Archive unwanted real data analyses

Author: ywwry66

real_data.org

@@ -115,129 +115,6 @@ $p=13$, $n=303$. Repeat $m=100,300$ times for average test error.
 |  0.7 | 17.33 (0.17) | 17.22 (0.18) | 20.26 (0.21) | 17.04 (0.18) | 16.90 (0.19) | 17.42 (0.18) |
 |  0.8 | 17.66 (0.23) | 17.54 (0.24) | 19.98 (0.26) | 17.29 (0.24) | 17.02 (0.24) | 17.46 (0.26) |
 
-* Accent Recognition
-
-#+name: accent recognition setup
-#+begin_src R
-  seed <- 2020
-  m <- 100
-  per <- c(0.4, 0.5, 0.6, 0.7, 0.8)
-  ar <- read.csv("real_data/accent-mfcc-data-1.csv")
-  ar <- ar[c(which(ar[, 1] == "GE"), which(ar[, 1] == "FR")), ]
-  y <- ar[, 1]
-  id0 <- which(y == "GE")
-  id1 <- which(y == "FR")
-  y[id0] <- 0
-  y[id1] <- 1
-  y <- as.numeric(y)
-  x <- data.matrix(ar[, 2:13])
-#+end_src
-
-#+call: data split()
-
-#+call: models evaluation(qda="FALSE",name="accent_rec")
-
-#+call: data summary[:results value replace :colnames yes](name="accent_rec")
-
-#+RESULTS:
-| pct. | DSDA         | LDA          | QDAP         | RDA          | SQDA         |
-|------+--------------+--------------+--------------+--------------+--------------|
-|  0.4 | 12.92 (0.80) | 14.58 (0.75) | 19.25 (1.11) | 12.58 (0.72) | 15.78 (0.82) |
-|  0.5 | 11.00 (0.70) | 11.10 (0.57) | 15.43 (1.14) | 10.03 (0.56) | 13.47 (0.63) |
-|  0.6 | 9.33 (0.59)  | 9.83 (0.50)  | 11.21 (0.69) | 9.17 (0.61)  | 12.67 (0.75) |
-|  0.7 | 8.61 (0.67)  | 7.83 (0.56)  | 7.94 (0.57)  | 8.56 (0.61)  | 12.06 (0.73) |
-|  0.8 | 7.42 (0.78)  | 7.50 (0.86)  | 7.42 (0.86)  | 6.92 (0.64)  | 10.00 (0.83) |
-
-* Parkinson Disease
-
-#+name: parkinson disease setup
-#+begin_src R
-  seed <- 2020
-  m <- 100
-  per <- c(0.4, 0.5, 0.6, 0.7, 0.8)
-  pd <- read.csv("real_data/ParkinsonDatabase.csv")
-  y <- pd[, 3]
-  id0 <- which(y == 0)
-  id1 <- which(y == 1)
-  x <- data.matrix(pd[, 5:48])
-  x[, 2] <- x[, 2] * 10000
-#+end_src
-
-#+call: data split()
-
-#+call: models evaluation(name="parkinson")
-
-#+call: data summary[:results value replace :colnames yes](name="parkinson")
-
-#+RESULTS:
-| pct. | DSDA         | LDA          | QDA          | QDAP         | RDA          | SQDA         |
-|------+--------------+--------------+--------------+--------------+--------------+--------------|
-|  0.4 | 22.15 (0.66) | 32.50 (1.29) | 39.58 (0.83) | 28.61 (1.04) | 21.18 (1.03) | 23.40 (1.36) |
-|  0.5 | 19.67 (1.25) | 27.92 (1.44) | 30.92 (1.18) | 25.42 (1.74) | 18.25 (0.92) | 26.50 (1.15) |
-|  0.6 | 20.10 (0.79) | 27.08 (1.39) | 27.40 (2.09) | 25.31 (1.60) | 19.48 (0.60) | 22.71 (0.90) |
-|  0.7 | 17.78 (1.07) | 22.50 (1.53) | 26.81 (1.66) | 20.14 (1.36) | 15.69 (1.49) | 21.81 (1.59) |
