Add simple linear regression

.gitignore

@@ -0,0 +1,2 @@
+venv
+

regression/simple_linear_regression/.gitignore

@@ -0,0 +1,2 @@
+*.csv
+.ipynb_checkpoints

regression/simple_linear_regression/simple_regression.md

@@ -0,0 +1,289 @@
+# Simple Regression
+
+Configure the project. Indeed you create a dataset in csv format.
+
+
+```python
+! rm -rf *.csv
+! unzip ./dataset/archive.zip
+! head -n 1 *.csv | head -n 2 | tail -n 1 > data.csv && for file in *.csv; do (tail -n +2 "$file"; echo) >> data.csv; done && sed -i '/^$/d' data.csv
+```
+
+    Archive:  ./dataset/archive.zip
+      inflating: score.csv               
+      inflating: score_updated.csv       
+
+
+Import needed libraries
+
+
+```python
+import matplotlib.pyplot as plt
+import pandas as pd
+import numpy as np
+import pylab as pl
+from sklearn.model_selection import train_test_split
+from sklearn import linear_model
+from sklearn.metrics import r2_score
+
+%matplotlib inline
+```
+
+Read data from data.csv using pandas and store in data frame structure. Also shuffle data to have uniform distribution. 
+
+
+```python
+df = pd.read_csv("data.csv")
+df.head()
+df = df.sample(frac=1.0, random_state=42).reset_index(drop=True)
+df.head()
+```
+
+
+
+
+<div>
+<style scoped>
+    .dataframe tbody tr th:only-of-type {
+        vertical-align: middle;
+    }
+
+    .dataframe tbody tr th {
+        vertical-align: top;
+    }
+
+    .dataframe thead th {
+        text-align: right;
+    }
+</style>
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>Hours</th>
+      <th>Scores</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>0</th>
+      <td>7.4</td>
+      <td>69</td>
+    </tr>
+    <tr>
+      <th>1</th>
+      <td>3.8</td>
+      <td>35</td>
+    </tr>
+    <tr>
+      <th>2</th>
+      <td>3.5</td>
+      <td>30</td>
+    </tr>
+    <tr>
+      <th>3</th>
+      <td>1.6</td>
+      <td>19</td>
+    </tr>
+    <tr>
+      <th>4</th>
+      <td>5.1</td>
+      <td>47</td>
+    </tr>
+  </tbody>
+</table>
+</div>
+
+
+
+
+```python
+# summarize data
+df.describe() 
+```
+
+
+
+
+<div>
+<style scoped>
+    .dataframe tbody tr th:only-of-type {
+        vertical-align: middle;
+    }
+
+    .dataframe tbody tr th {
+        vertical-align: top;
+    }
+
+    .dataframe thead th {
+        text-align: right;
+    }
+</style>
+<table border="1" class="dataframe">
+  <thead>
+    <tr style="text-align: right;">
+      <th></th>
+      <th>Hours</th>
+      <th>Scores</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <th>count</th>
+      <td>121.000000</td>
+      <td>121.000000</td>
+    </tr>
+    <tr>
+      <th>mean</th>
+      <td>5.214876</td>
+      <td>53.495868</td>
+    </tr>
+    <tr>
+      <th>std</th>
+      <td>2.499189</td>
+      <td>24.988705</td>
+    </tr>
+    <tr>
+      <th>min</th>
+      <td>1.000000</td>
+      <td>12.000000</td>
+    </tr>
+    <tr>
+      <th>25%</th>
+      <td>3.000000</td>
+      <td>30.000000</td>
+    </tr>
+    <tr>
+      <th>50%</th>
+      <td>5.100000</td>
+      <td>54.000000</td>
+    </tr>
+    <tr>
+      <th>75%</th>
+      <td>7.400000</td>
+      <td>75.000000</td>
+    </tr>
+    <tr>
+      <th>max</th>
+      <td>9.800000</td>
+      <td>99.000000</td>
+    </tr>
+  </tbody>
+</table>
+</div>
+
+
+
+Print the histogram chart of data
+
+
+```python
+viz = df[["Hours", "Scores"]]
+viz.hist()
+plt.show()
+```
+
+
+
+![png](simple_regression_files/simple_regression_9_0.png)
+
+
+
+Print scatter chart of data to recognize the patterns of data. Based on the below chart we must answer to this question "Is Linear Simple Regression good or not?"
+
+
+```python
+plt.scatter(df.Hours, df.Scores, color="blue")
+plt.ylabel("Scores")
+plt.xlabel("Hours of studying")
+plt.show()
+```
+
+
+
+![png](simple_regression_files/simple_regression_11_0.png)
+
+
+
+
+```python
+# print(df)
+train, temp = train_test_split(df, test_size=0.25, random_state=42)
+test, evaluate = train_test_split(temp, test_size=0.5, random_state=42)
+```
+
+
+```python
+fig = plt.figure()
+ax1 = fig.add_subplot()
+ax1.scatter(train.Hours, train.Scores, color="blue")
+ax1.scatter(test.Hours, test.Scores, color="red")
+ax1.scatter(evaluate.Hours, evaluate.Scores, color="green")
+plt.ylabel("Scores")
+plt.xlabel("Hours of studying")
+plt.show()
+```
+
+
+
+![png](simple_regression_files/simple_regression_13_0.png)
+
+
+
+Find the best fitted line based on distribution of data. 
+
+
+```python
+reg = linear_model.LinearRegression()
+train_x = np.asanyarray(train[['Hours']])
+train_y = np.asanyarray(train[['Scores']])
+reg.fit(train_x, train_y)
+
+print("Coefficients:\t", reg.coef_)
+print("Intercept:\t", reg.intercept_)
+```
+
+    Coefficients:	 [[9.86801899]]
+    Intercept:	 [1.90944816]
+
+
+
+```python
+plt.scatter(train.Hours, train.Scores, color="blue")
+plt.plot(train_x, reg.coef_[0][0]*train_x + reg.intercept_[0], "-r")
+#               y = theta1 x + theta0
+plt.ylabel("Scores")
+plt.xlabel("Hours of studying")
+```
+
+
+
+
+    Text(0.5, 0, 'Hours of studying')
+
+
+
+
+
+![png](simple_regression_files/simple_regression_16_1.png)
+
+
+
+Testing model based on Test data. Measure the R2 and MSE.
+
+
+```python
+test_x = np.asanyarray(test[['Hours']])
+test_y = np.asanyarray(test[['Scores']])
+
+test_y_ = reg.predict(test_x)
+
+print("Mean absolute error: %.2f" % np.mean(np.absolute(test_y_ - test_y)))
+print("Residual sum of squares (MSE): %.2f" % np.mean(test_y_ - test_y)**2)
+print("R2-score: %.2f" % r2_score(test_y_, test_y))
+```
+
+    Mean absolute error: 2.95
+    Residual sum of squares (MSE): 1.62
+    R2-score: 0.97
+