yashwants19 / v1 Goto Github PK

package main

import (
  "github.com/mlpack.org/v1/mlpack"
  "fmt"
)
func main() {

  // Download dataset.
  mlpack.DownloadFile("https://www.mlpack.org/datasets/covertype-small.data.csv.gz",
                      "data.csv.gz")
  mlpack.DownloadFile("https://www.mlpack.org/datasets/covertype-small.labels.csv.gz",
                      "labels.csv.gz")

  // Extract/Unzip the dataset.
  mlpack.UnZip("data.csv.gz", "data.csv")
  dataset, _ := mlpack.Load("data.csv")

  mlpack.UnZip("labels.csv.gz", "labels.csv")
  labels, _ := mlpack.Load("labels.csv")

  // Split the dataset using mlpack.
  params := mlpack.PreprocessSplitOptions()
  params.InputLabels = labels
  params.TestRatio = 0.3
  params.Verbose = true
  test, test_labels, train, train_labels :=
      mlpack.PreprocessSplit(dataset, params)

  // Train a random forest.
  rf_params := mlpack.RandomForestOptions()
  rf_params.NumTrees = 10
  rf_params.MinimumLeafSize = 3
  rf_params.PrintTrainingAccuracy = true
  rf_params.Training = train
  rf_params.Labels = train_labels
  rf_params.Verbose = true
  rf_model, _, _ := mlpack.RandomForest(rf_params)

  // Predict the labels of the test points.
  rf_params_2 := mlpack.RandomForestOptions()
  rf_params_2.Test = test
  rf_params_2.InputModel = &rf_model
  rf_params_2.Verbose = true
  _, predictions, _ := mlpack.RandomForest(rf_params_2)

  // Now print the accuracy.
  rows, _ := predictions.Dims()
  var sum int = 0
  for i := 0; i < rows; i++ {
    if (predictions.At(i, 0) == test_labels.At(i, 0)) {
      sum = sum + 1
    }
  }
  fmt.Print(sum, " correct out of ", rows, " (",
      (float64(sum) / float64(rows)) * 100, "%).\n")
}

package main

import (
  "github.com/frictionlessdata/tableschema-go/csv"
  "github.com/mlpack.org/v1/mlpack"
  "gonum.org/v1/gonum/mat"
  "fmt"
)
func main() {

  // Download dataset.
  mlpack.DownloadFile("https://www.mlpack.org/datasets/ml-20m/ratings-only.csv.gz",
                      "ratings-only.csv.gz")
  mlpack.DownloadFile("https://www.mlpack.org/datasets/ml-20m/movies.csv.gz",
                      "movies.csv.gz")

  // Extract dataset.
  mlpack.UnZip("ratings-only.csv.gz", "ratings-only.csv")
  ratings, _ := mlpack.Load("ratings-only.csv")

  mlpack.UnZip("movies.csv.gz", "movies.csv")
  table, _ := csv.NewTable(csv.FromFile("movies.csv"), csv.LoadHeaders())
  movies, _ := table.ReadColumn("title")

  // Split the dataset using mlpack.
  params := mlpack.PreprocessSplitOptions()
  params.TestRatio = 0.1
  params.Verbose = true
  ratings_test, _, ratings_train, _ := mlpack.PreprocessSplit(ratings, params)

  // Train the model.  Change the rank to increase/decrease the complexity of the
  // model.
  cf_params := mlpack.CfOptions()
  cf_params.Training = ratings_train
  cf_params.Test = ratings_test
  cf_params.Rank = 10
  cf_params.Verbose = true
  cf_params.Algorithm = "RegSVD"
  _, cf_model := mlpack.Cf(cf_params)

  // Now query the 5 top movies for user 1.
  cf_params_2 := mlpack.CfOptions()
  cf_params_2.InputModel = &cf_model
  cf_params_2.Recommendations = 10
  cf_params_2.Query = mat.NewDense(1, 1, []float64{1})
  cf_params_2.Verbose = true
  cf_params_2.MaxIterations = 10
  output, _ := mlpack.Cf(cf_params_2)

  // Get the names of the movies for user 1.
  fmt.Println("Recommendations for user 1")
  for i := 0; i < 10; i++ {
    fmt.Println(i, ":", movies[int(output.At(0 , i))])
  }
}