diff --git a/Assignment Colab/Agasi_Herbert_Updated.ipynb b/Assignment Colab/Agasi_Herbert_Updated.ipynb
new file mode 100644
index 0000000..79b46df
--- /dev/null
+++ b/Assignment Colab/Agasi_Herbert_Updated.ipynb
@@ -0,0 +1 @@
+{"cells":[{"metadata":{},"cell_type":"markdown","source":"# AGASI HERBERT\n# 2019/HD07/24875U"},{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load in \n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\n# Input data files are available in the \"../input/\" directory.\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\n\n## Profiling Report\nimport pandas_profiling as pp\n# DATA VISUALIZATION\nimport matplotlib.pyplot as plt\nimport matplotlib.gridspec as gridspec\nfrom matplotlib.pylab import rcParams\nimport seaborn as sns\n\n#scikit learn libraries\n#Lineaer Models\nfrom sklearn.linear_model import LogisticRegression\nfrom sklearn.neural_network import MLPClassifier\nfrom sklearn.linear_model import SGDClassifier\nfrom sklearn.discriminant_analysis import LinearDiscriminantAnalysis\n\n\nfrom sklearn.svm import SVC\nfrom sklearn.naive_bayes import GaussianNB\n# Tree\nfrom sklearn.tree import DecisionTreeClassifier\n\n# sklearn libraries\nfrom sklearn.neighbors import KNeighborsClassifier\nfrom sklearn.model_selection import train_test_split,KFold,cross_val_score\nfrom sklearn.preprocessing import normalize\nfrom sklearn.metrics import confusion_matrix,accuracy_score,precision_score,recall_score,f1_score,matthews_corrcoef,classification_report,roc_curve\nfrom sklearn.externals import joblib\nfrom sklearn.preprocessing import StandardScaler\nfrom sklearn.decomposition import PCA\n\n# Feature Selection\nfrom sklearn.feature_selection import SelectFromModel\n\n## esemble\nfrom sklearn.ensemble import RandomForestClassifier\nfrom sklearn.ensemble import VotingClassifier\n#let's remove the annoying warnings from our cells.\nimport warnings\nwarnings.filterwarnings('ignore')\n\n#TREE\nfrom sklearn import tree\n\n#XGBOOST\nimport xgboost as xgb\nfrom xgboost.sklearn import XGBClassifier \nfrom xgboost import plot_importance\nfrom xgboost import plot_tree\n\n#eli5\nimport eli5\nfrom eli5.sklearn import PermutationImportance\n\n## Gridsearch\nfrom sklearn.model_selection import GridSearchCV\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n for filename in filenames:\n print(os.path.join(dirname, filename))\n\n# Any results you write to the current directory are saved as output.","execution_count":1,"outputs":[{"output_type":"stream","text":"/opt/conda/lib/python3.6/site-packages/sklearn/externals/joblib/__init__.py:15: FutureWarning: sklearn.externals.joblib is deprecated in 0.21 and will be removed in 0.23. Please import this functionality directly from joblib, which can be installed with: pip install joblib. If this warning is raised when loading pickled models, you may need to re-serialize those models with scikit-learn 0.21+.\n warnings.warn(msg, category=FutureWarning)\nUsing TensorFlow backend.\n","name":"stderr"},{"output_type":"stream","text":"/kaggle/input/ace-class-assignment/Test.csv\n/kaggle/input/ace-class-assignment/AMP_TrainSet.csv\n","name":"stdout"}]},{"metadata":{"_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","trusted":true},"cell_type":"code","source":"## Loading the datasets\n\nTrain = pd.read_csv(\"../input/ace-class-assignment/AMP_TrainSet.csv\")\nTest = pd.read_csv(\"../input/ace-class-assignment/Test.csv\")\n\n## Taking a preview of the data\nTrain.head()","execution_count":2,"outputs":[{"output_type":"execute_result","execution_count":2,"data":{"text/plain":" FULL_Charge FULL_AcidicMolPerc FULL_AURR980107 FULL_DAYM780201 \\\n0 5.0 0.000 0.951 74.842 \n1 4.0 5.405 0.931 71.595 \n2 5.5 5.405 0.873 73.595 \n3 5.0 4.167 0.895 66.250 \n4 7.5 8.537 0.932 64.720 \n\n FULL_GEOR030101 FULL_OOBM850104 NT_EFC195 AS_MeanAmphiMoment \\\n0 0.975 -3.663 0 0.282 \n1 0.957 -4.011 1 0.600 \n2 0.961 -2.512 0 0.593 \n3 0.999 -1.362 0 0.614 \n4 0.979 -2.091 0 0.616 \n\n AS_DAYM780201 AS_FUKS010112 CT_RACS820104 CLASS \n0 73.444 5.661 1.041 1 \n1 68.222 6.537 1.453 1 \n2 69.444 4.934 1.722 1 \n3 67.222 4.316 1.382 1 \n4 72.944 4.540 1.539 1 ","text/html":"
\n\n
\n \n
\n
\n
FULL_Charge
\n
FULL_AcidicMolPerc
\n
FULL_AURR980107
\n
FULL_DAYM780201
\n
FULL_GEOR030101
\n
FULL_OOBM850104
\n
NT_EFC195
\n
AS_MeanAmphiMoment
\n
AS_DAYM780201
\n
AS_FUKS010112
\n
CT_RACS820104
\n
CLASS
\n
\n \n \n
\n
0
\n
5.0
\n
0.000
\n
0.951
\n
74.842
\n
0.975
\n
-3.663
\n
0
\n
0.282
\n
73.444
\n
5.661
\n
1.041
\n
1
\n
\n
\n
1
\n
4.0
\n
5.405
\n
0.931
\n
71.595
\n
0.957
\n
-4.011
\n
1
\n
0.600
\n
68.222
\n
6.537
\n
1.453
\n
1
\n
\n
\n
2
\n
5.5
\n
5.405
\n
0.873
\n
73.595
\n
0.961
\n
-2.512
\n
0
\n
0.593
\n
69.444
\n
4.934
\n
1.722
\n
1
\n
\n
\n
3
\n
5.0
\n
4.167
\n
0.895
\n
66.250
\n
0.999
\n
-1.362
\n
0
\n
0.614
\n
67.222
\n
4.316
\n
1.382
\n
1
\n
\n
\n
4
\n
7.5
\n
8.537
\n
0.932
\n
64.720
\n
0.979
\n
-2.091
\n
0
\n
0.616
\n
72.944
\n
4.540
\n
1.539
\n
1
\n
\n \n
\n
"},"metadata":{}}]},{"metadata":{},"cell_type":"markdown","source":"# Profiling Report"},{"metadata":{"trusted":true},"cell_type":"code","source":"Report = pp.ProfileReport(Train, title='Pandas Profiling Report', html={'style':{'full_width':True}})\nReport.to_notebook_iframe()\n#Report.to_widgets()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"## Loking at the dimensions of the dataset\n\nTrain.shape,Test.shape","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"## Looking for misssing and null values\nTrain.isnull().any().sum()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"## Looking the data types of the data\nTrain.dtypes","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"## Looking at the description of the data \nTrain.describe()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Looking at the data:\n* There are two categorical features CLASS and NT_EFC195\n* There is a big difference between the 75th percentile and the maximum values for columns FULL_Charge and FULL_AcidicMolPerc\n* The columns FULL_AcidicMolPerc and FULL_OOBM850104 have both negative and positive values."},{"metadata":{"trusted":true},"cell_type":"code","source":"# Looking at the ratio of the binary data for the class\n\nAll = Train.shape[0]\nPositive = Train[Train['CLASS'] == 1]\nNegative = Train[Train['CLASS'] == 0]\n\nx = len(Positive)/All\ny = len(Negative)/All\n\nprint('Positives :',x*100,'%')\nprint('Negatives :',y*100,'%')\n\n## The sets are evenly balanced and hence there will be no need to use SMOTE OR NEAR MISS","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Visualizing the data class\nlabels = ['Negatives','Positives']\nclasses = pd.value_counts(Train['CLASS'], sort = True)\nclasses.plot(kind = 'bar', rot=0)\nplt.title(\"Visualizing the data class\")\nplt.xticks(range(2), labels)\nplt.xlabel(\"Class\")\nplt.ylabel(\"Frequency\")","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Plot the distribution of features\n\nThis plot shows the distribution of the data by the binary class 1 or 0. The histogram is superimposed on the other. We can visibly see the distribution of the 0 and 1 on each feature.\n\n1. Most of the data follows a normal distribution based to the two categories. The Green has highest peak (7) among all the features in total. \n2. The blue(1)s have the highest peak for features : \n* Full_AcidicMolPerc\n* FULL_AURR980107\n\n3. The categories are evenly distributed to features 1 and 0\n3. There are feature with multiple peaks i.e Feature AS_MeanAmpiMoment has 3 peaks for data in the green color.\n"},{"metadata":{"trusted":true},"cell_type":"code","source":"# distribution of features\n# This feature works with numerical data not categorical data.\n## Removing feature\"NT_EFC195\" because it is categorical\nfeatures =Train.drop([\"CLASS\",\"NT_EFC195\"],axis=1).columns\n\nplt.figure(figsize=(12,12*4))\ngs = gridspec.GridSpec(12, 1)\nfor i, cn in enumerate(Train[features]):\n ax = plt.subplot(gs[i])\n sns.distplot(Train[cn][Train.CLASS == 1], bins=50, color = 'b')\n sns.distplot(Train[cn][Train.CLASS == 0], bins=50, color = 'g')\n ax.set_xlabel('')\n ax.set_title('histogram of feature: ' + str(cn))\nplt.show()","execution_count":4,"outputs":[{"output_type":"display_data","data":{"text/plain":"","image/png":"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\n"},"metadata":{}}]},{"metadata":{"trusted":true},"cell_type":"code","source":"\n# Data splitting\nX=Train.drop(\"CLASS\",axis=1)\ny=Train[\"CLASS\"]\nseed=42\n# splitting the faeture array and label array keeping 80% for the trainnig sets\nX_train,X_test,y_train,y_test = train_test_split(X,y,test_size=0.20,shuffle=True,random_state=seed)\n\n# normalize: Scale input vectors individually to unit norm (vector length).\nX_train = normalize(X_train)\nX_test=normalize(X_test)\n#FTest = normalize(FTest)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# Spot-Check Algorithms\n\nmodels = []\nmodels.append(('LR', LogisticRegression()))\nmodels.append(('LDA', LinearDiscriminantAnalysis()))\nmodels.append(('KNN', KNeighborsClassifier()))\nmodels.append(('CART', DecisionTreeClassifier()))\nmodels.append(('NB', GaussianNB()))\nmodels.append(('SVM', SVC()))\nmodels.append(('NN',MLPClassifier()))\nmodels.append(('SG',SGDClassifier()))\n\n# evaluate each model in turn\n\nresults = []\nnames = []\nseed =25\nfor name, model in models:\n kfold = KFold(n_splits=10, random_state=seed,shuffle=True)\n cv_results = cross_val_score(model, X_train, y_train, cv=kfold, \n scoring='accuracy')\n results.append(cv_results)\n names.append(name)\n print(\"%s: %f (%f)\" % (name, cv_results.mean(), cv_results.std()))\n#print(msg)\n\nfig = plt.figure()\nfig.suptitle('Algorithm Comparison')\nax = fig.add_subplot(111)\nplt.boxplot(results)\nax.set_xticklabels(names)\nplt.show()\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"\n# Model Evaluation\n\nIn this section I choose the best three perfoming models and then look at:\n* Accuracy Score\n* Precision Score\n* Recall Score\n* F1 Score\n* Mathew's Correlation Coefficient\n* Confusion Matrix\n\nAt this point I get to know the best performing model which is the Naive Bayes. I also try to combine the weights of the three algorithms to get and average using the Vector Classifier."},{"metadata":{},"cell_type":"markdown","source":"## Naive Bayes"},{"metadata":{"trusted":true},"cell_type":"code","source":"\nNB = GaussianNB()\nNB.fit(X_train,y_train)\nNB_predicted_test_labels= NB.predict(X_test)\n\n#scoring NB\nNB_accuracy_score = accuracy_score(y_test,NB_predicted_test_labels)\nNB_precison_score = precision_score(y_test,NB_predicted_test_labels)\nNB_recall_score = recall_score(y_test,NB_predicted_test_labels)\nNB_f1_score = f1_score(y_test,NB_predicted_test_labels)\nNB_MCC = matthews_corrcoef(y_test,NB_predicted_test_labels)\n\nprint(\"accuracy_score: {}\\nprecison_score: {}\\nrecall_score: {}\\nf1_score: {}\\nMCC: {}\".format(NB_accuracy_score,NB_precison_score,NB_recall_score,NB_f1_score,NB_MCC))\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## K Nearest Neighbours"},{"metadata":{"trusted":true},"cell_type":"code","source":"knn= KNeighborsClassifier()\nknn.fit(X_train,y_train)\nknn_predicted_test_labels = knn.predict(X_test)\n\n\n#scoring knn\nknn_accuracy_score = accuracy_score(y_test,knn_predicted_test_labels)\nknn_precison_score = precision_score(y_test,knn_predicted_test_labels)\nknn_recall_score = recall_score(y_test,knn_predicted_test_labels)\nknn_f1_score = f1_score(y_test,knn_predicted_test_labels)\nknn_MCC = matthews_corrcoef(y_test,knn_predicted_test_labels)\n\nprint(\"accuracy_score: {}\\nprecison_score: {}\\nrecall_score: {}\\nf1_score: {}\\nMCC: {}\".format(knn_accuracy_score,knn_precison_score,knn_recall_score,knn_f1_score,knn_MCC))","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Neural Networks"},{"metadata":{"trusted":true},"cell_type":"code","source":"NN = MLPClassifier()\nNN.fit(X_train,y_train)\nNN_predicted_test_labels = NN.predict(X_test)\n#scoring knn\nNN_accuracy_score = accuracy_score(y_test,NN_predicted_test_labels)\nNN_precison_score = precision_score(y_test,NN_predicted_test_labels)\nNN_recall_score = recall_score(y_test,NN_predicted_test_labels)\nNN_f1_score = f1_score(y_test,NN_predicted_test_labels)\nNN_MCC = matthews_corrcoef(y_test,NN_predicted_test_labels)\n\nprint(\"accuracy_score: {}\\nprecison_score: {}\\nrecall_score: {}\\nf1_score: {}\\nMCC: {}\".format(NN_accuracy_score,NN_precison_score,NN_recall_score,NN_f1_score,NN_MCC))","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## VOTING CLASSIFIER"},{"metadata":{},"cell_type":"markdown","source":"#### Voting Classifier and algorithms that combines the weights of two or more algorithms and ajusts the weights of the Intercept and coeficient to give a better prediction. In This case I combined the best performing algorithms \n* NB Naive Bayes\n* KNN K Nearest Neighbours\n* NN Neural Networks"},{"metadata":{"trusted":true},"cell_type":"code","source":"ensemble=VotingClassifier(estimators=[('NN', NN),('knn',knn),('NB',NB)], voting='hard')\ny_pred =ensemble.fit(X_train,y_train).predict(X_test)\nprint(confusion_matrix(y_test, y_pred))\n \n \n## Scoring the Vector Claissifier \naccuracy_score = accuracy_score(y_test,y_pred)\nprecison_score = precision_score(y_test,y_pred)\nrecall_score = recall_score(y_test,y_pred)\nf1_score = f1_score(y_test,y_pred)\nMCC = matthews_corrcoef(y_test,y_pred)\n\nprint(\"accuracy_score: {}\\nprecison_score: {}\\nrecall_score: {}\\nf1_score: {}\\nMCC: {}\".format(accuracy_score,precison_score,recall_score,f1_score,MCC))","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Testing\nAt this point I know the performance of my models the next thing I need to know is the Score upon uploading my CSV on Kaggle. After fitting of which I use all my data unlike the evaluation stage this is because I already know how my model is performing and there is no need for splitting my data other than increasing the changes of my algorithm learning with increased data. I then format my data into a CSV for all my four models as a standard for uploading the work.\n### Results\n\n Model | Scores\n ------------- | -------------\n No. | Name| Score\n 1. | Naive Bayes | 0.99559\n 2. | K Nearest Neighbours | 0.77\n 3. |Neural Networks | 0.77\n 4. |Vector Classifier | 0.86494\n \n"},{"metadata":{"trusted":true},"cell_type":"code","source":"\nNB = GaussianNB()\nNB.fit(X,y)\nNB_ypred= NB.predict(Test)\nfypred = pd.DataFrame(NB_ypred)\nfypred.to_csv(\"NaiveBayes.csv\", header =[\"CLASS\"], index_label=[\"Index\"], index = True,sep=\",\" )\n## Score 0.99559\n\nknn= KNeighborsClassifier()\nknn.fit(X,y)\nknn_ypred= knn.predict(Test)\nfypred = pd.DataFrame(knn_ypred)\nfypred.to_csv(\"KNearestN.csv\", header =[\"CLASS\"], index_label=[\"Index\"], index = True,sep=\",\" )\n## 77\n\n## Neural Networks\nNN = MLPClassifier()\nNN.fit(X,y)\nNN_predicted = NN.predict(Test)\nfypred = pd.DataFrame(knn_ypred)\nfypred.to_csv(\"NeuralNetworks.csv\", header =[\"CLASS\"], index_label=[\"Index\"], index = True,sep=\",\" )\n## 77\n\n## Voting Classifier\nensemble=VotingClassifier(estimators=[('NN', NN),('knn',knn),('NB',NB)], voting='hard')\ny_pred =ensemble.fit(X,y).predict(Test)\nfypred = pd.DataFrame(y_pred)\nfypred.to_csv(\"VotingClass.csv\", header =[\"CLASS\"], index_label=[\"Index\"], index = True,sep=\",\" )\n## 0.86494","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"\n# Model Tunning\n\nIn this section I am making adjustments to both my model and data in search of better Scores to **Accuracy,MCC,F1, Precision and Recall scores.**\n* Feature Selection\n* PCA\n* Hyper parameter Tunning\n* Outliers"},{"metadata":{},"cell_type":"markdown","source":"## Principal Component Analysis\n\n\nRunning clustering n a raw expression matrix can be very challenging due to the high dimentionality.\nTherefore is almost recommended to perform any sort of dimentionality reduction before proceeding with cluster analysis. Examples of clustering algorithms\n* K means\n* Gausian Mixture Models\n* Hierarchical Clustering\n* Spectral clustering\n* SC3\n* Boostrap Concesus Clustering\n* SNN-cliq\n* Seurat\n\nIn this I am going to use kneares neighbours because it of those i got when i did the spot check on the algorithms and it had an **accuracy of 91% and MCC value of 82%.** "},{"metadata":{"trusted":true},"cell_type":"code","source":"# PCA (Principal Component Analysis) mainly using to reduce the size of the feature space while\n# retaining as much of the information as possible.\n# In here all the features transformed into 2 features using PCA.\n\nX = Train.drop(['CLASS'], axis = 1)\ny = Train['CLASS']\n\nS = Test\n#Sy =Test['CLASS']\n\npca = PCA(n_components=2)\nprincipalComponents = pca.fit_transform(X.values)\nprincipalComponents_Test = pca.fit_transform(S.values)\nprincipalDf = pd.DataFrame(data = principalComponents\n , columns = ['principal component 1', 'principal component 2'])\nprincipalDf_test= pd.DataFrame(data = principalComponents_Test\n , columns = ['principal component 1', 'principal component 2'])\nFTrain = pd.concat([principalDf, y], axis = 1)\nFTest = pd.DataFrame(principalDf_test)\nFTrain.head()\nFTest.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### 2D visualization\nThe purpose of the 2d visualization is to ensure that my data is still balanced\nafter doing the Dimention reduction.\nIt is visible that the data is still balanced."},{"metadata":{"trusted":true},"cell_type":"code","source":"# 2D visualization\nfig = plt.figure(figsize = (8,8))\nax = fig.add_subplot(1,1,1) \nax.set_xlabel('Principal Component 1', fontsize = 15)\nax.set_ylabel('Principal Component 2', fontsize = 15)\nax.set_title('2 component PCA', fontsize = 20)\ntargets = [0, 1]\ncolors = ['r', 'g']\nfor target, color in zip(targets,colors):\n indicesToKeep = FTrain['CLASS'] == target\n ax.scatter(FTrain.loc[indicesToKeep, 'principal component 1']\n , FTrain.loc[indicesToKeep, 'principal component 2']\n , c = color\n , s = 50\n , alpha=0.5)\nax.legend(targets)\nax.grid()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## K Neares Neighbours with PCA"},{"metadata":{},"cell_type":"markdown","source":"### K Nearest Neighbours with PCA\n\nIn here I am going to run the algorithm on KNN and check for an improvement on the Score. However here I am going to use all my data to increase the chances of getting a better Score on Kaggle. The output data is going to be written to a csv file \"KNearestN.csv\" "},{"metadata":{"trusted":true},"cell_type":"code","source":"# K Neares Neighbours with PCA\nPCA_X_Train = FTrain.drop(\"CLASS\",axis=1)\nPCA_Y_Train = FTrain[\"CLASS\"]\n\nknn= KNeighborsClassifier()\nknn.fit(PCA_X_Train,PCA_Y_Train)\nknn_ypred= knn.predict(FTest)\nfypred = pd.DataFrame(knn_ypred)\nfypred.to_csv(\"KNearestN.csv\", header =[\"CLASS\"], index_label=[\"Index\"], index = True,sep=\",\" )","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# XGBOOST"},{"metadata":{},"cell_type":"markdown","source":"#### XGBoost is an optimized distributed gradient boosting library designed to be highly efficient, flexible and portable. It implements machine learning algorithms under the Gradient Boosting framework. XGBoost provides a parallel tree boosting (also known as GBDT, GBM) that solve many data science problems in a fast and accurate way."},{"metadata":{"trusted":true},"cell_type":"code","source":"param={'eta':0.3,'max_depth':3,'objective':'multi:softprob', 'num_class':3}\nsteps = 40\nD_train = xgb.DMatrix(X_train,label=y_train)\nD_test = xgb.DMatrix(X_test,label=y_test)\nD_FTest = xgb.DMatrix(FTest)\nXGB = xgb.train(param,D_train,steps)\n##Predictions\n\npreds =XGB.predict(D_test)\n#y\n\nbest_preds =pd.DataFrame(np.asarray([np.argmax(line) for line in preds]))\n#y_test=pd.DataFrame(y_test)\n\n#accuracy_score = accuracy_score(y_test,best_preds)\nprecison_score = precision_score(y_test,best_preds)\n#recall_score = recall_score(y_test,best_preds)\n#f1_score = f1_score(y_test,best_preds)\nMCC = matthews_corrcoef(y_test,best_preds)\nprint(confusion_matrix(y_test,best_preds))\n\nprint(\"precison_score: {}\\nMCC: {}\".format(precison_score,MCC))\n\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Feature Selection\nUnder feature selection I try to choose the features with the highest variance other those with the highest correlation. This is in exclusion of the Target column. For this use a feature importance plot using XGBOOST CLASSIFIER. The feature with the lowest score is the **NT_EFC195** which is also evident given that it is categorical and it has an unbalanced dataset. If used it could bias the algorithm. \n\n\n\n"},{"metadata":{},"cell_type":"markdown","source":"## Visualizing Feature with Permuattion Importance & Feature weight\nWith this code we try to look at the weights added by the features"},{"metadata":{},"cell_type":"markdown","source":"## Feature Importance plot"},{"metadata":{"trusted":true},"cell_type":"code","source":"xgb1=XGBClassifier()\nxgb1.fit(X_train,y_train)\n# plot feature importance\nplot_importance(xgb1)\nplt.show()\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"#### Looking at the feature with the lowest score **NT_EFC195**"},{"metadata":{"trusted":true},"cell_type":"code","source":"Train.NT_EFC195.value_counts().plot(kind=\"bar\")","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Feature weights plots"},{"metadata":{"trusted":true},"cell_type":"code","source":"X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)\nxgb2 = XGBClassifier().fit(X_train, y_train)\nperm = PermutationImportance(xgb2).fit(X_test, y_test)\neli5.show_weights(perm)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Ploting Descision Trees\nIn this section we look at the algorithm in a tree format on how the XGB Classifier grouped our data. "},{"metadata":{"trusted":true},"cell_type":"code","source":"xgb3 = XGBClassifier().fit(X_train, y_train)\n#set up the parameters\nrcParams['figure.figsize'] = 150,300\nplot_tree(xgb3)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"### Running XGBOOST without feature **NT_EFC195**"},{"metadata":{"trusted":true},"cell_type":"code","source":"## Running XGBOOST after filtering out the featuring with the lowest.\nXX=Train.drop([\"CLASS\",\"NT_EFC195\"],axis=1)\nTest_NT95 = Test.drop(\"NT_EFC195\",axis=1)\nsteps = 40\nmodel = XGBClassifier()\nmodel.fit(XX,y)\ny_pred = model.predict(Test_NT95)\nfypred = pd.DataFrame(y_pred)\nfypred.to_csv(\"XGB-NT_EFC195.csv\", header =[\"CLASS\"], index_label=[\"Index\"], index = True,sep=\",\" )","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Grid Search\nModel hyperparameter is a characteristic of a model that is external to the model and whose value cannot be estimated from data. The value of the hyperparameter has to be set before the learning process begins. With the Grid search we provide a range of values (hyperparameters) and the grid algorithm continuously loops through the set of values to get the most optimal values for the model."},{"metadata":{"trusted":true},"cell_type":"code","source":"##\n#This module works perfectly but i have commented it out to reduce the time required to run the jobs.\n\nGS = xgb.XGBClassifier()\nparameters={\"eta\":[0.05,0.10,.015,0.20,0.25,0.30],\"max_depth\":[3,4,5,6,8,10,12,15],\"min_child_weight\":[1,3,5,7],\"gamma\":[0.0,0.1,0.2,0.3,0.4],\"colsample_bytree\":[0.3,0.4,0.5,0.7]}\ngrid=GridSearchCV(GS,parameters,n_jobs=4,scoring=\"neg_log_loss\",cv=3)\ngrid.fit(X_train,y_train)\n\npreds = grid.predict(X_test)\nmatthews_corrcoef(y_test,preds)\n\n## Makinga prediction on the Test data\ngrid.fit(X,y)\ngrid_preds =grid.predict(Test)\nfypred = pd.DataFrame(grid_preds)\nfypred.to_csv(\"Grid.csv\", header =[\"CLASS\"], index_label=[\"Index\"], index = True,sep=\",\" )\n\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Submition\n## XGB scores best among all the models that I have tried to tune.\n### Challenge is inconsistent results each time i run the algorithms."},{"metadata":{"trusted":true},"cell_type":"code","source":"#param={'eta':0.3,'max_depth':3,'objective':'multi:softprob', 'num_class':3}\nsteps = 40\nmodel = XGBClassifier()\nmodel.fit(X,y)\ny_pred = model.predict(Test)\nfypred = pd.DataFrame(y_pred)\nfypred.to_csv(\"XGB_ALL_FEAT.csv\", header =[\"CLASS\"], index_label=[\"Index\"], index = True,sep=\",\" )\n#XGB = model.train(param,model,steps)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Model Tunning Scores Kaggle\n\n\n\nModel | Private | Public\n------- |---------|------\nKNN_PCA | 0.71371 |0.61176\nXGBOOST | 0.91494 |0.81723\nXGBOOst-N195| 0.91463 |0.80901\nGrid |0.92764 | 0.82605\n\n\n\nBased on the scores on kaggle. The Grid has the best score on note only the public score board but also the private score board.\n> The difference between the two is that the public is tested on the 80% of the data whereas the private is tested on the 20%."},{"metadata":{},"cell_type":"markdown","source":"# References\n1. https://github.com/atwine/ace-class-notes.git \n2. https://scikit-learn.org/stable/supervised_learning.html#supervised-learning\n3. https://towardsdatascience.com/grid-search-for-model-tuning-3319b259367e\n4. https://towardsdatascience.com/k-means-clustering-with-scikit-learn-6b47a369a83c\n5. https://towardsdatascience.com/dive-into-pca-principal-component-analysis-with-python-43ded13ead21\n\n"}],"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":4,"nbformat_minor":4}
\ No newline at end of file