{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"##########################################################\n# 1. IMPORT ALL PACKAGES\n##########################################################\nimport pandas as pd\nimport numpy as np\nfrom sklearn.linear_model import Perceptron\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.metrics import confusion_matrix\nimport math\nfrom collections import Counter\nimport seaborn as sns\nimport matplotlib.pyplot as plt #for plotting\nfrom sklearn.ensemble import RandomForestClassifier\nfrom sklearn.metrics import multilabel_confusion_matrix\n# importing mean()\nfrom statistics import mean","metadata":{"_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","papermill":{"duration":1.324169,"end_time":"2021-11-14T19:57:51.601002","exception":false,"start_time":"2021-11-14T19:57:50.276833","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-11-24T13:46:08.860722Z","iopub.execute_input":"2021-11-24T13:46:08.861342Z","iopub.status.idle":"2021-11-24T13:46:10.329199Z","shell.execute_reply.started":"2021-11-24T13:46:08.861222Z","shell.execute_reply":"2021-11-24T13:46:10.328289Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"##########################################################\n# 2. LOAD TESTING AND TRAINING DATASET\n##########################################################\ndf_train = pd.read_csv(\"../input/digit-recognizer/train.csv\",header=0)# header 0 means the first row is name of the coloumn \ndf_test = pd.read_csv(\"../input/digit-recognizer/test.csv\",header=0)# header 0 means the first row is name of the coloumn \ndf_sample_submission = pd.read_csv(\"../input/digit-recognizer/sample_submission.csv\",header=0)# header 0 means the first row is name of the coloumn \n\n# View sample data\ndf_train.head(10) ","metadata":{"papermill":{"duration":4.7778,"end_time":"2021-11-14T19:57:56.388601","exception":false,"start_time":"2021-11-14T19:57:51.610801","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-11-24T13:46:10.330365Z","iopub.execute_input":"2021-11-24T13:46:10.330566Z","iopub.status.idle":"2021-11-24T13:46:15.327437Z","shell.execute_reply.started":"2021-11-24T13:46:10.330542Z","shell.execute_reply":"2021-11-24T13:46:15.326500Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"# Check all classes/labels in trainiing data\nall_label = set(df_train['label'].tolist())\nprint(\"All labels: {0}\".format(all_label))\n\nall_label_sub = set(df_sample_submission['Label'].tolist())\nprint(\"'sample_submission.csv' only has one class: {0}\".format(all_label_sub))\n\n# Data distribution for each class\ndst_train = Counter(df_train['label'])\nprint(dst_train)\n\n# Plot distribution\nsns.countplot(df_train['label'])","metadata":{"papermill":{"duration":0.278882,"end_time":"2021-11-14T19:57:56.675671","exception":false,"start_time":"2021-11-14T19:57:56.396789","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-11-24T13:46:15.328940Z","iopub.execute_input":"2021-11-24T13:46:15.329258Z","iopub.status.idle":"2021-11-24T13:46:15.611886Z","shell.execute_reply.started":"2021-11-24T13:46:15.329217Z","shell.execute_reply":"2021-11-24T13:46:15.610995Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"##########################################################\n# 3. SHARE TO TEST AND TRAIN DATA\n##########################################################\n# We share train data because this \"sample_submission.csv\" only has one class/label\nx = df_train.iloc[:, 1:]\ny = df_train['label'].tolist()\n\n# # Select 10000 rows data as a testing dataset\nx_test = x.iloc[0:10000, :].values.astype('float32') # all pixel values \ny_test = y[0:10000] # Select label for testing data\nx_train = x.iloc[10000:, :].values.astype('float32') # all pixel values \ny_train = y[10000:]\n\n# View one row data\n#df_train.iloc[0:1, 1:].values","metadata":{"papermill":{"duration":0.101717,"end_time":"2021-11-14T19:57:56.788858","exception":false,"start_time":"2021-11-14T19:57:56.687141","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-11-24T13:46:15.613731Z","iopub.execute_input":"2021-11-24T13:46:15.614575Z","iopub.status.idle":"2021-11-24T13:46:15.689390Z","shell.execute_reply.started":"2021-11-24T13:46:15.614515Z","shell.execute_reply":"2021-11-24T13:46:15.688341Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"##########################################################\n# 4. FEATURE NORMALIZATION FOR BOTH (TEST & TRAIN)\n##########################################################\n# Proceed to normalize the features because the pixel intensities are currently between the range of 0 and 255\nprint((min(x_train[2]), max(x_train[2])))\n\n# Normalizing the data\nx_train = x_train/255.0\nx_test = x_test/255.0\n\n# Printing the shape of the Datasets\nprint('x_train shape:', x_train.shape)\nprint(x_train.shape[0], 'train samples')\nprint(x_test.shape[0], 'test samples')","metadata":{"papermill":{"duration":0.078489,"end_time":"2021-11-14T19:57:56.876577","exception":false,"start_time":"2021-11-14T19:57:56.798088","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-11-24T13:46:15.690551Z","iopub.execute_input":"2021-11-24T13:46:15.690842Z","iopub.status.idle":"2021-11-24T13:46:15.752883Z","shell.execute_reply.started":"2021-11-24T13:46:15.690812Z","shell.execute_reply":"2021-11-24T13:46:15.751904Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"##########################################################\n# 5. TRAIN RANDOM FOREST ALGORITHM\n##########################################################\n# Create a RandomForestClassifier object with the parameters over the data\n# n_estimators (default=100) = the number of trees in the forest.