extracts Weka Options helpers into own ns due to circ dep issue

project.clj

@@ -1,4 +1,4 @@
-(defproject com.leadtune/clj-ml "0.1.2"
+(defproject com.leadtune/clj-ml "0.1.3-SNAPSHOT"
   :description "Machine Learning library for Clojure built around Weka and friends"
   :repositories {"leadtune-repo" "http://c0026236.cdn1.cloudfiles.rackspacecloud.com/repo"}
   :java-source-path "src/java"
@@ -11,7 +11,8 @@
                  [incanter/incanter-charts "1.2.3"]
                  [lt/weka "3.6.3"]
                  [hr.irb/fastRandomForest "0.98"]]
-  :dev-dependencies [[autodoc "0.7.0"
+  :dev-dependencies [[midje "1.3-alpha1"]
+                     [autodoc "0.7.0"
                       :exclusions [org.clojure/clojure org.clojure/clojure-contrib
                                    ant/ant ant/ant-launcher]] ;; older ant breaks newer lein
                      [lein-javac "1.2.1-SNAPSHOT"

src/clj_ml/classifiers.clj

@@ -60,7 +60,7 @@
     (serialize-to-file *classifier*
      \"/Users/antonio.garrote/Desktop/classifier.bin\")
 "
-  (:use [clj-ml utils data kernel-functions])
+  (:use [clj-ml utils data kernel-functions options-utils])
   (:import (java.util Date Random)
            (hr.irb.fastRandomForest FastRandomForest)
            (weka.core Instance Instances)

src/clj_ml/clusterers.clj

@@ -13,7 +13,7 @@
    having the full data set in main memory. Functions for evaluating the clusterer
    as well as for clustering new instances are also supported
 "
-  (:use [clj-ml utils data distance-functions])
+  (:use [clj-ml utils data distance-functions options-utils])
   (:import (java.util Date Random)
            (weka.clusterers ClusterEvaluation SimpleKMeans Cobweb EM)))
 

src/clj_ml/distance_functions.clj

@@ -9,7 +9,7 @@
    classifiers and clusterers like K-Means.
 
    Euclidean, Manhattan and Chebysev distance functions are supported."
-  (:use [clj-ml utils data])
+  (:use [clj-ml utils data options-utils])
   (:import (weka.core EuclideanDistance ManhattanDistance ChebyshevDistance)))
 
 

src/clj_ml/filters.clj

@@ -36,7 +36,7 @@
    The previous sample of code could be rewritten with the make-apply-filter function:
 
      (def filtered-ds (make-apply-filter :remove-attributes {:attributes [:a :c]} ds))"
-  (:use [clj-ml data utils]
+  (:use [clj-ml data utils options-utils]
         [clojure.contrib [def :only [defvar defvar-]]])
   (:require [clojure.contrib [string :as str]])
   (:import (weka.filters Filter)
@@ -51,13 +51,6 @@
   "Creates the right parameters for a filter. Returns a clojure vector."
   (fn [kind map] kind))
 
-(defn- extract-attributes
-  "Transforms the :attributes value from m into the appropriate weka flag"
-  [m]
-  ["-R" (str/join ","
-                  (for [attr (:attributes m)]
-                    (inc (dataset-index-attr (:dataset-format m) attr))))])
-
 (declare make-apply-filter)
 ;TODO: consider passing in the make-filter-options body here as well in additon to the docstring.
 (defmacro deffilter

src/clj_ml/kernel_functions.clj

@@ -8,7 +8,7 @@
   "Kernel functions that can be passed as parameters to support vector machines classifiers.
 
