aphd87

Adding-and-Dropping-columns

Add column, drop column, creating a new column, creating a where function for column outcomes, combining conditions and rules with select, and dropping a column

Adding-text-box-to-data

Add text to data, data point call out, annotate, textcoords, call outs, xytext, bounding box, fill color, edge color, arrowstyle, arrow props

Advanced-K-Means-Clustering

Create data, merge data with inner join, drop rows with non food names, group by, pivot table, missing values, normalize, WCSS to find k value, instantiate and fit model, cluster profile breakdown

Advanced-linear-regression

Pickle file, shuffle data, missing values, drop values, outliers, box plot, split, categorical variables and one hot encoder, feature selection and RFECV, feature selection, optimal number of features plot, r-square, cross validation scores, adjusted r-square, model coefficients table and intercept, final interpretation

Aggregating_Data

Merging, value counts, sum, groupby, quantile, reset_index, sum and mean for multiple columns, aggregate, specify aggregate items for which columns

Airlines-Data-Set-

Running size, column view, replace NaN, data shape, group, and other related functions on data set.

ANN-with-Desc-null-balance-bootstrap-plots-seab-label-enc-corr-compile-early-stop-hyppar.

ANN with Desc, null, balance (bootstrap), plots, seab, label enc, corr , compile, early stop, hyppar.

Apply-Map-Reduce

Map different dictionary keys to values, run replace, write a function and view length, use function to code column, apply changes to data frame, rows and columns

NLP-index-tokenize-sentence-detection-NER-POS-chunking-tagging-parsing-vectors-similarities.

NLP index, tokenize, sentence detection, NER, POS, chunking, tagging, parsing, vectors, similarities, and pipelines with spaCY words.

Use-of-correlations-heatmap-plotted-logistic-regression-chart.

Use of correlations, heatmap, plotted logistic regression chart in examination of insurance data.

aphd87/Advanced-K-Means-Clustering

Create data, merge data with inner join, drop rows with non food names, group by, pivot table, missing values, normalize, WCSS to find k value, instantiate and fit model, cluster profile breakdown

aphd87/Advanced-linear-regression

Pickle file, shuffle data, missing values, drop values, outliers, box plot, split, categorical variables and one hot encoder, feature selection and RFECV, feature selection, optimal number of features plot, r-square, cross validation scores, adjusted r-square, model coefficients table and intercept, final interpretation

aphd87/Artificial-neural-network

Artificial neural network, feature scaling, network architecture, compile, summary, training network, plot loss accuracy, make predictions on new data

aphd87/Bank-Churn-Project

test file, train file, submission file, describe with highlights, outliers, missing values, plots, XGB, RF, GB, CatBoost, column transformers, NN, test data flattened, processed into submission file

aphd87/Basic-regression

split, instantiate, train, r2 score for accuracy

aphd87/Causal-Impact-Analysis

Import data tables, aggregate by group by, merge, pivot table, frequency, change column order, rename columns, apply causal impact, print out casual impact report

aphd87/Classification-Tree

Classification tree, split, instantiate, train, assess model accuracy, plot decision tree

aphd87/Classification-Tree-Advanced

Classification tree, shuffle data, class balance, missing values, model shape, categorical variables, model training, decision tree plotmax_depth, max_accuracymodel assessment, y_probabilities, confusion_matrix, accuracy_score, recall, F1 score, precision,

aphd87/CNN-Fruit-Project-Keras-Tuner

Architecture tuning, random search, tuner search, tuner results, hyperparameters, best models, sequential, plots, accuracy, class probabilities, matrix, accuracy scores

aphd87/Creating-our-regression

Combine excel databases and sheets, merge files, left join, groupby and aggregate new variables, inner join, pickle dump files

aphd87/dsi-streamlit-web-app

aphd87/Fruit-classification-augmentation

Data augmentation to improve accuracy, shape of class probs, class probs, predicted class, predicted label, predictions, test accuracy, confusion matrix

aphd87/Fruit-classification-dropout

Training and validating data, network architecture with layers added, training parameters, plot learning, predict class prob, class index, label, accuracy, and confusion matrix

aphd87/Grid-Search

Grid Search, hyperparameter optimization, get best cv score, best parameters, best estimator, create optimal model object

aphd87/Image-search-engine

Vgg16 architecture, pre-trained model, model summary, preprocess image, featurise image, pass in images, pickle dump, load file, search results, cosine scores, image distances, image indices, print out

aphd87/K-Means-Clustering

Plotting cluster, instantiate cluster, adding labels to cluster, value counts, cluster centroid locations, plotting clusters and centroids

aphd87/KNN-Basic-Classification

Train test split, instantiate, basic cluster analysis and accuracy score

aphd87/KNN-cluster-analysis-

Data model, shuffle data, class balance, missing values, model shape, outliers, splits, categorical variables, feature scaling, feature selection, confusion matrix, F1, accuracy, precision, recall, optimal value of k,

aphd87/Logistic-Regression-Advanced

Pickle upload, drop data, shuffle, class balance, value counts, missing values, shape, outliers, categorical, feature selection, model training, data probabilities, probability of class, confusion matrix, acc, precision, recall, f1, ROC AUC, optimal threshold, plotting

aphd87/Logistic-Regression-Basic

Logistic regression, train test split, instantiate, train, assess with test, individual probabilities, plt seaborn, confusion matrix

aphd87/PCA

PCA, drop column, shuffle data, class balance, missing values, shape, split variables, feature scaling, instantiate PCA, explained variance, plot variance explained, plot cumulative variance, accuracy score

aphd87/Pipelines

Pipelines, specify numeric and categorical features, transform numeric and categorical features, preprocess, logistic regression, random forest, save pipeline, import pipeline and predict on new data, pass new data and receive predictions

aphd87/Predicting-Missing-Loyalty-Scores

Pickle load, drop columns, drop missing values, apply one hot encoding, concatenate, make loyalty customer predictions

aphd87/Random-Forest-Classification

Random forest, train test split, instantiate, train, accuracy, notes around parameters we can use

aphd87/Random-Forest-Classification-Advanced

Pickle file, drop column, shuffle data, class balance, value counts, dealing with missing values, shape, split, dealing with categorical variables, model training, model assessment, class probabilities, confusion matrix, accuracy, precision, recall, f1 score, feature importance, permutation importance,

aphd87/Random_Forest_Regression_Advanced

Pickle file, data drop, shuffle data, missing values, drop data, outlier investigation, split, one hot encoding, R2, cross validation, adjusted R2, feature importance, permutation importance, exam one customer observation, predictions under hood

aphd87/Regression-Tree-Advanced

Pickle load, column drop, missing values, outliers, test train split, one hot encoding, R square, cross validation, cross validation, overfitting, max_depth, accuracy score, optimal score, decision tree

aphd87/streamlit

2019 Deloitte data set with Streamlit web app

aphd87/Streamlit-2024-Project

Streamlit dashboard that uses Deloitte media survey data to help predict who will have a streaming service subscription.

aphd87/Transferred-learning-via-vgg-frozen-layers-accuracy-matrix-and-matrix-percentages

Transferred learning with CNN