completed part 1 of deep learning

python/Deep Learning/Introduction to Neural Networks/Student Admissions(Neural Network)/StudentAdmissionsSolutions.ipynb

@@ -0,0 +1,115 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Solutions"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### One-hot encoding the rank"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "# Make dummy variables for rank\n",
+    "one_hot_data = pd.concat([data, pd.get_dummies(data['rank'], prefix='rank')], axis=1)\n",
+    "\n",
+    "# Drop the previous rank column\n",
+    "one_hot_data = one_hot_data.drop('rank', axis=1)\n",
+    "\n",
+    "# Print the first 10 rows of our data\n",
+    "one_hot_data[:10]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Scaling the data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "# Copying our data\n",
+    "processed_data = one_hot_data[:]\n",
+    "\n",
+    "# Scaling the columns\n",
+    "processed_data['gre'] = processed_data['gre']/800\n",
+    "processed_data['gpa'] = processed_data['gpa']/4.0\n",
+    "processed_data[:10]"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Backpropagating the data"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "def error_term_formula(x, y, output):\n",
+    "    return (y - output)*sigmoid_prime(x)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "## alternative solution ##\n",
+    "# you could also *only* use y and the output \n",
+    "# and calculate sigmoid_prime directly from the activated output!\n",
+    "\n",
+    "# below is an equally valid solution (it doesn't utilize x)\n",
+    "def error_term_formula(x, y, output):\n",
+    "    return (y-output) * output * (1 - output)"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.6.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}

python/Deep Learning/Introduction to Neural Networks/Student Admissions(Neural Network)/student_admissions.py

@@ -0,0 +1,181 @@
+# Importing pandas and numpy
+import pandas as pd
+import numpy as np
+import matplotlib.pyplot as plt
+
+# Reading the csv file into a pandas DataFrame
+data = pd.read_csv('student_data.csv')
+
+# Printing out the first 10 rows of our data
+print(data[:10])
+
+
+# Importing matplotlib
+
+# Function to help us plot
+def plot_points(data):
+    X = np.array(data[["gre", "gpa"]])
+    y = np.array(data["admit"])
+    admitted = X[np.argwhere(y == 1)]
+    rejected = X[np.argwhere(y == 0)]
+    plt.scatter([s[0][0] for s in rejected], [s[0][1]
+                                              for s in rejected],
+                s=25, color='red', edgecolor='k')
+    plt.scatter([s[0][0] for s in admitted], [s[0][1]
+                                              for s in admitted],
+                s=25, color='cyan', edgecolor='k')
+    plt.xlabel('Test (GRE)')
+    plt.ylabel('Grades (GPA)')
+
+
+# Plotting the points
+plot_points(data)
+plt.show()
+
+
+# Separating the ranks
+data_rank1 = data[data["rank"] == 1]
+data_rank2 = data[data["rank"] == 2]
+data_rank3 = data[data["rank"] == 3]
+data_rank4 = data[data["rank"] == 4]
+
+# Plotting the graphs
+plot_points(data_rank1)
+plt.title("Rank 1")
+plt.show()
