features checking

.gitignore

@@ -1,2 +1,3 @@
 data/
-.DS_Store
+.DS_Store
+.Rproj.user

Electrocardiogram.Rproj

@@ -0,0 +1,13 @@
+Version: 1.0
+
+RestoreWorkspace: Default
+SaveWorkspace: Default
+AlwaysSaveHistory: Default
+
+EnableCodeIndexing: Yes
+UseSpacesForTab: Yes
+NumSpacesForTab: 2
+Encoding: UTF-8
+
+RnwWeave: Sweave
+LaTeX: pdfLaTeX

feature_selection_results.csv

@@ -0,0 +1,32 @@
+"Num_Features","Features","Accuracy","F1_class1","F2_class1","Precision_class1","Recall_class1"
+1,"freq_median_freq",0.9773,0.4111,0.3103,0.8966,0.2667
+2,"freq_median_freq, time_kurtosis",0.9806,0.5544,0.454,0.8778,0.4051
+3,"freq_median_freq, time_kurtosis, freq_peak_freq",0.9812,0.5859,0.4932,0.8529,0.4462
+4,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay",0.9808,0.5743,0.4824,0.8416,0.4359
+5,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp",0.9844,0.6928,0.627,0.8394,0.5897
+6,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid",0.9863,0.7305,0.6638,0.8777,0.6256
+7,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp",0.9847,0.7024,0.6406,0.8369,0.6051
+8,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95",0.9855,0.7164,0.6522,0.8571,0.6154
+9,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs",0.985,0.7118,0.6541,0.8345,0.6205
+10,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std",0.986,0.7278,0.6663,0.8601,0.6308
+11,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio",0.9863,0.7353,0.6757,0.8621,0.641
+12,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio, time_min",0.9852,0.7104,0.6467,0.85,0.6103
+13,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio, time_min, time_skew",0.9867,0.7449,0.6857,0.8699,0.6513
+14,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio, time_min, time_skew, acf_integral_time",0.986,0.7278,0.6663,0.8601,0.6308
+15,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio, time_min, time_skew, acf_integral_time, ar_r2",0.9875,0.7602,0.7012,0.8844,0.6667
+16,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio, time_min, time_skew, acf_integral_time, ar_r2, time_var",0.987,0.7493,0.6872,0.8819,0.6513
+17,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio, time_min, time_skew, acf_integral_time, ar_r2, time_var, acf_val_dom",0.986,0.7294,0.6703,0.8552,0.6359
+18,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio, time_min, time_skew, acf_integral_time, ar_r2, time_var, acf_val_dom, ar_resid_var",0.9864,0.739,0.6803,0.863,0.6462
+19,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio, time_min, time_skew, acf_integral_time, ar_r2, time_var, acf_val_dom, ar_resid_var, time_iqr",0.9855,0.7181,0.6562,0.8521,0.6205
+20,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio, time_min, time_skew, acf_integral_time, ar_r2, time_var, acf_val_dom, ar_resid_var, time_iqr, freq_bandwidth",0.9858,0.7257,0.6656,0.8542,0.6308
+21,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio, time_min, time_skew, acf_integral_time, ar_r2, time_var, acf_val_dom, ar_resid_var, time_iqr, freq_bandwidth, time_max",0.9861,0.7362,0.6828,0.8467,0.6513
+22,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio, time_min, time_skew, acf_integral_time, ar_r2, time_var, acf_val_dom, ar_resid_var, time_iqr, freq_bandwidth, time_max, freq_entropy",0.9863,0.7368,0.6796,0.8571,0.6462
+23,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio, time_min, time_skew, acf_integral_time, ar_r2, time_var, acf_val_dom, ar_resid_var, time_iqr, freq_bandwidth, time_max, freq_entropy, time_p5",0.986,0.7278,0.6663,0.8601,0.6308
+24,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio, time_min, time_skew, acf_integral_time, ar_r2, time_var, acf_val_dom, ar_resid_var, time_iqr, freq_bandwidth, time_max, freq_entropy, time_p5, ar1_coef",0.9864,0.739,0.6803,0.863,0.6462
+25,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio, time_min, time_skew, acf_integral_time, ar_r2, time_var, acf_val_dom, ar_resid_var, time_iqr, freq_bandwidth, time_max, freq_entropy, time_p5, ar1_coef, time_p95",0.9857,0.7219,0.6609,0.8531,0.6256
+26,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio, time_min, time_skew, acf_integral_time, ar_r2, time_var, acf_val_dom, ar_resid_var, time_iqr, freq_bandwidth, time_max, freq_entropy, time_p5, ar1_coef, time_p95, time_mean",0.9857,0.7251,0.6688,0.8435,0.6359
+27,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio, time_min, time_skew, acf_integral_time, ar_r2, time_var, acf_val_dom, ar_resid_var, time_iqr, freq_bandwidth, time_max, freq_entropy, time_p5, ar1_coef, time_p95, time_mean, time_p25",0.9861,0.7347,0.6789,0.8514,0.6462
+28,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio, time_min, time_skew, acf_integral_time, ar_r2, time_var, acf_val_dom, ar_resid_var, time_iqr, freq_bandwidth, time_max, freq_entropy, time_p5, ar1_coef, time_p95, time_mean, time_p25, freq_top1_freq",0.9866,0.7381,0.6732,0.8794,0.6359
+29,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio, time_min, time_skew, acf_integral_time, ar_r2, time_var, acf_val_dom, ar_resid_var, time_iqr, freq_bandwidth, time_max, freq_entropy, time_p5, ar1_coef, time_p95, time_mean, time_p25, freq_top1_freq, time_median",0.9855,0.7214,0.6641,0.8425,0.6308
+30,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio, time_min, time_skew, acf_integral_time, ar_r2, time_var, acf_val_dom, ar_resid_var, time_iqr, freq_bandwidth, time_max, freq_entropy, time_p5, ar1_coef, time_p95, time_mean, time_p25, freq_top1_freq, time_median, time_p75",0.9861,0.7284,0.663,0.8714,0.6256
+31,"freq_median_freq, time_kurtosis, freq_peak_freq, acf_time_decay, time_ptp, freq_centroid, freq_top1_amp, freq_rolloff_95, acf_mean_abs, time_std, time_range_ratio, time_min, time_skew, acf_integral_time, ar_r2, time_var, acf_val_dom, ar_resid_var, time_iqr, freq_bandwidth, time_max, freq_entropy, time_p5, ar1_coef, time_p95, time_mean, time_p25, freq_top1_freq, time_median, time_p75, time_rms",0.9869,0.7485,0.6904,0.8707,0.6564

main.R

@@ -0,0 +1,222 @@
+#install.packages("lightgbm", repos = "https://cran.r-project.org")
+#install.packages("MLmetrics")
+
+library(lightgbm)
+library(MLmetrics)
+
+# 1. Load your data
+df <- read.csv("./data/Ketamine_icp_no_x.csv")
+
+# --- 2. Data Preparation ---
+target_name <- "label" 
+target_index <- which(names(df) == target_name)
+
+# Prepare target variable
+if (is.factor(df[, target_index])) {
+  y <- as.numeric(df[, target_index]) - 1
+} else {
+  y <- df[, target_index]
+}
+
+# Create the data matrix for features
+X <- as.matrix(df[, -target_index]) 
+
+# --- 3. Split Data into Training and Testing Sets ---
+set.seed(42) 
+train_index <- sample(nrow(X), size = 0.8 * nrow(X)) 
+
+X_train <- X[train_index, ]
+X_test <- X[-train_index, ]
+y_train <- y[train_index]
+y_test <- y[-train_index]
+
+# --- 4. Get Feature Importance from Full Model ---
+lgb_train_full <- lgb.Dataset(data = X_train, label = y_train)
+
+params <- list(
+  objective = "binary",
+  metric = "binary_logloss",
+  boosting_type = "gbdt",
+  num_leaves = 20,
+  learning_rate = 0.05,
+  feature_fraction = 0.8
+)
+
+bst_full <- lgb.train(
+  params = params,
+  data = lgb_train_full,
+  nrounds = 100,
+  verbose = -1
+)
+
+# Get feature importance
+importance <- lgb.importance(bst_full)
+num_features <- nrow(importance)
+
+# Create a data frame to store results
+results_df <- data.frame(
+  Num_Features = integer(),
+  Features = character(),
+  Accuracy = numeric(),
+  F1_class1 = numeric(),
+  F2_class1 = numeric(),
+  Precision_class1 = numeric(),
+  Recall_class1 = numeric()
+)
+
+# --- 5. Loop through different numbers of top features ---
+cat("Training models with different numbers of top features...\n")
+cat("=====================================================\n")
+
+for (i in 1:num_features) {
+  cat(paste("Training model with top", i, "features...\n"))
+  
+  # Select top i features
+  top_features <- importance$Feature[1:i]
+  
+  # Subset training and test data
+  X_train_sub <- X_train[, top_features, drop = FALSE]
+  X_test_sub <- X_test[, top_features, drop = FALSE]
+  
+  # Create LightGBM dataset
+  lgb_train_sub <- lgb.Dataset(data = X_train_sub, label = y_train)
+  
+  # Train model with subset of features
+  bst_sub <- lgb.train(
+    params = params,
+    data = lgb_train_sub,
+    nrounds = 100,
+    verbose = -1
+  )
+  
+  # Make predictions
+  pred_prob_sub <- predict(bst_sub, X_test_sub)
+  pred_class_sub <- as.numeric(pred_prob_sub > 0.5)
+  
+  # Calculate metrics
+  accuracy <- mean(pred_class_sub == y_test)
+  
+  # For binary classification
+  if (length(unique(y_test)) == 2) {
+    # F1 score for class 1
+    f1 <- F1_Score(y_true = y_test, y_pred = pred_class_sub, positive = 1)
+    
+    # Precision and Recall for class 1
+    precision <- Precision(y_true = y_test, y_pred = pred_class_sub, positive = 1)
+    recall <- Recall(y_true = y_test, y_pred = pred_class_sub, positive = 1)
+    
+    # F2-score (beta = 2)
+    beta <- 2
+    f2 <- (1 + beta^2) * (precision * recall) / (beta^2 * precision + recall)
+    
+    # Handle cases where precision or recall might be NaN
+    if (is.na(f2)) {
+      f2 <- 0
+    }
+  } else {
+    # For multi-class classification
+    f1 <- NA
+    precision <- NA
+    recall <- NA
+    f2 <- NA
+  }
+  
+  # Store results
+  results_df <- rbind(results_df, data.frame(
+    Num_Features = i,
+    Features = paste(top_features, collapse = ", "),
