I am comparing the training accuracies of two different neural networks. How can I set the scales so that they are comparable. (like setting both y axis to 1 for both so that the graphs are comparable)
The code I used is below:
def NeuralNetwork(X_train, Y_train, X_val, Y_val, epochs, nodes, lr):
hidden_layers = len(nodes) - 1
weights = InitializeWeights(nodes)
Training_accuracy=[]
Validation_accuracy=[]
for epoch in range(1, epochs+1):
weights = Train(X_train, Y_train, lr, weights)
if (epoch % 1 == 0):
print("Epoch {}".format(epoch))
print("Training Accuracy:{}".format(Accuracy(X_train, Y_train, weights)))
if X_val.any():
print("Validation Accuracy:{}".format(Accuracy(X_val, Y_val, weights)))
Training_accuracy.append(Accuracy(X_train, Y_train, weights))
Validation_accuracy.append(Accuracy(X_val, Y_val, weights))
plt.plot(Training_accuracy)
plt.plot((Validation_accuracy),'#008000')
plt.legend(["Training_accuracy", "Validation_accuracy"])
plt.xlabel("Epoch")
plt.ylabel("Accuracy")
return weights , Training_accuracy , Validation_accuracy
and the two graphs are as below :
try use matplotlib.pyplot.ylim(low, high) refer this link https://www.geeksforgeeks.org/matplotlib-pyplot-ylim-in-python/