+plot_points(data_rank2)
+plt.title("Rank 2")
+plt.show()
+plot_points(data_rank3)
+plt.title("Rank 3")
+plt.show()
+plot_points(data_rank4)
+plt.title("Rank 4")
+plt.show()
+
+
+# TODO:  Make dummy variables for rank
+one_hot_data = pd.concat([data, pd.get_dummies(data['rank'], prefix='rank')],
+                         axis=1)
+
+# TODO: Drop the previous rank column
+one_hot_data = one_hot_data.drop('rank', axis=1)
+
+# Print the first 10 rows of our data
+one_hot_data[:10]
+
+
+# Making a copy of our data
+processed_data = one_hot_data[:]
+
+# TODO: Scale the columns
+processed_data['gre'] = processed_data['gre'] / 800
+processed_data['gpa'] = processed_data['gpa'] / 4.0
+processed_data[:10]
+
+# Printing the first 10 rows of our procesed data
+processed_data[:10]
+
+
+sample = np.random.choice(processed_data.index, size=int(
+    len(processed_data) * 0.9), replace=False)
+train_data, test_data = processed_data.iloc[sample], processed_data.drop(
+    sample)
+
+print("Number of training samples is", len(train_data))
+print("Number of testing samples is", len(test_data))
+print(train_data[:10])
+print(test_data[:10])
+
+
+features = train_data.drop('admit', axis=1)
+targets = train_data['admit']
+features_test = test_data.drop('admit', axis=1)
+targets_test = test_data['admit']
+
+print(features[:10])
+print(targets[:10])
+
+
+# Activation (sigmoid) function
+def sigmoid(x):
+    return 1 / (1 + np.exp(-x))
+
+
+def sigmoid_prime(x):
+    return sigmoid(x) * (1 - sigmoid(x))
+
+
+def error_formula(y, output):
+    return - y * np.log(output) - (1 - y) * np.log(1 - output)
+
+
+# TODO: Write the error term formula
+def error_term_formula(x, y, output):
+    return (y - output) * sigmoid_prime(x)
+
+
+# Neural Network hyperparameters
+epochs = 1000
+learnrate = 0.5
+
+# Training function
+
+
+def train_nn(features, targets, epochs, learnrate):
+
+    # Use to same seed to make debugging easier
+    np.random.seed(42)
+
+    n_records, n_features = features.shape
+    last_loss = None
+
+    # Initialize weights
+    weights = np.random.normal(scale=1 / n_features**.5, size=n_features)
+
+    for e in range(epochs):
+        del_w = np.zeros(weights.shape)
+        for x, y in zip(features.values, targets):
+            # Loop through all records, x is the input, y is the target
+
+            # Activation of the output unit
+            #   Notice we multiply the inputs and the weights here
+            #   rather than storing h as a separate variable
+            output = sigmoid(np.dot(x, weights))
+
+            # The error, the target minus the network output
+            error = error_formula(y, output)
+
+            # The error term
+            error_term = error_term_formula(x, y, output)
+
+            # The gradient descent step, the error times the gradient times the inputs
+            del_w += error_term * x
+
+        # Update the weights here. The learning rate times the
+        # change in weights, divided by the number of records to average
+        weights += learnrate * del_w / n_records
+
+        # Printing out the mean square error on the training set
+        if e % (epochs / 10) == 0:
