Commit Inicial

score_all.py

@@ -0,0 +1,78 @@
+import sys
+import time
+import warnings
+
+from experimentation.Sets import Datasets
+from odte import Odte
+from sklearn.ensemble import RandomForestClassifier
+from sklearn.svm import SVC, LinearSVC
+from sklearn.tree import DecisionTreeClassifier
+from stree import Stree
+
+classifiers = {
+    "Stree ": Stree(C=256, max_iter=1e5, random_state=1),
+    "D.Tree": DecisionTreeClassifier(),
+    "SVC(L)": LinearSVC(),
+    "SVC(R)": SVC(kernel="rbf"),
+    "SVC(P)": SVC(kernel="poly"),
+    "R.For.": RandomForestClassifier(),
+    "Odte  ": Odte(random_state=1, n_jobs=-1),
+    # "Oc2   ": ObliqueTree()
+}
+
+
+def header():
+    print(f"Score {dtype} files")
+    initial = f"{'Dataset':30s}"
+    sec_line = "=" * 30
+    for classifier in classifiers:
+        initial += " Time(s)  " + classifier
+        sec_line += " ======== " + "=" * len(classifier)
+    print(initial)
+    print(sec_line)
+
+
+warnings.filterwarnings("ignore")
+dtype = "all"
+if len(sys.argv) > 1:
+    set_name = sys.argv[1]
+    if len(sys.argv) > 2:
+        dtype = sys.argv[2]
+else:
+    set_name = "aaai"
+if set_name != "aaai" and set_name != "tanveer":
+    print("First parameter has to be one of: {aaai, tanveer}")
+    exit(1)
+datasets = Datasets(False, False, set_name)
+clf = None
+header()
+better = worse = equal = 0
+for dataset in datasets:
+    if dataset[1] == dtype or dtype == "all" or dtype == "any":
+        X, y = datasets.load(dataset[0])
+        output = ""
+        odte_score = stree_score = 0.0
+        for model_name, clf in classifiers.items():
+            now = time.time()
+            clf.set_params(random_state=0)
+            score = clf.fit(X, y).score(X, y)
+            if model_name.strip() == "Stree":
+                stree_score = score
+            if model_name.strip() == "Odte":
+                odte_score = score
+            output += f" {time.time()-now:-7.2f} {score:.4f} "
+        line = f"{dataset[0]:30s} " + output
+        if stree_score > odte_score:
+            line += "-"
+            worse += 1
+        elif stree_score < odte_score:
+            line += "+"
+            better += 1
+        else:
+            line += "="
+            equal += 1
+        print(line)
+print(f"Odte is better than Stree {better:2d} times")
+print(f"Odte is worse  than Stree {worse:2d} times")
+print(f"Odte is equal  to   Stree {equal:2d} times")
+print(clf.get_params() if clf is not None else "No dataset match criteria")