Workshop IA

Slides

01_Workshop_IA

 

 

---

Laboratório

Ferramentas

Python

Iris dataset

Scikit Learn

SciKit – Install

Exemplos de algoritmos supervisionados do Scikit com o Iris dataset:

KNN

SVM

Neural Networks

Código de exemplo:

import numpy as np
import matplotlib.pyplot as plt
import time

from sklearn import datasets, svm, neighbors, neural_network



iris = datasets.load_iris()

#print iris

X = iris.data
Y = iris.target

np.random.seed(2)
indices = np.random.permutation(len(X))

#print "indices:", (indices)


x_train = X[indices[:-10]]
y_train = Y[indices[:-10]]
x_test = X[indices[-10:]]
y_test = Y[indices[-10:]]

#model = CLASSIFIER_MODEL


# KNN
model = neighbors.KNeighborsClassifier()

#SVM
model = svm.SVC(kernel='rbf', gamma=1)

# MLP
model = neural_network.MLPClassifier(activation='relu', alpha=1e-05, batch_size='auto',
beta_1=0.9, beta_2=0.999, early_stopping=False,
epsilon=1e-08, hidden_layer_sizes=(5, 5), learning_rate='constant',
learning_rate_init=0.001, max_iter=200, momentum=0.9,
nesterovs_momentum=True, power_t=0.5, random_state=1, shuffle=True,
solver='lbfgs', tol=0.0001, validation_fraction=0.1, verbose=False,
warm_start=False)



before = time.time()

# Treina o modelo
model.fit(x_train,y_train)

after = time.time()

print
print "Time of training", after - before
print
print "Predictions:", model.predict(x_test)
print

---

 

---

Material Extra

Espaço de características:

https://jcaip.github.io/Machine-Learning/

Professor Rodrigo Mello:

http://conteudo.icmc.usp.br/pessoas/mello/