+    Accuracy = round(accuracy, 4),
+    F1_class1 = round(f1, 4),
+    F2_class1 = round(f2, 4),
+    Precision_class1 = round(precision, 4),
+    Recall_class1 = round(recall, 4)
+  ))
+  
+  # Print progress
+  cat(paste("  Accuracy:", round(accuracy, 4), 
+            "| F1:", round(f1, 4),
+            "| F2:", round(f2, 4),
+            "| Precision:", round(precision, 4),
+            "| Recall:", round(recall, 4), "\n"))
+}
+
+cat("=====================================================\n")
+
+# --- 6. Display Results ---
+cat("\nSummary of Results:\n")
+cat("===================\n")
+print(results_df)
+
+# Find best performing models based on different metrics
+cat("\nBest Performing Models:\n")
+cat("=======================\n")
+
+# Best by F1 score
+if (!all(is.na(results_df$F1_class1))) {
+  best_f1_idx <- which.max(results_df$F1_class1)
+  cat(paste("Best F1-score (", results_df$F1_class1[best_f1_idx], 
+            ") with", results_df$Num_Features[best_f1_idx], "features\n"))
+}
+
+# Best by F2 score
+if (!all(is.na(results_df$F2_class1))) {
+  best_f2_idx <- which.max(results_df$F2_class1)
+  cat(paste("Best F2-score (", results_df$F2_class1[best_f2_idx], 
+            ") with", results_df$Num_Features[best_f2_idx], "features\n"))
+}
+
+# Best by Accuracy
+best_acc_idx <- which.max(results_df$Accuracy)
+cat(paste("Best Accuracy (", results_df$Accuracy[best_acc_idx], 
+          ") with", results_df$Num_Features[best_acc_idx], "features\n"))
+
+# --- 7. Optional: Plot metrics vs number of features ---
+if (require(ggplot2)) {
+  library(ggplot2)
+  
+  # Plot F1 and F2 scores
+  p1 <- ggplot(results_df, aes(x = Num_Features)) +
+    geom_line(aes(y = F1_class1, color = "F1 Score"), size = 1) +
+    geom_line(aes(y = F2_class1, color = "F2 Score"), size = 1) +
+    geom_point(aes(y = F1_class1, color = "F1 Score"), size = 2) +
+    geom_point(aes(y = F2_class1, color = "F2 Score"), size = 2) +
+    labs(title = "F1 and F2 Scores vs Number of Features",
+         x = "Number of Top Features",
+         y = "Score Value") +
+    theme_minimal() +
+    scale_color_manual(values = c("F1 Score" = "blue", "F2 Score" = "red"))
+  
+  # Plot Accuracy
+  p2 <- ggplot(results_df, aes(x = Num_Features, y = Accuracy)) +
+    geom_line(color = "darkgreen", size = 1) +
+    geom_point(color = "darkgreen", size = 2) +
+    labs(title = "Accuracy vs Number of Features",
+         x = "Number of Top Features",
+         y = "Accuracy") +
+    theme_minimal()
+  
+  # Plot Precision and Recall
+  p3 <- ggplot(results_df, aes(x = Num_Features)) +
+    geom_line(aes(y = Precision_class1, color = "Precision"), size = 1) +
+    geom_line(aes(y = Recall_class1, color = "Recall"), size = 1) +
+    geom_point(aes(y = Precision_class1, color = "Precision"), size = 2) +
+    geom_point(aes(y = Recall_class1, color = "Recall"), size = 2) +
+    labs(title = "Precision and Recall (Class 1) vs Number of Features",
+         x = "Number of Top Features",
+         y = "Score Value") +
+    theme_minimal() +
+    scale_color_manual(values = c("Precision" = "purple", "Recall" = "orange"))
+  
+  # Display plots
+  print(p1)
+  print(p2)
+  print(p3)
+}
+
+# --- 8. Save results to CSV ---
+write.csv(results_df, "feature_selection_results.csv", row.names = FALSE)
+cat("\nResults saved to 'feature_selection_results.csv'\n")
+
+# --- 9. Display top 20 feature importance plot ---
+lgb.plot.importance(importance, top_n = min(20, num_features))
+