    Polynomic, radial basis and string kernels are supported"
-  (:use [clj-ml utils data])
+  (:use [clj-ml utils data options-utils])
   (:import (weka.classifiers.functions.supportVector PolyKernel RBFKernel StringKernel)))
 
 (defmulti make-kernel-function-options

src/clj_ml/options_utils.clj

@@ -0,0 +1,68 @@
+;;
+;; Utilities for converting clojure hash maps into Weka string options
+;; @author Ben Mabey
+;;
+
+(ns #^{:author "Ben Mabey <ben@benmabey.com>"
+       :skip-wiki true}
+  clj-ml.options-utils
+  (:use [clj-ml data])
+  (:require [clojure.contrib [string :as str]]))
+
+;; Manipulation of array of options
+
+(defn check-option [opts val flag map]
+  "Sets an option for a filter"
+  (let [val-in-map (get map val)]
+    (if (nil? val-in-map)
+      opts
+      (conj opts flag))))
+
+(defn check-option-value [opts val flag map]
+  "Sets an option with value for a filter"
+  (let [val-in-map (get map val)]
+    (if (nil? val-in-map)
+      opts
+      (conj  (conj opts flag) (str val-in-map)))))
+
+
+(defn extract-attributes
+  "Transforms the :attributes value from m into the appropriate weka flag"
+  ([m] (extract-attributes "-R" m))
+  ([flag m] (extract-attributes flag :attributes m))
+  ([flag key-name m]
+     (if-let [attributes (key-name m)]
+       [flag (str/join ","
+                       (for [attr attributes]
+                         (inc (dataset-index-attr (:dataset-format m) attr))))]
+       [])))
+
+
+; TODO: Raise a helpful exception when the keys don't match up with the provided flags.
+(defn check-options
+  "Checks the presence of a set of options for a filter"
+  ([args-map opts-map] (check-options args-map opts-map []))
+  ( [args-map opts-map tmp]
+      (loop [rem (keys opts-map)
+             acum tmp]
+        (if (empty? rem)
+          acum
+          (let [k (first rem)
+                vk (get opts-map k)
+                rst (rest rem)]
+            (recur rst
+                   (check-option acum k vk args-map)))))))
+
+(defn check-option-values
+  "Checks the presence of a set of options with value for a filter"
+  ([args-map opts-map] (check-option-values args-map opts-map []))
+  ([args-map opts-map val]
+      (loop [rem (keys opts-map)
+             acum val]
+        (if (empty? rem)
+          acum
+          (let [k (first rem)
+                vk (get opts-map k)
+                rst (rest rem)]
+            (recur rst
+                   (check-option-value acum k vk args-map)))))))

src/clj_ml/utils.clj

@@ -67,51 +67,6 @@
       (catch NoSuchAlgorithmException e
         (throw (new RuntimeException e))))))
 
-;; Manipulation of array of options
-
-(defn check-option [opts val flag map]
-  "Sets an option for a filter"
-  (let [val-in-map (get map val)]
-    (if (nil? val-in-map)
-      opts
-      (conj opts flag))))
-
-(defn check-option-value [opts val flag map]
-  "Sets an option with value for a filter"
-  (let [val-in-map (get map val)]
-    (if (nil? val-in-map)
-      opts
-      (conj  (conj opts flag) (str val-in-map)))))
-
-; TODO: Raise a helpful exception when the keys don't match up with the provided flags.
-(defn check-options
-  "Checks the presence of a set of options for a filter"
-  ([args-map opts-map] (check-options args-map opts-map []))
-  ( [args-map opts-map tmp]
-      (loop [rem (keys opts-map)
-             acum tmp]
-        (if (empty? rem)
-          acum
-          (let [k (first rem)
-                vk (get opts-map k)
-                rst (rest rem)]
-            (recur rst
-                   (check-option acum k vk args-map)))))))
-
-(defn check-option-values
-  "Checks the presence of a set of options with value for a filter"
-  ([args-map opts-map] (check-option-values args-map opts-map []))
-  ([args-map opts-map val]
-      (loop [rem (keys opts-map)
-             acum val]
-        (if (empty? rem)
-          acum
-          (let [k (first rem)
-                vk (get opts-map k)
-                rst (rest rem)]
-            (recur rst
-                   (check-option-value acum k vk args-map)))))))
-
 ;; Serializing classifiers
 
 (defn serialize