandardScaler() approach
24 | - "minmax" - MinMaxScaler() approach
25 | - "maxabs" - MaxAbsScaler() approach
26 | - "robust" - RobustScaler() approach
27 |
28 | * Encoding : Category features encoding, default settings
29 | - "onehot" : OnehotEncoder() approach, with dummy trap consideration in regression problem
30 | - "label" : LabelEncoder() approach
31 | - "frequency" : Frequency calculation approach
32 | - "mean" : Mean calculation approach
33 |
34 | * winsorization : Default limits settings
35 | - (0.01,0.01) : Top 1% and bottom 1% will be excluded
36 | - (0.05,0.05) : Top 5% and bottom 5% will be excluded
37 |
38 | dynapipePreprocessing
39 | ---------------------
40 |
41 | .. autoclass:: optimalflow.autoPP.dynaPreprocessing
42 | :members:
43 |
44 | PPtools
45 | -------
46 |
47 | .. autoclass:: optimalflow.funcPP.PPtools
48 | :members:
--------------------------------------------------------------------------------
/docs/autoPipe.rst:
--------------------------------------------------------------------------------
1 | ===============
2 | autoPipe Module
3 | ===============
4 |
5 | Description :
6 | - This module is used to build *Pipeline Cluster Traversal Experiments*:
7 | * Create sequential components of *Pipeline Cluster Traversal Experiments*
8 | * Apply traversal experiments through pipeline cluster to find the best baseline model
9 | * Generate comparable and parameter-tracable dictionaies and reports to support autoVIZ and autoFlow modules
10 |
11 | - Build Steps:
12 | * autoPP - dynaPreprocessing() Class in autoPP module
13 | * Datasets Splitting - pipeline_splitting_rule() Function in utilis_funs module
14 | * autoFS - dynaFS_clf() or dynaFS_reg() Class in autoFS module
15 | * autoCV - dynaClassifier() or dynaRegressor() Class in autoCV module
16 | * Model Evaluate - evaluate_model() Class in autoCV module
17 |
18 | .. image:: PipelineClusterTraversalExperiments.PNG
19 | :width: 980
20 |
21 | autoPipe
22 | ---------------------
23 |
24 | .. automodule:: optimalflow.autoPipe
25 | :members:
26 |
--------------------------------------------------------------------------------
/docs/autoViz.rst:
--------------------------------------------------------------------------------
1 | ==============
2 | autoViz Module
3 | ==============
4 |
5 | Description :
6 | - This module is used for outputs visualization:
7 | * Visualize and retrieve each pipeline's generating steps & performance;
8 | * Using this module as the open interface for further interactive visualization development.
9 |
10 | autoViz
11 | ---------------------
12 |
13 | .. automodule:: optimalflow.autoViz
14 | :members:
15 |
16 |
--------------------------------------------------------------------------------
/docs/autoViz_Demo.PNG:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/tonyleidong/OptimalFlow/8c38b2f6681ba8754f4d3aeb0785d55e8d8310ba/docs/autoViz_Demo.PNG
--------------------------------------------------------------------------------
/docs/autoViz_Demo_Dynamic_Table.PNG:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/tonyleidong/OptimalFlow/8c38b2f6681ba8754f4d3aeb0785d55e8d8310ba/docs/autoViz_Demo_Dynamic_Table.PNG
--------------------------------------------------------------------------------
/docs/autoViz_Demo_Model_Retrieval_Diagram.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/tonyleidong/OptimalFlow/8c38b2f6681ba8754f4d3aeb0785d55e8d8310ba/docs/autoViz_Demo_Model_Retrieval_Diagram.png
--------------------------------------------------------------------------------
/docs/comparison_table.PNG:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/tonyleidong/OptimalFlow/8c38b2f6681ba8754f4d3aeb0785d55e8d8310ba/docs/comparison_table.PNG
--------------------------------------------------------------------------------
/docs/conf.py:
--------------------------------------------------------------------------------
1 | # Configuration file for the Sphinx documentation builder.
2 | #
3 | # This file only contains a selection of the most common options. For a full
4 | # list see the documentation:
5 | # https://www.sphinx-doc.org/en/master/usage/configuration.html
6 |
7 | # -- Path setup --------------------------------------------------------------
8 |
9 | # If extensions (or modules to document with autodoc) are in another directory,
10 | # add these directories to sys.path here. If the directory is relative to the
11 | # documentation root, use os.path.abspath to make it absolute, like shown here.
12 | #
13 | import os
14 | import sys
15 | # sys.path.insert(0, os.path.abspath('..'))
16 |
17 | sys.path.append(os.path.join(os.path.abspath(os.pardir)))
18 | autodoc_mock_imports = ['category_encoders','IPython','xgboost','plotly']
19 |
20 | # Get the project root dir, which is the parent dir of this
21 | cwd = os.getcwd()
22 | project_root = os.path.dirname(cwd)
23 |
24 | # Insert the project root dir as the first element in the PYTHONPATH.
25 | # This lets us ensure that the source package is imported, and that its
26 | # version is used.
27 | sys.path.insert(0, project_root)
28 |
29 |
30 |
31 |
32 |
33 | # -- Project information -----------------------------------------------------
34 |
35 | project = 'Optimal Flow'
36 | copyright = '2020, Tony Dong'
37 | author = 'Tony Dong'
38 |
39 | # The master toctree document.
40 | master_doc = 'index'
41 |
42 | # -- General configuration ---------------------------------------------------
43 |
44 | # Add any Sphinx extension module names here, as strings. They can be
45 | # extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
46 | # ones.
47 | extensions = [
48 | 'sphinx.ext.autodoc',
49 | 'sphinx.ext.coverage',
50 | 'sphinx.ext.napoleon',
51 | 'sphinx.ext.autosummary',
52 | 'sphinx.ext.doctest',
53 | 'sphinx.ext.mathjax',
54 | 'sphinx.ext.githubpages',
55 | 'sphinx.ext.viewcode',
56 | ]
57 |
58 | # Add any paths that contain templates here, relative to this directory.
59 | templates_path = ['_templates']
60 |
61 | # List of patterns, relative to source directory, that match files and
62 | # directories to ignore when looking for source files.
63 | # This pattern also affects html_static_path and html_extra_path.
64 | exclude_patterns = ['_build', 'Thumbs.db', '.DS_Store']
65 |
66 |
67 | # -- Options for HTML output -------------------------------------------------
68 |
69 | # The theme to use for HTML and HTML Help pages. See the documentation for
70 | # a list of builtin themes.
71 | #
72 | #html_theme = 'sphinx_rtd_theme'
73 | html_theme = 'nature'
74 | # html_show_sourcelink = True
75 |
76 | # Add any paths that contain custom static files (such as style sheets) here,
77 | # relative to this directory. They are copied after the builtin static files,
78 | # so a file named "default.css" will overwrite the builtin "default.css".
79 | html_static_path = ['_static']
80 |
--------------------------------------------------------------------------------
/docs/history.rst:
--------------------------------------------------------------------------------
1 | =======
2 | History
3 | =======
4 |
5 | 0.1.11 (2020-09-29)
6 | ------------------
7 | * Added SearchinSpace settings page in Web App. Users could custom set estimators/regressors' parameters for optimal tuning outputs.
8 | * Modified some layouts of existing pages in Web App.
9 |
10 | 0.1.10 (2020-09-16)
11 | ------------------
12 | * Created a Web App, based on flask framework, as OptimalFlow's GUI. Users could build Automated Machine Learning workflow all clicks, without any coding at all!
13 | * Web App included PCTE workflow bulder, LogsViewer, Visualization, Documentation sections.
14 | * Fix the filename issues in autoViz module, and remove auto_open function when generating new html format plots.
