knn_cv 0.1.2 → 0.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 82ad55344b760f5b5fc48453c165ac225f943b5d
-  data.tar.gz: f95e098c3a20dae28375f3e61df444aa054daf33
+  metadata.gz: ae7308b1dc9b6844bfe35f260b0112db5200a1cd
+  data.tar.gz: 6d299e33cd783d6ef734728e50c440d7ab108536
 SHA512:
-  metadata.gz: aeeba2a91ea65ec5ce46c68c6050ccaf67579b0ae80428b4b5484c7df2b0d035e76c6246b08ec3758babd4c48df41462c133faaaf231303ef8d6ce8579d38eb7
-  data.tar.gz: ea440252cf2b59f64671ffff77bd48df5544d66c4b691d77c5720ee84664e1e6f2ce5ad39ad3a8b4161dc4f49b7987199ef094f9b714c46d3c0ce1b384355c6c
+  metadata.gz: b716e72c5029a23262b6c41c6e0bafa5fcf53357bbb89876a7e181e13200b61174ee705f64a84399eb94895251f34af46ae5e655b68e89aba0dbbb7ed6ae29c6
+  data.tar.gz: da2335e2a1d07010ef7a0827d3078314925569c7bdc62a5e6e82297f504edcba7d5f1fd1ef1821d64720a797fca7ba362ead7a80f35f9e114d6bc7873f088faf

data/ext/c_knn/knn.c

@@ -184,55 +184,59 @@ VALUE method_c_knn_leaveoneout(VALUE self, VALUE rb_features) {
   return rb_float_new(fitness);
 }
 
-VALUE method_c_knn_initialize(VALUE self, VALUE rb_k, VALUE rb_dataset, VALUE rb_random_par) {
-  int ncol, nrow;
+VALUE method_c_knn_initialize(VALUE self, VALUE rb_k, VALUE data, VALUE rb_class, VALUE rb_numeric, VALUE rb_random_par) {
+  long int ncol, nrow;
 
   double * instances = NULL;
   int * classes = NULL;
   int * which_numeric = NULL;
 
-  VALUE data = rb_funcall(rb_dataset, rb_intern("instances"), 0);
-  VALUE rb_class = rb_funcall(rb_dataset, rb_intern("classes"), 0);
-  VALUE rb_numeric = rb_funcall(rb_dataset, rb_intern("numeric_attrs"), 0);
+  // VALUE data = rb_funcall(rb_dataset, rb_intern("instances"), 0);
+  // VALUE rb_class = rb_funcall(rb_dataset, rb_intern("classes"), 0);
+  // VALUE rb_numeric = rb_funcall(rb_dataset, rb_intern("numeric_attrs"), 0);
 
   // Define global variables
   rb_iv_set(self, "@num_neighbors", rb_k);
   nrow = RARRAY_LEN(data);
-  rb_iv_set(self, "@nrow", INT2NUM(nrow));
-  ncol = RARRAY_LEN(rb_ary_entry(data, 0));
-  rb_iv_set(self, "@ncol", INT2NUM(ncol));
-  rb_iv_set(self, "@nclass", rb_funcall(rb_dataset, rb_intern("class_count"), 0));
-  FLOAT_MAX = NUM2DBL(rb_intern("Float::MAX"));
-  rb_iv_set(self, "@rng", rb_random_par);
+  if (nrow > 0) {
+    rb_iv_set(self, "@nrow", INT2NUM(nrow));
+    ncol = RARRAY_LEN(rb_ary_entry(data, 0));
+    rb_iv_set(self, "@ncol", INT2NUM(ncol));
+    rb_iv_set(self, "@nclass", rb_funcall(rb_funcall(rb_class, rb_intern("uniq"), 0), rb_intern("length"), 0));
+    FLOAT_MAX = NUM2DBL(rb_intern("Float::MAX"));
+    rb_iv_set(self, "@rng", rb_random_par);
+
+    instances = (double*) malloc(sizeof(double) * nrow * ncol);
+
+    int i, j;
+    for (i = 0; i < nrow; i++) {
+      for (j = 0; j < ncol; j++) {
+        if (TYPE(rb_ary_entry(rb_ary_entry(data, i), j)) == T_STRING) {
+          rb_raise(rb_eArgError, "A string was found within the dataset. Aborting...");
+        } else
+        instances[i * ncol + j] = NUM2DBL(rb_ary_entry(rb_ary_entry(data, i), j));
+      }
+    }
 
