#!/usr/bin/env python
# coding: utf-8

# In[1]:


import numpy as np
import matplotlib.pyplot as plt
import sklearn.datasets as dt
import sklearn.model_selection as ms
import sklearn.svm as sv


# In[8]:


IRIS = dt.load_iris()
x = IRIS.data
y = IRIS.target
xtr, xte, ytr, yte, = ms.train_test_split(x, y, train_size = 0.75, random_state = 55)


# In[9]:


#Different Kernels:
trACC = []
teACC = []
kernel = []
for i in ["linear", "poly", "rbf", "sigmoid"]:
    clsfr = sv.SVC(kernel = i, degree = 2)
    clsfr.fit(xtr, ytr)
    trACC.append(clsfr.score(xtr, ytr))
    teACC.append(clsfr.score(xte, yte))
    kernel.append(i)


# In[10]:


plt.plot(trACC, label = "Train", c = "#00FFFF")
plt.plot(teACC, label = "Test", c = "#FF4040")
plt.xticks([0, 1, 2, 3], kernel)
plt.xlabel("Kernel")
plt.ylabel("Accuracy")
plt.legend()
plt.show()


# In[11]:


#Different degrees:
trACC = []
teACC = []
degree = []
for i in range (1, 11):
    clsfr = sv.SVC(kernel = "poly", degree = i)
    clsfr.fit(xtr, ytr)
    trACC.append(clsfr.score(xtr, ytr))
    teACC.append(clsfr.score(xte, yte))
    degree.append(i)


# In[12]:


plt.plot(trACC, label = "Train", c = "#00FFFF")
plt.plot(teACC, label = "Test", c = "#FF4040")
plt.xlabel("Degree")
plt.ylabel("Accuracy")
plt.legend()
plt.show()


# In[ ]: