{"metadata":{"kernelspec":{"name":"ir","display_name":"R","language":"R"},"language_info":{"name":"R","codemirror_mode":"r","pygments_lexer":"r","mimetype":"text/x-r-source","file_extension":".r","version":"4.0.5"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"# G2Net Gravitational Wave Detection in R\n\n> #### All programming by John Akwei, ECMp ERMp Data Scientist\n> #### https://contextbase.github.io","metadata":{}},{"cell_type":"markdown","source":"# Section 1 - Required Libraries","metadata":{}},{"cell_type":"code","source":"suppressMessages({library(tidyverse)\n                  library(tidyr)\n                  library(readr)\n                  library(data.table)\n                  library(magrittr)\n                  library(dplyr)\n                  library(caret)\n                  library(reshape2)\n                  library(kernlab)\n                  library(corrplot)\n                  library(ggplot2)})","metadata":{"execution":{"iopub.status.busy":"2022-06-18T22:28:38.482336Z","iopub.execute_input":"2022-06-18T22:28:38.520043Z","iopub.status.idle":"2022-06-18T22:28:38.56258Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Section 2 - Import Datasets","metadata":{}},{"cell_type":"code","source":"train_labels <- fread(\"../input/g2net-gravitational-wave-detection/training_labels.csv\")\nsample_submission <- fread(\"../input/g2net-gravitational-wave-detection/sample_submission.csv\")\n\ntrain_labels <- train_labels[1:56000,]","metadata":{"execution":{"iopub.status.busy":"2022-06-18T22:29:43.643333Z","iopub.execute_input":"2022-06-18T22:29:43.645174Z","iopub.status.idle":"2022-06-18T22:29:45.271577Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Section 3 - Exploratory Data Analysis","metadata":{}},{"cell_type":"code","source":"cat(\"First Five Rows of Train Labels Data\")\nhead(train_labels)\n\ncat(\"Dimensions of Train Labels Data\")\ndim(train_labels)\n\ncat(\"First Five Rows of Sample Submission Data\")\nhead(sample_submission)\n\ncat(\"Dimensions of Sample Submission Data\")\ndim(sample_submission)\n\ncat(\"Plot of First 500 Target Values\")\nplot(train_labels$target[1:500], type = \"l\")","metadata":{"execution":{"iopub.status.busy":"2022-06-18T22:30:05.944426Z","iopub.execute_input":"2022-06-18T22:30:05.946303Z","iopub.status.idle":"2022-06-18T22:30:06.022429Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Section 4 - CrossValidation","metadata":{}},{"cell_type":"code","source":"set.seed(123)\ndataSubset <- createDataPartition(train_labels$target, p=0.8, list=F)\ntrain  <- train_labels[dataSubset,]\ntest <- train_labels[-dataSubset,]\n\ncat(\"Dimensions of the train dataset\")\ndim(train)\n\ncat(\"Dimensions of the test dataset\")\ndim(test)\n\ncat('It is ', nrow(train)+nrow(test)==nrow(train_labels),\n    ' that the crossValidated Train and Test datasets are an accurate partition of the source dataset',\n    '.', sep='')","metadata":{"execution":{"iopub.status.busy":"2022-06-18T22:30:41.51989Z","iopub.execute_input":"2022-06-18T22:30:41.521674Z","iopub.status.idle":"2022-06-18T22:30:41.636421Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Section 5 - Predictive Analytics","metadata":{}},{"cell_type":"code","source":"train$id <- as.numeric(as.factor(train$id))\ntest$id <- as.numeric(as.factor(test$id))\n\nmodel_SVM <- ksvm(target ~ id,\n                  type = 'nu-svr', kernel =\"anovadot\", train)\npred_SVM <- predict(model_SVM, newdata = test)\n\ncat(\"Parameters of model_SVM:\")\nmodel_SVM\n\n# R^2 accuracy\nR2 <- function(y_actual,y_predict){cor(y_actual,y_predict)^2}\n\ncat(\"The R2 accuracy of SVM:\")\nrsquared <- R2(test$target, pred_SVM)\nas.numeric(rsquared)","metadata":{"execution":{"iopub.status.busy":"2022-06-18T22:37:20.384545Z","iopub.execute_input":"2022-06-18T22:37:20.38623Z","iopub.status.idle":"2022-06-18T22:41:16.327445Z"},"trusted":true},"execution_count":null,"outputs":[]}]}