Delete sandra_loss_visualization.py

Code/sandra_loss_visualization.py

-# loss visualization
-
-import matplotlib.pyplot as plt
-import re
-import numpy as np
-from statistics import mean
-
-with open("output_loss_epoch_data.txt", "r") as f:
-    loss_file = f.read()
-
-loss_pattern = r"\((\d+\.\d+)"
-
-epochs_pattern = r"Epoche: *(\d+)"
-
-loss_values = re.findall(loss_pattern, loss_file)
-
-eps = [1, 2, 3, 4, 5]
-#epochs = re.findall(epochs_pattern, loss_file)
-# test test test
-
-single_loss_batch = int(len(loss_values)/5)
-
-
-loss_1 = loss_values[:single_loss_batch]
-loss_1 = [eval(i) for i in loss_1]
-loss_1_1 = loss_1[:84]
-loss_1_2 = loss_1[len(loss_1_1):len(loss_1_1)+84]
-loss_1_3 = loss_1[len(loss_1_2):len(loss_1_2)+84]
-loss_1_4 = loss_1[len(loss_1_3):len(loss_1_3)+84]
-loss_1_5 = loss_1[len(loss_1_4):len(loss_1_4)+84]
-mean_1_1 = mean(loss_1_1)
-mean_1_2 = mean(loss_1_2)
-mean_1_3 = mean(loss_1_3)
-mean_1_4 = mean(loss_1_4)
-mean_1_5 = mean(loss_1_5)
-
-
-loss_2 = loss_values[len(loss_1):len(loss_1)+single_loss_batch]
-loss_2 = [eval(i) for i in loss_2]
-loss_2_1 = loss_2[:84]
-loss_2_2 = loss_2[len(loss_2_1):len(loss_2_1)+84]
-loss_2_3 = loss_2[len(loss_2_2):len(loss_2_2)+84]
-loss_2_4 = loss_2[len(loss_2_3):len(loss_2_3)+84]
-loss_2_5 = loss_2[len(loss_2_4):len(loss_2_4)+84]
-mean_2_1 = mean(loss_2_1)
-mean_2_2 = mean(loss_2_2)
-mean_2_3 = mean(loss_2_3)
-mean_2_4 = mean(loss_2_4)
-mean_2_5 = mean(loss_2_5)
-
-
-loss_3 = loss_values[len(loss_2):len(loss_2)+single_loss_batch]
-loss_3 = [eval(i) for i in loss_3]
-loss_3_1 = loss_3[:84]
-loss_3_2 = loss_3[len(loss_3_1):len(loss_3_1)+84]
-loss_3_3 = loss_3[len(loss_3_2):len(loss_3_2)+84]
-loss_3_4 = loss_3[len(loss_3_3):len(loss_3_3)+84]
-loss_3_5 = loss_3[len(loss_3_4):len(loss_3_4)+84]
-mean_3_1 = mean(loss_3_1)
-mean_3_2 = mean(loss_3_2)
-mean_3_3 = mean(loss_3_3)
-mean_3_4 = mean(loss_3_4)
-mean_3_5 = mean(loss_3_5)
-
-loss_4 = loss_values[len(loss_3):len(loss_3)+single_loss_batch]
-loss_4 = [eval(i) for i in loss_4]
-loss_4_1 = loss_4[:84]
-loss_4_2 = loss_4[len(loss_4_1):len(loss_4_1)+84]
-loss_4_3 = loss_4[len(loss_4_2):len(loss_4_2)+84]
-loss_4_4 = loss_4[len(loss_4_3):len(loss_4_3)+84]
-loss_4_5 = loss_4[len(loss_4_4):len(loss_4_4)+84]
-mean_4_1 = mean(loss_4_1)
-mean_4_2 = mean(loss_4_2)
-mean_4_3 = mean(loss_4_3)
-mean_4_4 = mean(loss_4_4)
-mean_4_5 = mean(loss_4_5)
-
-
-loss_5 = loss_values[len(loss_4):len(loss_4)+single_loss_batch]
-loss_5 = [eval(i) for i in loss_5]
-loss_5_1 = loss_2[:84]
-loss_5_2 = loss_5[len(loss_5_1):len(loss_5_1)+84]
-loss_5_3 = loss_5[len(loss_5_2):len(loss_5_2)+84]
-loss_5_4 = loss_5[len(loss_5_3):len(loss_5_3)+84]
-loss_5_5 = loss_5[len(loss_5_4):len(loss_5_4)+84]
-mean_5_1 = mean(loss_5_1)
-mean_5_2 = mean(loss_5_2)
-mean_5_3 = mean(loss_5_3)
-mean_5_4 = mean(loss_5_4)
-mean_5_5 = mean(loss_5_5)
-
-
-    
-plt.ylabel("Loss")
-plt.xlabel("Epochs")
-plt.title("Loss of Baselines")
-
-
-plt.plot(eps, [mean_1_1, mean_1_2, mean_1_3, mean_1_4, mean_1_5], label = "loss set 1")
-
-plt.plot(eps, [mean_2_1, mean_2_2, mean_2_3, mean_2_4, mean_2_5], label = "loss set 2")
-
-plt.plot(eps, [mean_3_1, mean_3_2, mean_3_3, mean_3_4, mean_3_5], label = "loss set 3")
-
-plt.plot(eps, [mean_4_1, mean_4_2, mean_4_3, mean_4_4, mean_4_5], label = "loss set 4")
-
-plt.plot(eps, [mean_5_1, mean_5_2, mean_5_3, mean_5_4, mean_5_5], label = "loss set 5")
-
-plt.legend()
-plt.show()