+            out = sigmoid(np.dot(features, weights))
+            loss = np.mean((out - targets) ** 2)
+            print("Epoch:", e)
+            if last_loss and last_loss < loss:
+                print("Train loss: ", loss, "  WARNING - Loss Increasing")
+            else:
+                print("Train loss: ", loss)
+            last_loss = loss
+            print("=========")
+    print("Finished training!")
+    return weights
+
+
+weights = train_nn(features, targets, epochs, learnrate)
+
+
+# Calculate accuracy on test data
+test_out = sigmoid(np.dot(features_test, weights))
+predictions = test_out > 0.5
+accuracy = np.mean(predictions == targets_test)
+print("Prediction accuracy: {:.3f}".format(accuracy))

python/Deep Learning/Introduction to Neural Networks/Student Admissions(Neural Network)/student_data.csv

@@ -0,0 +1,401 @@
+admit,gre,gpa,rank
+0,380,3.61,3
+1,660,3.67,3
+1,800,4,1
+1,640,3.19,4
+0,520,2.93,4
+1,760,3,2
+1,560,2.98,1
+0,400,3.08,2
+1,540,3.39,3
+0,700,3.92,2
+0,800,4,4
+0,440,3.22,1
+1,760,4,1
+0,700,3.08,2
+1,700,4,1
+0,480,3.44,3
+0,780,3.87,4
+0,360,2.56,3
+0,800,3.75,2
+1,540,3.81,1
+0,500,3.17,3
+1,660,3.63,2
+0,600,2.82,4
+0,680,3.19,4
+1,760,3.35,2
+1,800,3.66,1
+1,620,3.61,1
+1,520,3.74,4
+1,780,3.22,2
+0,520,3.29,1
+0,540,3.78,4
+0,760,3.35,3
+0,600,3.4,3
+1,800,4,3
+0,360,3.14,1
+0,400,3.05,2
+0,580,3.25,1
+0,520,2.9,3
+1,500,3.13,2
+1,520,2.68,3
+0,560,2.42,2
+1,580,3.32,2
+1,600,3.15,2
+0,500,3.31,3
+0,700,2.94,2
+1,460,3.45,3
+1,580,3.46,2
+0,500,2.97,4
+0,440,2.48,4
+0,400,3.35,3
+0,640,3.86,3
+0,440,3.13,4
+0,740,3.37,4
+1,680,3.27,2
+0,660,3.34,3
+1,740,4,3
+0,560,3.19,3
+0,380,2.94,3
+0,400,3.65,2
+0,600,2.82,4
+1,620,3.18,2
+0,560,3.32,4
+0,640,3.67,3
+1,680,3.85,3
+0,580,4,3
+0,600,3.59,2
+0,740,3.62,4
+0,620,3.3,1
+0,580,3.69,1
+0,800,3.73,1
+0,640,4,3
+0,300,2.92,4
+0,480,3.39,4
+0,580,4,2
+0,720,3.45,4
+0,720,4,3
+0,560,3.36,3
+1,800,4,3
+0,540,3.12,1
+1,620,4,1
+0,700,2.9,4
+0,620,3.07,2
+0,500,2.71,2
+0,380,2.91,4
+1,500,3.6,3
+0,520,2.98,2
+0,600,3.32,2
+0,600,3.48,2
+0,700,3.28,1
+1,660,4,2
+0,700,3.83,2
+1,720,3.64,1
+0,800,3.9,2
+0,580,2.93,2
+1,660,3.44,2
+0,660,3.33,2
+0,640,3.52,4
+0,480,3.57,2
+0,700,2.88,2
+0,400,3.31,3
+0,340,3.15,3
+0,580,3.57,3
+0,380,3.33,4
+0,540,3.94,3
+1,660,3.95,2
+1,740,2.97,2
+1,700,3.56,1
+0,480,3.13,2
+0,400,2.93,3
+0,480,3.45,2
+0,680,3.08,4
+0,420,3.41,4
+0,360,3,3
+0,600,3.22,1
+0,720,3.84,3
+0,620,3.99,3
+1,440,3.45,2
+0,700,3.72,2
+1,800,3.7,1
+0,340,2.92,3
+1,520,3.74,2
+1,480,2.67,2
+0,520,2.85,3
+0,500,2.98,3
+0,720,3.88,3
+0,540,3.38,4
+1,600,3.54,1
+0,740,3.74,4
+0,540,3.19,2
+0,460,3.15,4
+1,620,3.17,2
+0,640,2.79,2
+0,580,3.4,2
+0,500,3.08,3
+0,560,2.95,2
+0,500,3.57,3
+0,560,3.33,4
+0,700,4,3
+0,620,3.4,2
+1,600,3.58,1
+0,640,3.93,2
+1,700,3.52,4
+0,620,3.94,4
+0,580,3.4,3
+0,580,3.4,4
+0,380,3.43,3
+0,480,3.4,2
+0,560,2.71,3
+1,480,2.91,1
+0,740,3.31,1
+1,800,3.74,1
+0,400,3.38,2
+1,640,3.94,2
+0,580,3.46,3
+0,620,3.69,3
+1,580,2.86,4
+0,560,2.52,2
+1,480,3.58,1
+0,660,3.49,2
+0,700,3.82,3
+0,600,3.13,2
+0,640,3.5,2
+1,700,3.56,2
+0,520,2.73,2
+0,580,3.3,2
+0,700,4,1
+0,440,3.24,4
+0,720,3.77,3
+0,500,4,3
