Update Hands-On-Quantum-Machine-Learning-With-Python-Vol-1

Author: frankzickert

chapter_01.ipynb

@@ -34,7 +34,6 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "\n",
     "sudo apt-get update\n",
     "sudo apt-get upgrade\n",
     "sudo apt-get install -y build-essential wget python3-dev \\\n",

chapter_02.ipynb

@@ -64,6 +64,7 @@
     }
    ],
    "source": [
+    "# Listing 2.1: Load the data from the csv‐files\n",
     "import pandas as pd\n",
     "\n",
     "train = pd.read_csv('./data/train.csv')\n",
@@ -86,6 +87,7 @@
     }
    ],
    "source": [
+    "# Listing 2.2: The shapes of the Titanic datasets\n",
     "print('train has {} rows and {} columns'.format(*train.shape))\n",
     "print('test has {} rows and {} columns'.format(*test.shape))"
    ]
@@ -106,6 +108,7 @@
     }
    ],
    "source": [
+    "# Listing 2.3: The structure of the train dataset\n",
     "train.info()"
    ]
   },
@@ -125,6 +128,7 @@
     }
    ],
    "source": [
+    "# Listing 2.4: The structure of the test dataset\n",
     "test.info()"
    ]
   },
@@ -170,6 +174,7 @@
     }
    ],
    "source": [
+    "# Listing 2.5: Look at the data\n",
     "train.head()"
    ]
   },
@@ -196,6 +201,7 @@
     }
    ],
    "source": [
+    "# Listing 2.6: Cope with missing values\n",
     "# option 1\n",
     "# We only have two passengers without it. This is bearable\n",
     "train = train.dropna(subset=[\"Embarked\"])    \n",
@@ -234,6 +240,7 @@
     }
    ],
    "source": [
+    "# Listing 2.7: Unique values in columns\n",
     "print('There are {} different (unique) PassengerIds in the data'\n",
     "    .format(train[\"PassengerId\"].nunique()))\n",
     "print('There are {} different (unique) names in the data'\n",
@@ -258,6 +265,7 @@
     }
    ],
    "source": [
+    "# Listing 2.8: Remove identifying data\n",
     "train = train.drop(\"PassengerId\", axis=1)\n",
     "train = train.drop(\"Name\", axis=1)\n",
     "train = train.drop(\"Ticket\", axis=1)\n",
@@ -293,6 +301,7 @@
     }
    ],
    "source": [
+    "# Listing 2.9: Transforming textual data into numbers\n",
     "from sklearn.preprocessing import LabelEncoder\n",
     "le = LabelEncoder()\n",
     "\n",
@@ -319,6 +328,7 @@
     }
    ],
    "source": [
+    "# Listing 2.10: The maximum values\n",
     "print('The maximum age is {}'.format(train[\"Age\"].max()))\n",
     "print('The maximum fare is {}'.format(train[\"Fare\"].max()))"
    ]
@@ -339,6 +349,7 @@
     }
    ],
    "source": [
+    "# Listing 2.11: Normalization of the data.\n",
     "from sklearn.preprocessing import MinMaxScaler\n",
     "\n",
     "scaler = MinMaxScaler()\n",
@@ -365,6 +376,7 @@
     }
    ],
    "source": [
+    "# Listing 2.12: Separating input from labels and training from testing sets\n",
     "from sklearn.model_selection import train_test_split\n",
     "\n",
     "input_data = train[:, 1:8]\n",
@@ -383,6 +395,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "# Listing 2.13: Save the data to the filesystem\n",
     "import numpy as np\n",
     "\n",
     "with open('data/train.npy', 'wb') as f:\n",
@@ -408,6 +421,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "# Listing 2.14: A random classifier\n",
     "import random\n",
     "random.seed(a=None, version=2)\n",
     " \n",
@@ -421,6 +435,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "# Listing 2.15: The classification runner\n",
     "def run(f_classify, x):\n",
     "    return list(map(f_classify, x))"
    ]
@@ -433,6 +448,7 @@
    },
    "outputs": [],
    "source": [
+    "# Listing 2.16: Run the classifier\n",
     "result = run(classify, train_input)"
    ]
   },
@@ -461,6 +477,7 @@
     }
    ],
    "source": [
+    "# Listing 2.17: Evaluate the classifier\n",
     "def evaluate(predictions, actual):\n",
     "    correct = list(filter(\n",
     "        lambda item: item[0] == item[1],\n",
@@ -488,6 +505,7 @@
     }
    ],
    "source": [
+    "# Listing 2.18: Always predict a passenger died\n",
     "def predict_death(item):\n",
     "    return 0\n",
