https://github.com/firmai/atspy
Tip revision: bae5326e3c2803f8458919b70f034c57567be5c6 authored by Derek Snow on 18 December 2021, 09:26:18 UTC
Update README.md
Update README.md
Tip revision: bae5326
new.txt
1. ##### **Model Interpretability**
* Model Respecification
Protected Values Prediction
Model Constraints
Hyperparameter Modelling
Interpretable Model
Global Explanations
Level Two Monotonicity Feature Explanations
* Quantitative Validation
Level Two Monotonicity
Relationship Analysis
Partial Dependence (LV1) Monotonicity
Feature Interaction
Metrics
1. ##### **Model Robustness**
* Residual Deviation
* Residual Explanations
* Benchmark Competition
* Adversarial Attack
1. ##### **Regulatory Fairness**
* Group
Disparate Error Analysis
Parity Indicators
Fair Lending Measures
Model Agnostic Processing
Reweighing Preprocessing
Disparate Impact Preprocessing
Calibrate Equalized Odds
Feature Decomposition
* Individual
1. Reasoning
* Individual Disparity
* Reasoning Codes
1. Example Based
* Prototypical
* Counterfactual
* Contrastive
#### [Table Processing](#scrollTo=VO6NxvYjofas&line=1&uniqifier=1)
- [Configure Pandas](#configure-pandas)
- [Data Frame Formatting](#data-frame-formatting)
- [Data Frames for Testing](#data-frames-for-testing)
- [Lower Case Columns](#lower-case-columns)
- [Front and Back Column Selection](#front-and-back-columns)
- [Fast Data Frame Split](#fast-data-frame-split)
- [Create Features and Labels List](#create-features-and-labels-list)
- [Short Basic Commands](#short-basic-commands)
- [Read Commands](#read-commands)
- [Create Ordered Categories](#create-ordered-categories)
- [Select Columns Based on Regex](#select-columns-based-on-regex)
- [Accessing Group of Groupby Object](#accessing-group-of-groupby-object)
- [Multiple External Selection Criteria](#multiple-external-selection-criteria)
- [Memory Reduction Script](#memory-reduction-script)
- [Verify Primary Key](#verify-primary-key)
- [Shift Columns to Front](#shift-columns-to-front)
- [Multiple Column Assignment](#multiple-column-assignment)
- [Method Changing Technique](#method-chaning-event)
- [Load Multiple Files](#load-multiple-files)
- [Drop Rows and Column Substring](#drop-rows-and-column-substring)
- [Explode a Column](#explode-a-column)
- [Nest List Back into Column](#nest-list-back-into-column)
- [Split Cells with List](#split-cells-with-list)
#### [Table Exploration](#table-exploration)
- [Groupby Functionality](#groupby-functionality)
- [Cross Correlation Series Without Duplicates](#cross-correlation-series-without-duplicates)
- [Missing Data Report](#missing-data-report)
- [Duplicated Rows Report](#duplicated-rows-report)
- [Skewness](#skewness)
#### [Feature Processing](#feature-processing)
- [Replace Infrequently Occurring Categories](#replace-infrequently-occuring-categories)
- [Quasi-Constant Feature Detection](#quasi-constant-feature-detection)
- [Filling Missing Values Separately](#filling-missing-values-separately)
- [Conditioned Column Value Replacement](#conditioned-column-value-replacement)
- [Remove Non-numeric Values in Data Frame](#remove-non-numeric-values-in-data-frame)
- [Feature Scaling, Normalisation, Standardisation](#feature-scaling-normalisation-standardisation)
#### [Feature Engineering](#feature-engineering)
- [Automated Dummy Encoding](#automate-dummy-encodings)
- [Binarise Empty Columns](#binarise-empty-columns)
- [Polynomials](#polynomials)
- [Transformations](#transformations)
- [Genetic Programming](#genetic-programming)
- [Principal Component](#principal-component)
- [Multiple Lags](#multiple-lags)
- [Multiple Rolling](#multiple-rolling)
- [Date Features](#data-features)
- [Haversine Distance](#havervsine-distance)
- [Parse Address](#parse-address)
- [Processing Strings in Pandas](#processing-strings-in-pandas)
- [Filtering Strings in Pandas](#filtering-strings-in-pandas)
#### [Model Validation](#scrollTo=VO6NxvYjofas&line=1&uniqifier=1)
- [Classification Metrics](#classification-metrics)
I currently have you listed as a potential contributing member to Medium FirmAI.Is this something you would be interested in? For contributors - https://medium.com/firmai/example-3a-short-format-bulletin-7cbd94d8754c
. Our first publications would be from Vasant Dhar and Foster Provost. Is this something you would be interested in? For contributors - https://medium.com/firmai/example-3a-short-format-bulletin-7cbd94d8754c and for quality examples https://medium.com/@derek_snow/quality-post-examples-f52a887ecb35
I currently have you listed as a potential contributing member to Medium FirmAI. Is this something you would be interested in? Contributing Fields - https://medium.com/firmai/example-3a-short-format-bulletin-7cbd94d8754c - Our first contributors are Vasant Dhar, Foster Provost, and Isil Erel.
Hi, I listed you as a potential contributor to Medium FirmAI. Is this something you would be interested in? Contributing Fields - https://medium.com/firmai/example-3a-short-format-bulletin-7cbd94d8754c - Our first contributors are Vasant Dhar, Foster Provost, and Isil Erel; starting 4 May.
#### overall performance
'Prevalence ': '(tp + fn) / (tp + tn +fp + fn)', # how much default actually happens for this group
#'Adverse Impact': '(tp + fp) / (tp + tn + fp + fn)', # how often the model predicted default for each group
'Accuracy ': '(tp + tn) / (tp + tn + fp + fn)', # how often the model predicts default and non-default correctly for this group
#### predicting default will happen
# (correctly)
'True Positive Rate ': 'tp / (tp + fn)', # out of the people in the group *that did* default, how many the model predicted *correctly* would default
'Precision ': 'tp / (tp + fp)', # out of the people in the group the model *predicted* would default, how many the model predicted *correctly* would default
#### predicting default won't happen
# (correctly)
'Specificity ': 'tn / (tn + fp)', # out of the people in the group *that did not* default, how many the model predicted *correctly* would not default
'Negative Predicted Value ': 'tn / (tn + fn)', # out of the people in the group the model *predicted* would not default, how many the model predicted *correctly* would not default
#### analyzing errors - type I
# false accusations
'False Positive Rate ': 'fp / (tn + fp)', # out of the people in the group *that did not* default, how many the model predicted *incorrectly* would default
'False Discovery Rate ': 'fp / (tp + fp)', # out of the people in the group the model *predicted* would default, how many the model predicted *incorrectly* would default
#### analyzing errors - type II
# costly ommisions
'False Negative Rate ': 'fn / (tp + fn)', # out of the people in the group *that did* default, how many the model predicted *incorrectly* would not default
'False Omissions Rate ':'fn / (tn + fn)' # out of the people in the group the model *predicted* would not default, how many the model predicted *incorrectly* would not default
}