+0,600,3.62,3
+0,400,3.51,3
+0,540,2.81,3
+0,680,3.48,3
+1,800,3.43,2
+0,500,3.53,4
+1,620,3.37,2
+0,520,2.62,2
+1,620,3.23,3
+0,620,3.33,3
+0,300,3.01,3
+0,620,3.78,3
+0,500,3.88,4
+0,700,4,2
+1,540,3.84,2
+0,500,2.79,4
+0,800,3.6,2
+0,560,3.61,3
+0,580,2.88,2
+0,560,3.07,2
+0,500,3.35,2
+1,640,2.94,2
+0,800,3.54,3
+0,640,3.76,3
+0,380,3.59,4
+1,600,3.47,2
+0,560,3.59,2
+0,660,3.07,3
+1,400,3.23,4
+0,600,3.63,3
+0,580,3.77,4
+0,800,3.31,3
+1,580,3.2,2
+1,700,4,1
+0,420,3.92,4
+1,600,3.89,1
+1,780,3.8,3
+0,740,3.54,1
+1,640,3.63,1
+0,540,3.16,3
+0,580,3.5,2
+0,740,3.34,4
+0,580,3.02,2
+0,460,2.87,2
+0,640,3.38,3
+1,600,3.56,2
+1,660,2.91,3
+0,340,2.9,1
+1,460,3.64,1
+0,460,2.98,1
+1,560,3.59,2
+0,540,3.28,3
+0,680,3.99,3
+1,480,3.02,1
+0,800,3.47,3
+0,800,2.9,2
+1,720,3.5,3
+0,620,3.58,2
+0,540,3.02,4
+0,480,3.43,2
+1,720,3.42,2
+0,580,3.29,4
+0,600,3.28,3
+0,380,3.38,2
+0,420,2.67,3
+1,800,3.53,1
+0,620,3.05,2
+1,660,3.49,2
+0,480,4,2
+0,500,2.86,4
+0,700,3.45,3
+0,440,2.76,2
+1,520,3.81,1
+1,680,2.96,3
+0,620,3.22,2
+0,540,3.04,1
+0,800,3.91,3
+0,680,3.34,2
+0,440,3.17,2
+0,680,3.64,3
+0,640,3.73,3
+0,660,3.31,4
+0,620,3.21,4
+1,520,4,2
+1,540,3.55,4
+1,740,3.52,4
+0,640,3.35,3
+1,520,3.3,2
+1,620,3.95,3
+0,520,3.51,2
+0,640,3.81,2
+0,680,3.11,2
+0,440,3.15,2
+1,520,3.19,3
+1,620,3.95,3
+1,520,3.9,3
+0,380,3.34,3
+0,560,3.24,4
+1,600,3.64,3
+1,680,3.46,2
+0,500,2.81,3
+1,640,3.95,2
+0,540,3.33,3
+1,680,3.67,2
+0,660,3.32,1
+0,520,3.12,2
+1,600,2.98,2
+0,460,3.77,3
+1,580,3.58,1
+1,680,3,4
+1,660,3.14,2
+0,660,3.94,2
+0,360,3.27,3
+0,660,3.45,4
+0,520,3.1,4
+1,440,3.39,2
+0,600,3.31,4
+1,800,3.22,1
+1,660,3.7,4
+0,800,3.15,4
+0,420,2.26,4
+1,620,3.45,2
+0,800,2.78,2
+0,680,3.7,2
+0,800,3.97,1
+0,480,2.55,1
+0,520,3.25,3
+0,560,3.16,1
+0,460,3.07,2
+0,540,3.5,2
+0,720,3.4,3
+0,640,3.3,2
+1,660,3.6,3
+1,400,3.15,2
+1,680,3.98,2
+0,220,2.83,3
+0,580,3.46,4
+1,540,3.17,1
+0,580,3.51,2
+0,540,3.13,2
+0,440,2.98,3
+0,560,4,3
+0,660,3.67,2
+0,660,3.77,3
+1,520,3.65,4
+0,540,3.46,4
+1,300,2.84,2
+1,340,3,2
+1,780,3.63,4
+1,480,3.71,4
+0,540,3.28,1
+0,460,3.14,3
+0,460,3.58,2
+0,500,3.01,4
+0,420,2.69,2
+0,520,2.7,3
+0,680,3.9,1
+0,680,3.31,2
+1,560,3.48,2
+0,580,3.34,2
+0,500,2.93,4
+0,740,4,3
+0,660,3.59,3
+0,420,2.96,1
+0,560,3.43,3
+1,460,3.64,3
+1,620,3.71,1
+0,520,3.15,3
+0,620,3.09,4
+0,540,3.2,1
+1,660,3.47,3
+0,500,3.23,4
+1,560,2.65,3
+0,500,3.95,4
+0,580,3.06,2
+0,520,3.35,3
+0,500,3.03,3
+0,600,3.35,2
+0,580,3.8,2
+0,400,3.36,2
+0,620,2.85,2
+1,780,4,2
+0,620,3.43,3
+1,580,3.12,3
+0,700,3.52,2
+1,540,3.78,2
+1,760,2.81,1
+0,700,3.27,2
+0,720,3.31,1
+1,560,3.69,3
+0,720,3.94,3
+1,520,4,1
+1,540,3.49,1
+0,680,3.14,2
+0,460,3.44,2
+1,560,3.36,1
+0,480,2.78,3
+0,460,2.93,3
+0,620,3.63,3
+0,580,4,1
+0,800,3.89,2
+1,540,3.77,2
+1,680,3.76,3
+1,680,2.42,1
+1,620,3.37,1
+0,560,3.78,2
+0,560,3.49,4
+0,620,3.63,2
+1,800,4,2
+0,640,3.12,3
+0,540,2.7,2
+0,700,3.65,2
+1,540,3.49,2
+0,540,3.51,2
+0,660,4,1
+1,480,2.62,2
+0,420,3.02,1
+1,740,3.86,2
+0,580,3.36,2
+0,640,3.17,2
+0,640,3.51,2
+1,800,3.05,2
+1,660,3.88,2
+1,600,3.38,3
+1,620,3.75,2
+1,460,3.99,3
+0,620,4,2
+0,560,3.04,3
+0,460,2.63,2
+0,700,3.65,2
+0,600,3.89,3