From 8103eabd8f6ba03ff9b4f893ab2d32783d7bb503 Mon Sep 17 00:00:00 2001
From: Fabian Poker <fp168637@fh-muenster.de>
Date: Sun, 7 May 2023 21:22:31 +0200
Subject: [PATCH] Evaluation
---
aufgaben/p5/pla_applied.py | 68 ++++++++++++++++++++++++++++++++++----
1 file changed, 62 insertions(+), 6 deletions(-)
diff --git a/aufgaben/p5/pla_applied.py b/aufgaben/p5/pla_applied.py
index 306838d..e8c6518 100644
--- a/aufgaben/p5/pla_applied.py
+++ b/aufgaben/p5/pla_applied.py
@@ -1,3 +1,5 @@
+import math
+
from algorithm.pla.perceptron_learning_algorithm import train, train_pocket
from algorithm.pla.perceptron import Perceptron
from algorithm.pla.transfer_functions import normalized_tanh
@@ -26,10 +28,37 @@ def apply_pla():
max_iterations = 100
learning_rate = 0.10
train(perceptron, training_data, max_iterations, learning_rate)
- print("\nPLA: Zunächst Gewichte, dann Klassifizierung. Erwartet: 1 & 0")
+ print("\nPLA: Zunächst Gewichte, dann Klassifizierung. Gehen = 1, Kniebeuge = 0")
print(perceptron.weights)
- print(perceptron.classify(joggen.werte[15:18].tolist()))
- print(perceptron.classify(kniebeuge.werte[55:58].tolist()))
+ cnt_true_pos = 0
+ cnt_true_neg = 0
+ cnt_false_pos = 0
+ cnt_false_neg = 0
+ unclassified = 0
+ for i in range(500):
+ jog_res = perceptron.classify(joggen.werte[i:i+3].tolist())
+ knie_res = perceptron.classify(kniebeuge.werte[i:i+3].tolist())
+ if math.isclose(1, jog_res, abs_tol=0.1):
+ cnt_true_pos += 1
+ elif math.isclose(0, jog_res, abs_tol=0.1):
+ cnt_false_neg += 1
+ else:
+ unclassified += 1
+ if math.isclose(0, knie_res, abs_tol=0.1):
+ cnt_true_neg += 1
+ elif math.isclose(1, knie_res, abs_tol=0.1):
+ cnt_false_pos += 1
+ else:
+ unclassified += 1
+ print(f"""
+ Anzahl Prüfungen: 1000
+ True Positive: {cnt_true_pos}
+ False Positive: {cnt_false_pos}
+ True Negative: {cnt_true_neg}
+ False Negative: {cnt_false_neg}
+ Unclassifiable: {unclassified}
+ Fehlerrate: {(cnt_false_neg+cnt_false_pos) / 10}% bzw. {(cnt_false_neg+cnt_false_pos+unclassified) / 10}%
+ """)
def apply_pocket():
@@ -54,10 +83,37 @@ def apply_pocket():
max_iterations = 1
learning_rate = 0.10
train_pocket(perceptron, training_data, max_iterations, learning_rate)
- print("\nPocket: Zunächst Gewichte, dann Klassifizierung. Erwartet: 1 & 0")
+ print("\nPocket: Zunächst Gewichte, dann Klassifizierung. Gehen = 1, Kniebeuge = 0")
print(perceptron.weights)
- print(perceptron.classify(joggen.werte[15:18].tolist()))
- print(perceptron.classify(kniebeuge.werte[55:58].tolist()))
+ cnt_true_pos = 0
+ cnt_true_neg = 0
+ cnt_false_pos = 0
+ cnt_false_neg = 0
+ unclassified = 0
+ for i in range(500):
+ jog_res = perceptron.classify(joggen.werte[i:i+3].tolist())
+ knie_res = perceptron.classify(kniebeuge.werte[i:i+3].tolist())
+ if math.isclose(1, jog_res, abs_tol=0.1):
+ cnt_true_pos += 1
+ elif math.isclose(0, jog_res, abs_tol=0.1):
+ cnt_false_neg += 1
+ else:
+ unclassified += 1
+ if math.isclose(0, knie_res, abs_tol=0.1):
+ cnt_true_neg += 1
+ elif math.isclose(1, knie_res, abs_tol=0.1):
+ cnt_false_pos += 1
+ else:
+ unclassified += 1
+ print(f"""
+ Anzahl Prüfungen: 1000
+ True Positive: {cnt_true_pos}
+ False Positive: {cnt_false_pos}
+ True Negative: {cnt_true_neg}
+ False Negative: {cnt_false_neg}
+ Unclassifiable: {unclassified}
+ Fehlerrate: {(cnt_false_neg+cnt_false_pos) / 10}% bzw. {(cnt_false_neg+cnt_false_pos+unclassified) / 10}%
+ """)
if __name__ == '__main__':
--
GitLab