     "\n",
@@ -519,6 +537,7 @@
     }
    ],
    "source": [
+    "# Listing 2.19: Confustion matrix of the predict death classifier\n",
     "from sklearn.metrics import confusion_matrix\n",
     "\n",
     "predictions = run(predict_death, train_input)\n",
@@ -541,6 +560,7 @@
     }
    ],
    "source": [
+    "# Listing 2.20: The precision score\n",
     "from sklearn.metrics import precision_score\n",
     "print('The precision score of the predict_death classifier is {}'\n",
     "    .format(precision_score(train_labels, predictions)))"
@@ -562,6 +582,7 @@
     }
    ],
    "source": [
+    "# Listing 2.21: The recall score\n",
     "from sklearn.metrics import recall_score\n",
     "print('The recall score of the predict_death classifier is {}'\n",
     "    .format(recall_score(train_labels, predictions)))"
@@ -583,6 +604,7 @@
     }
    ],
    "source": [
+    "# Listing 2.22: The specificity and the npv\n",
     "def specificity(matrix):\n",
     "    return matrix[0][0]/(matrix[0][0]+matrix[0][1]) if (matrix[0][0]+matrix[0][1] > 0) else 0\n",
     "\n",
@@ -611,6 +633,7 @@
     }
    ],
    "source": [
+    "# Listing 2.23: The scores of the random classifier\n",
     "random_predictions = run(classify, train_input)\n",
     "random_cm = confusion_matrix(train_labels, random_predictions)\n",
     "\n",
@@ -637,6 +660,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "# Listing 2.24: A hypocrite classifier\n",
     "def hypocrite(passenger, weight):\n",
     "    return round(min(1,max(0,weight*0.5+random.uniform(0, 1))))"
    ]
@@ -657,6 +681,7 @@
     }
    ],
    "source": [
+    "# Listing 2.25: The scores of the hypocrite classifier\n",
     "w_predictions = run(lambda passenger: hypocrite(passenger, -0.5), train_input)\n",
     "w_cm = confusion_matrix(train_labels, w_predictions)\n",
     "\n",
@@ -678,6 +703,7 @@
    },
    "outputs": [],
    "source": [
+    "# Listing 2.26: Run the hypocrite classifiers\n",
     "import numpy as np\n",
     "\n",
     "# number of steps to consider between -1 and 1\n",
@@ -736,6 +762,7 @@
     }
    ],
    "source": [
+    "# Listing 2.27: Plot the distribution of predictions\n",
     "import matplotlib.pyplot as plt\n",
     "import matplotlib\n",
     "\n",
@@ -768,6 +795,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "# Listing 2.28: Metrics of the hypocrite classifier\n",
     "l_precision = list(map(lambda step: precision_score(train_labels, l_predictions[step]),steps))\n",
     "l_recall = list(map(lambda step: recall_score(train_labels, l_predictions[step]),steps))\n",
     "l_specificity = list(map(lambda step: specificity(l_cm[step]),steps))\n",
@@ -808,6 +836,7 @@
     }
    ],
    "source": [
+    "# Listing 2.29: Plot the performance measures\n",
     "m_precision, = plt.plot(weights, l_precision, 'pink', label=\"precision\")\n",
     "m_recall, = plt.plot(weights, l_recall, 'cyan', label=\"recall\")\n",
     "m_specificity, = plt.plot(weights, l_specificity, 'gold', label=\"specificity\")\n",
@@ -851,6 +880,7 @@
     }
    ],
    "source": [
+    "# Listing 2.30: Calculating the mean of the measures\n",
     "l_mean = list(map(lambda step: sum(step)*0.25, zip(l_precision, l_recall, l_specificity, l_npv)))\n",
     "m_mean, = plt.plot(weights, l_mean, 'pink', label=\"Mean of the measures\")\n",
     "\n",
@@ -866,6 +896,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "# Listing 2.31: A reusable function to unmask the hypocrite classifier\n",
     "def classifier_report(name, run, classify, input, labels):\n",
     "    cr_predictions = run(classify, input)\n",
     "    cr_cm = confusion_matrix(labels, cr_predictions)\n",
@@ -904,13 +935,13 @@
     }
    ],
    "source": [
+    "# Listing 2.32: The report of the random classifier\n",
     "classifier_report(\n",
     "    \"Random PQC\", \n",
     "    run,\n",
     "    classify,\n",
     "    train_input,\n",
-    "    train_labels)\n",
-    "#CAPTION The report of the random classifier"
+    "    train_labels)"
    ]
   }
  ]