15 |
16 | 0.1.7 (2020-08-31)
17 | ------------------
18 | * Modify autoPP's default_parameters: Remove "None" in "scaler", modify "sparsity" : [0.50], modify "cols" : [100]
19 | * Modify autoViz clf_table_report()'s coloring settings
20 | * Fix bugs in autoViz reg_table_report()'s gradient coloring function
21 |
22 | 0.1.6 (2020-08-28)
23 | ------------------
24 | * Remove evaluate_model() function's round() bugs in coping with classification problem
25 | * Move out SVM based algorithm from fastClassifier & fastRegressor's default estimators settings
26 | * Move out SVM based algorithm from autoFS class's default selectors settings
27 |
28 | 0.1.5 (2020-08-26)
29 | ------------------
30 | * Fix evaluate_model() function's bugs in coping with regression problem
31 | * Add reg_table_report() function to create dynamic table report for regression problem in autoViz
32 |
33 | 0.1.4 (2020-08-24)
34 | ------------------
35 | * Fix evaluate_model() function's precision_score issue when running modelmulti-class classification problems
36 | * Add custom_selectors args for customized algorithm settings with autoFS's 2 classes(dynaFS_reg, dynaFS_clf)
37 |
38 | 0.1.3 (2020-08-20)
39 | ------------------
40 | * Add Dynamic Table for Pipeline Cluster Model Evaluation Report in autoViz module
41 | * Add custom_estimators args for customized algorithm settings with autoCV's 4 classes(dynaClassifier,dynaRegressor,fastClassifier, and fastRegressor)
42 |
43 | 0.1.2 (2020-08-14)
44 | ------------------
45 |
46 | * Add *fastClassifier*, and *fastRegressor* class which are both random parameter search based
47 | * Modify the display settings when using dynaClassifier in non in_pipeline mode
48 |
49 | 0.1.1 (2020-08-10)
50 | ------------------
51 |
52 | * Add classifiers: LinearSVC, HistGradientBoostingClassifier, SGDClassifier, RidgeClassifierCV.
53 | * Modify Readme.md file.
54 |
55 | 0.1.0 (2020-08-10)
56 | ------------------
57 |
58 | * First release on PyPI.
--------------------------------------------------------------------------------
/docs/installation.rst:
--------------------------------------------------------------------------------
1 | .. highlight:: shell
2 |
3 | ============
4 | Installation
5 | ============
6 |
7 |
8 |
9 | To install OptimalFlow's latest version, run this command in your terminal:
10 |
11 | .. code-block:: console
12 |
13 | $ pip install --upgrade optimalflow
14 |
15 | This is the preferred method to install OptimalFlow, as it will always install the most recent stable release.
16 |
17 | You can find more details about the package info at PYPI
7 |
Currently only support Pipeline Cluster Retrieval Diagram for Classification Problem...
8 |
9 |
You can connect with me on my LinkedIn or GitHub .
10 |
18 | --------
19 | One
20 | Two
21 |
22 |
23 | Child 1:
24 | --------
25 | Eleven
26 | Twelve
27 | Thirteen
28 | fourteen
29 | fifteen
30 |
31 |
32 | *
33 | * Note: A page can have multiple sets of this
34 | *
35 | */
36 |
37 |
38 | var handle_dependent_selects = function($parent) {
39 | var $child = document.getElementById($parent.getAttribute('data-child-id')),
40 | $selected = $parent.options[$parent.selectedIndex],
41 | parent_val = $selected.value;
42 |
43 | for (var i=0; i<$child.options.length; i++) {
44 | var $option = $child.options[i];
45 | if($option.value != '') {
46 | $option.setAttribute('hidden',true);
47 | }
48 | };
49 |
50 | if(parent_val) {
51 | var child_options = $selected.getAttribute('data-child-options'),
52 | child_options_array = child_options.split('|#');
53 |
54 | for (i=0; i<$child.options.length; i++) {
55 | var $option = $child.options[i];
56 | if ($option.value == "") {
57 | $option.innerText = "--------";
58 | continue;
59 | }
60 | if(child_options_array.indexOf($option.value) != -1) {
61 | $option.removeAttribute('hidden');
62 | }
63 | };
64 |
65 | } else {
66 | var show_text = $child.getAttribute('data-text-if-parent-empty');
67 | if(!show_text) {
68 | show_text = 'Select ' + $parent.name;
69 | }
70 | for (i=0; i<$child.options.length; i++) {
71 | var $option = $child.options[$child.selectedIndex];
72 | if ($option.value == "") {
73 | $option.innerText = '- ' + show_text + ' -';
74 | break;
75 | }
76 | };
77 | }
78 | }
79 |
80 | document.addEventListener('DOMContentLoaded', function() {
81 | var $parents = document.getElementsByClassName('dependent-selects__parent');
82 | for (var i=0; i<$parents.length; i++) {
83 | handle_dependent_selects($parents[i]);
84 | $parents[i].addEventListener('change', function() {
85 | handle_dependent_selects(this)
86 | })
87 | }
88 | }, false);
--------------------------------------------------------------------------------
/optimalflow/webapp/templates/about.html:
--------------------------------------------------------------------------------
1 | {% extends 'base.html' %}
2 |
3 | {% block title %} OptimalFlow About Author{% endblock %}
4 |
5 | {% block body %}
6 |
7 |
About Me
8 |
9 |
I am a healthcare & pharmaceutical data scientist and big data Analytics & AI enthusiast, living in Boston area.
10 |
In my spare time, I developed OptimalFlow library to help data scientists building optimal models in an easy way, and automate Machine Learning workflow with simple codes.
11 |
As a big data insights seeker, process optimizer, and AI professional with years of analytics experience, I use machine learning and problem-solving skills in data science to turn data into actionable insights while providing strategic and quantitative products as solutions for optimal outcomes.
12 |
You can connect with me on my LinkedIn or GitHub .
13 |
{% block title %} {% endblock %}
6 |
7 |
14 |
15 |
16 |
OptimalFlow
17 |
18 |
20 |
21 |
43 |
44 |
Fork This
48 |
49 |
50 | {% block body %}
51 |
52 |
53 | {% endblock %}
54 |
55 |
56 |
57 |
58 |
59 |
--------------------------------------------------------------------------------
/optimalflow/webapp/templates/docs.html:
--------------------------------------------------------------------------------
1 | {% extends 'base.html' %}
2 |
3 | {% block title %} OptimalFlow Documentation{% endblock %}
4 |
5 | {% block body %}
6 |
7 |
8 |
Documentation
9 |
Find Official OptimalFlow Manual Docs Here:
10 |
11 |
12 |
13 |
14 |
8 |
9 |
Logs Viewer
10 |
Quickly Check the Logs of OptimalFlow, which is Supported by autoFlow Module.
11 |
12 |
28 |
33 |
34 |
35 | {% if log_flag %}
36 |
37 | {% else %}
38 |
39 | {% endif %}
40 |
41 |
2 |
Select the Log File Above, and Click the Button to Review.
3 | NOTE: The Logs Files Will Only be Available When You've Done the PCTE Workflow Step.
4 |
--------------------------------------------------------------------------------
/optimalflow/webapp/templates/viz.html:
--------------------------------------------------------------------------------
1 | {% extends 'base.html' %}
2 |
3 | {% block title %} OptimalFlow Visualization{% endblock %}
4 |
5 | {% block body %}
6 |
7 |
8 |
Visualization
9 |
Quickly Generate PCTE Model Evaluation Report or Retrieval Diagram, which are Supported by autoViz Module.
10 |
11 |
12 |
13 |
14 |
You can find more use demos from Documentation or from OptimalFlow's GitHub .
15 |
16 |
17 |
18 |
7 |
Currently only support Pipeline Cluster Retrieval Diagram for Classification Problem...
8 |
9 |
You can connect with me on my LinkedIn or GitHub .
