python 3.6

In this project, we are going to work with a fictitious charity organization located in Silicon Valley. The goal of the creation of this fictitious company is to provide financial support to pleople who are interested in learning machine learning.

• CharityML determined that every donation they received came from someone that was making more than $50,000 annually.

• pandas
• numpy
• time
• IPython.display -> allows the use of display() for DataFrames
• visuals -> supplementary visualization code visuals.py

• sklearn.preprocessing import MinMaxScaler -> transforms features by scaling each feature to a given range.

• sklearn.cross_validation import train_test_split -> Split arrays or matrices into random train and test subsets

• sklearn.metrics import accuracy_score, fbeta_score: fbeta = The F-beta score is the weighted harmonic mean of precision and recall, reaching its optimal value at 1 and its worst value at 0.

• sklearn.svm import SVC -> Suport Vector Classifier
• sklearn.tree import DecisionTreeClassifier -> Decision Tree Classifier
• sklearn.ensemble import RandomForestClassifier -> Random Forest Classifier

• sklearn.metrics import make_scorer -> make a scorer from a performance metric or loss function.
• sklearn.model_selection import GridSearchCV -> process the hyper parameter tuning to determine the optimal values
• sklearn.ensemble import AdaBoostClassifier -> adaptive boosting used in conjunction with many other types of learning algorithms to improve performance

• sklearn.base import clone -> for cloing the model and use a subset of all the available features in the data and compare models.

Finding Donors starts with a brief description of the company, the target, the dataset origin and the article by Ron Kohavi from where the project it's inspired.

This part is designated for the data exploration: number of values in the dataset, null values, featureset exploration.

Transforming skewed continuous features, normalizing numerical features, data preprocessing (get dummies), shuffle and split data.

Implementation if 3 different algorithms for comparison purposes through the creation of a training and predictive pipeline.

In this final section, the best out of three supervised learning models is selected. Then we perform a grid search optimization for the model over the entire training set (X_train and y_train) by tuning at least one parameter to improve upon the untuned model's F-score.

Final conclusions of why of the decision taken on which model we picked