-|  0.8 | 21.88 (2.20) | 23.75 (2.02) | 24.17 (1.36) | 23.33 (1.88) | 20.21 (2.01) | 23.54 (1.39) |
-
-* Heart Failure Clinical Records
-
-#+name: heart failure setup
-#+begin_src R
-  seed <- 2020
-  m <- 100
-  per <- c(0.4, 0.5, 0.6, 0.7, 0.8)
-  hf <- read.csv("real_data/heart_failure_clinical_records_dataset.csv")
-  y <- hf[, 13]
-  id0 <- which(y == 0)
-  id1 <- which(y == 1)
-  x <- data.matrix(hf[, 1:12])
-#+end_src
-
-#+call: data split()
-
-#+call: models evaluation(name="heart_failure")
-
-#+call: data summary[:results value replace :colnames yes](name="heart_failure")
-
-#+RESULTS:
-| pct. | DSDA         | LDA          | QDA          | QDAP         | RDA          | SQDA         |
-|------+--------------+--------------+--------------+--------------+--------------+--------------|
-|  0.4 | 21.71 (0.58) | 19.34 (0.22) | 26.71 (0.26) | 19.73 (0.30) | 19.12 (0.27) | 18.39 (0.31) |
-|  0.5 | 22.06 (0.64) | 19.00 (0.29) | 25.26 (0.29) | 19.30 (0.32) | 18.57 (0.29) | 18.56 (0.33) |
-|  0.6 | 20.55 (0.56) | 18.39 (0.27) | 24.61 (0.28) | 19.63 (0.34) | 18.03 (0.28) | 18.03 (0.32) |
-|  0.7 | 18.71 (0.56) | 17.50 (0.37) | 25.11 (0.37) | 18.23 (0.39) | 16.92 (0.39) | 17.88 (0.44) |
-|  0.8 | 19.18 (0.60) | 17.65 (0.43) | 25.00 (0.47) | 18.33 (0.44) | 17.72 (0.47) | 17.88 (0.45) |
-
-* Breast Cancer Coimbra
-
-#+name: breast cancer coimbre setup
-#+begin_src R
-  seed <- 2020
-  m <- 100
-  per <- c(0.4, 0.5, 0.6, 0.7, 0.8)
-  bcc <- read.csv("real_data/Breast Cancer Coimbra.csv")
-  y <- bcc[, 10]
-  id0 <- which(y == 2)
-  id1 <- which(y == 1)
-  y[id0] <- 0
-  y[id1] <- 1
-  x <- data.matrix(bcc[, 1:9])
-#+end_src
-
-#+call: data split()
-
-#+call: models evaluation(name="breast_cancer_coimbra")
-
-#+call: data summary[:results value replace :colnames yes](name="breast_cancer_coimbra")
-
-#+RESULTS:
-| pct. | DSDA         | LDA          | QDA          | QDAP         | RDA          | SQDA         |
-|------+--------------+--------------+--------------+--------------+--------------+--------------|
-|  0.4 | 31.99 (0.53) | 31.29 (0.52) | 32.83 (0.47) | 35.71 (0.73) | 31.75 (0.57) | 36.09 (0.55) |
-|  0.5 | 32.14 (0.59) | 29.67 (0.54) | 31.34 (0.59) | 36.76 (0.79) | 31.33 (0.71) | 34.52 (0.65) |
-|  0.6 | 32.94 (0.69) | 30.19 (0.66) | 33.89 (0.60) | 38.79 (0.74) | 30.26 (0.62) | 35.53 (0.64) |
-|  0.7 | 31.20 (0.79) | 29.49 (0.66) | 32.29 (0.65) | 36.03 (0.97) | 29.37 (0.66) | 33.71 (0.70) |
-|  0.8 | 30.57 (1.06) | 29.43 (0.99) | 37.48 (0.83) | 42.13 (0.90) | 30.04 (0.93) | 36.57 (0.89) |
-
 * Various Code Blocks
 :PROPERTIES:
 :APPENDIX: t