10 |
18 | --------
19 | One
20 | Two
21 |
22 |
23 | Child 1:
24 | --------
25 | Eleven
26 | Twelve
27 | Thirteen
28 | fourteen
29 | fifteen
30 |
31 |
32 | *
33 | * Note: A page can have multiple sets of this
34 | *
35 | */
36 |
37 |
38 | var handle_dependent_selects = function($parent) {
39 | var $child = document.getElementById($parent.getAttribute('data-child-id')),
40 | $selected = $parent.options[$parent.selectedIndex],
41 | parent_val = $selected.value;
42 |
43 | for (var i=0; i<$child.options.length; i++) {
44 | var $option = $child.options[i];
45 | if($option.value != '') {
46 | $option.setAttribute('hidden',true);
47 | }
48 | };
49 |
50 | if(parent_val) {
51 | var child_options = $selected.getAttribute('data-child-options'),
52 | child_options_array = child_options.split('|#');
53 |
54 | for (i=0; i<$child.options.length; i++) {
55 | var $option = $child.options[i];
56 | if ($option.value == "") {
57 | $option.innerText = "--------";
58 | continue;
59 | }
60 | if(child_options_array.indexOf($option.value) != -1) {
61 | $option.removeAttribute('hidden');
62 | }
63 | };
64 |
65 | } else {
66 | var show_text = $child.getAttribute('data-text-if-parent-empty');
67 | if(!show_text) {
68 | show_text = 'Select ' + $parent.name;
69 | }
70 | for (i=0; i<$child.options.length; i++) {
71 | var $option = $child.options[$child.selectedIndex];
72 | if ($option.value == "") {
73 | $option.innerText = '- ' + show_text + ' -';
74 | break;
75 | }
76 | };
77 | }
78 | }
79 |
80 | document.addEventListener('DOMContentLoaded', function() {
81 | var $parents = document.getElementsByClassName('dependent-selects__parent');
82 | for (var i=0; i<$parents.length; i++) {
83 | handle_dependent_selects($parents[i]);
84 | $parents[i].addEventListener('change', function() {
85 | handle_dependent_selects(this)
86 | })
87 | }
88 | }, false);
--------------------------------------------------------------------------------
/tests/webapp/templates/about.html:
--------------------------------------------------------------------------------
1 | {% extends 'base.html' %}
2 |
3 | {% block title %} OptimalFlow About Author{% endblock %}
4 |
5 | {% block body %}
6 |
7 |
About Me
8 |
9 |
I am a healthcare & pharmaceutical data scientist and big data Analytics & AI enthusiast, living in Boston area.
10 |
In my spare time, I developed OptimalFlow library to help data scientists building optimal models in an easy way, and automate Machine Learning workflow with simple codes.
11 |
As a big data insights seeker, process optimizer, and AI professional with years of analytics experience, I use machine learning and problem-solving skills in data science to turn data into actionable insights while providing strategic and quantitative products as solutions for optimal outcomes.
12 |
You can connect with me on my LinkedIn or GitHub .
13 |
{% block title %} {% endblock %}
6 |
7 |
14 |
15 |
16 |
OptimalFlow
17 |
18 |
20 |
21 |
43 |
44 |
Fork This
48 |
49 |
50 | {% block body %}
51 |
52 |
53 | {% endblock %}
54 |
55 |
56 |
57 |
58 |
59 |
--------------------------------------------------------------------------------
/tests/webapp/templates/docs.html:
--------------------------------------------------------------------------------
1 | {% extends 'base.html' %}
2 |
3 | {% block title %} OptimalFlow Documentation{% endblock %}
4 |
5 | {% block body %}
6 |
7 |
8 |
Documentation
9 |
Find Official OptimalFlow Manual Docs Here:
10 |
11 |
12 |
13 |
14 |
8 |
9 |
Logs Viewer
10 |
Quickly Check the Logs of OptimalFlow, which is Supported by autoFlow Module.
11 |
12 |
28 |
33 |
34 |
35 | {% if log_flag %}
36 |
37 | {% else %}
38 |
39 | {% endif %}
40 |
41 |
2 |
Select the Log File Above, and Click the Button to Review.
3 | NOTE: The Logs Files Will Only be Available When You've Done the PCTE Workflow Step.
4 |
--------------------------------------------------------------------------------
/tests/webapp/templates/viz.html:
--------------------------------------------------------------------------------
1 | {% extends 'base.html' %}
2 |
3 | {% block title %} OptimalFlow Visualization{% endblock %}
4 |
5 | {% block body %}
6 |
7 |
8 |
Visualization
9 |
Quickly Generate PCTE Model Evaluation Report or Retrieval Diagram, which are Supported by autoViz Module.
10 |
11 |
12 |
13 |
14 |
You can find more use demos from Documentation or from OptimalFlow's GitHub .
15 |
16 |
17 |
18 |
}})