\n# max_depth (default=None) = the maximum depth of the tree.\nmodel_clf = RandomForestClassifier(n_estimators=500, max_depth=2, random_state=0)\n\n# Train the Random Forest algorithm\nmodel_clf.fit(x_train, y_train)","metadata":{"papermill":{"duration":13.31791,"end_time":"2021-11-14T19:58:10.203899","exception":false,"start_time":"2021-11-14T19:57:56.885989","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-11-24T13:46:15.754260Z","iopub.execute_input":"2021-11-24T13:46:15.754552Z","iopub.status.idle":"2021-11-24T13:46:29.444280Z","shell.execute_reply.started":"2021-11-24T13:46:15.754510Z","shell.execute_reply":"2021-11-24T13:46:29.443471Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"##########################################################\n# 6. APPLY THE TRAINED LEARNER TO TEST NEW DATA\n##########################################################\n# Apply the trained perceptron to make prediction of test data\ny_pred = model_clf.predict(x_test)","metadata":{"papermill":{"duration":0.471867,"end_time":"2021-11-14T19:58:10.685990","exception":false,"start_time":"2021-11-14T19:58:10.214123","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-11-24T13:46:29.445575Z","iopub.execute_input":"2021-11-24T13:46:29.445890Z","iopub.status.idle":"2021-11-24T13:46:29.917622Z","shell.execute_reply.started":"2021-11-24T13:46:29.445859Z","shell.execute_reply":"2021-11-24T13:46:29.916757Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"##########################################################\n# 7. MULTI-CLASS CONFUSION MATRIX FOR EACH CLASS\n##########################################################\n# Actual and predicted classes\nlst_actual_class = y_test\nlst_predicted_class = y_pred\n\n# Class = digit labels 0, 1, 2, 3, 4, 5, 6, 7, 8, 9\nlst_classes = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9] # Must in order\n\n# Compute multi-class confusion matrix\narr_out_matrix = multilabel_confusion_matrix(lst_actual_class, lst_predicted_class, labels=lst_classes)\n\n# Temp store results\nstore_sens = [];\nstore_spec = [];\nstore_acc = [];\nstore_bal_acc = [];\nstore_prec = [];\nstore_fscore = [];\nstore_mcc = [];\n\n# Loop for each taget label\nfor no_class in range(len(lst_classes)):\n    arr_data = arr_out_matrix[no_class];\n    print(\"Predicted Performance of Digit Label/Class: {0}\".format(no_class));\n\n    tp = arr_data[1][1]\n    fp = arr_data[0][1]\n    tn = arr_data[0][0]\n    fn = arr_data[1][0]\n\n    sensitivity = round(tp/(tp+fn), 3);\n    specificity = round(tn/(tn+fp), 3);\n    accuracy = round((tp+tn)/(tp+fp+tn+fn), 3);\n    balanced_accuracy = round((sensitivity+specificity)/2, 3);\n    precision = round(tp/(tp + fp), 3)\n    fscore = round((2 * ((precision * sensitivity) / (precision + sensitivity))), 3)\n    mcc = round((((tp * tn)-(fp * fn))/ math.sqrt((tp + fp)*(tp + fn)*(tn + fp)*(tn + fn))),3)\n    \n    store_sens.append(sensitivity);\n    store_spec.append(specificity);\n    store_acc.append(accuracy);\n    store_bal_acc.append(balanced_accuracy);\n    store_prec.append(precision);\n    store_fscore.append(fscore);\n    store_mcc.append(mcc);\n    \n    print(\"TP={0}, FP={1}, TN={2}, FN={3}\".format(tp, fp, tn, fn));\n    print(\"Sensitivity: {0}\".format(sensitivity));\n    print(\"Specificity: {0}\".format(specificity));\n    print(\"Accuracy: {0}\".format(accuracy));\n    print(\"Balanced Accuracy: {0}\".format(balanced_accuracy));\n    print(\"Precision: {0}\".format(precision));\n    print(\"F1-Score: {0}\".format(fscore))\n    print(\"MCC: {0}\\n\".format(mcc))","metadata":{"papermill":{"duration":0.046267,"end_time":"2021-11-14T19:58:10.742013","exception":false,"start_time":"2021-11-14T19:58:10.695746","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2021-11-24T13:46:29.918811Z","iopub.execute_input":"2021-11-24T13:46:29.919047Z","iopub.status.idle":"2021-11-24T13:46:29.951240Z","shell.execute_reply.started":"2021-11-24T13:46:29.919020Z","shell.execute_reply":"2021-11-24T13:46:29.950107Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{"papermill":{"duration":0.009844,"end_time":"2021-11-14T19:58:10.763133","exception":false,"start_time":"2021-11-14T19:58:10.753289","status":"completed"},"tags":[]},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{"papermill":{"duration":0.009812,"end_time":"2021-11-14T19:58:10.783046","exception":false,"start_time":"2021-11-14T19:58:10.773234","status":"completed"},"tags":[]},"execution_count":null,"outputs":[]}]}