Random Forest Classifier Scikit-learn Titanic Example

Here’s a quick example of how to leverage Scikit-learn on JetML.

Workflow Setup

Add the following python packages to your workflow configuration.

    scikit-learn
    pandas
dtreeviz
    ```

1. Add the following Bash command

```python
    apt-get install -y graphviz
    ```

1. Run the workflow.
2. Copy over the example code below into a notebook on the running workflow (download link at the bottom).

##### Example Notebook

Titanic Random Forest Classifier Example

Import libraries


```python
from sklearn.ensemble import RandomForestClassifier

from sklearn import tree

import dtreeviz

import matplotlib.pyplot as plt

import pandas as pd

Load titanic dataset

passenger_dataframe = pd.read_csv('https://web.stanford.edu/class/archive/cs/cs109/cs109.1166/stuff/titanic.csv')

passenger_dataframe.head()

      Survived
      Pclass
      Name
      Sex
      Age
      Siblings/Spouses Aboard
      Parents/Children Aboard
      Fare

0
      0
      3
      Mr. Owen Harris Braund
      male
      22.0
      1
      0
      7.2500

1
      1
      1
      Mrs. John Bradley (Florence Briggs Thayer) Cum...
      female
      38.0
      1
      0
      71.2833

2
      1
      3
      Miss. Laina Heikkinen
      female
      26.0
      0
      0
      7.9250

3
      1
      1
      Mrs. Jacques Heath (Lily May Peel) Futrelle
      female
      35.0
      1
      0
      53.1000

4
      0
      3
      Mr. William Henry Allen
      male
      35.0
      0
      0
      8.0500

Encode string characters for scikit learn

full_df = pd.get_dummies(passenger_dataframe)

Let's choose which features to use

features = ['Sex_female','Sex_male','Fare','Age','Pclass']

X = full_df.loc[:, features]

X.head()

Lets set if this person survived or not, this is what we'll predict

y = passenger_dataframe.Survived

y.head()

0    0
1    1
2    1
3    1
4    0
Name: Survived, dtype: int64

Let's create a random forest with 100 trees

rf_n_estimators = 100 #100 trees will be created

rf_depth= 4 #4 decision nodes per tree

model = RandomForestClassifier(n_estimators=rf_n_estimators, random_state=1, max_depth=rf_depth) #define the model

model.fit(X, y) #Train/fit the model, map the passenger's features (X) to the survival (y)

predicted = model.predict(X) #Use the model to predict

passenger_dataframe['predict'] = predicted #add the new predictions to the initial passenger dataframe

Inspect first tree

viz = dtreeviz.model(model.estimators_[0], X, y, feature_names=X.columns,class_names=["died", "survived"], target_name="Target")

viz.view()

Notes about this project

This was a simple example of how to run scikit-learn's Random Forest on JetML. Easily adapt this project to your own dataset and build out your full ML workflow.

converted by JetML

download original notebook .ipynb file

Download Jupyter Notebook