-  instances = (double*) malloc(sizeof(double) * nrow * ncol);
+    classes = (int*) malloc(sizeof(int) * nrow);
 
-  int i, j;
-  for (i = 0; i < nrow; i++) {
-    for (j = 0; j < ncol; j++) {
-      if (TYPE(rb_ary_entry(rb_ary_entry(data, i), j)) == T_STRING) {
-        rb_raise(rb_eStandardError, "A string was found within the dataset. Aborting...");
-      } else
-      instances[i * ncol + j] = NUM2DBL(rb_ary_entry(rb_ary_entry(data, i), j));
+    for (i = 0; i < nrow; i++) {
+      classes[i] = NUM2INT(rb_ary_entry(rb_class, i));
     }
-  }
 
-  classes = (int*) malloc(sizeof(int) * nrow);
+    which_numeric = (int*) malloc(sizeof(int) * ncol);
 
-  for (i = 0; i < nrow; i++) {
-    classes[i] = NUM2INT(rb_ary_entry(rb_class, i));
-  }
-
-  which_numeric = (int*) malloc(sizeof(int) * ncol);
+    for (j = 0; j < ncol; j++) {
+      which_numeric[j] = NUM2INT(rb_ary_entry(rb_numeric, j));
+    }
 
-  for (j = 0; j < ncol; j++) {
-    which_numeric[j] = NUM2INT(rb_ary_entry(rb_numeric, j));
+    rb_iv_set(self, "@instances", Data_Wrap_Struct(C_instances, NULL, c_knn_free, instances));
+    rb_iv_set(self, "@classes", Data_Wrap_Struct(C_classes, NULL, c_knn_free, classes));
+    rb_iv_set(self, "@which_numeric", Data_Wrap_Struct(C_numerics, NULL, c_knn_free, which_numeric));
+  } else {
+    rb_raise(rb_eArgError, "Attempted to create a classifier for an empty dataset. Aborting...");
   }
 
-  rb_iv_set(self, "@instances", Data_Wrap_Struct(C_instances, NULL, c_knn_free, instances));
-  rb_iv_set(self, "@classes", Data_Wrap_Struct(C_classes, NULL, c_knn_free, classes));
-  rb_iv_set(self, "@which_numeric", Data_Wrap_Struct(C_numerics, NULL, c_knn_free, which_numeric));
-
   return self;
 }
 
@@ -245,6 +249,6 @@ void Init_c_knn(void) {
   C_classes = rb_define_class_under(Classifier, "Classes", rb_cObject);
   C_numerics = rb_define_class_under(Classifier, "Numerics", rb_cObject);
 
-  rb_define_method(Classifier, "initialize", method_c_knn_initialize, 3);
+  rb_define_method(Classifier, "initialize", method_c_knn_initialize, 5);
   rb_define_method(Classifier, "fitness_for", method_c_knn_leaveoneout, 1);
 }

data/lib/knn_cv/version.rb

@@ -1,3 +1,3 @@
 module KnnCv
-  VERSION = "0.1.2"
+  VERSION = "0.2"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: knn_cv
 version: !ruby/object:Gem::Version
-  version: 0.1.2
+  version: '0.2'
 platform: ruby
 authors:
 - David Charte
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2016-04-07 00:00:00.000000000 Z
+date: 2016-05-01 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler