Simple neural network implementation using Chainer-Data preparation-
From Last time, I am writing an article to actually build a neural network using Chainer, which is a framework for deep learning. This time
-
Data preparation
-
Model description
-
Optimization algorithm settings
-
Learning
-
Result output Of
-
Data preparation
I will write about.
Iris data
This time, Iris data is used as sample data. It seems that Iris data is often used in machine learning. This Iris data can be easily introduced from Python's machine learning library scikit-learn. scikit-learn itself can be installed with pip.
$ pip install scikit-learn
By the way, as of November 18, 2016, as a condition to install scikit-learn
Python (>= 2.6 or >= 3.3),
NumPy (>= 1.6.1),
SciPy (>= 0.9).
Is required to be.
The content of this Iris data is a four-dimensional vector. Let's check the contents.
>>> from sklearn import datasets
>>> iris = datasets.load_iris()
>>> iris
>>> iris
{'target_names': array(['setosa', 'versicolor', 'virginica'],
dtype='<U10'), 'target': array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2]), 'feature_names': ['sepal length (cm)', 'sepal width (cm)', 'petal length (cm)', 'petal width (cm)'], 'data': array([[ 5.1, 3.5, 1.4, 0.2],
[ 4.9, 3. , 1.4, 0.2],
[ 4.7, 3.2, 1.3, 0.2],
[ 4.6, 3.1, 1.5, 0.2],
[ 5. , 3.6, 1.4, 0.2],
[ 5.4, 3.9, 1.7, 0.4],
[ 4.6, 3.4, 1.4, 0.3],
[ 5. , 3.4, 1.5, 0.2],
[ 4.4, 2.9, 1.4, 0.2],
[ 4.9, 3.1, 1.5, 0.1],
[ 5.4, 3.7, 1.5, 0.2],
[ 4.8, 3.4, 1.6, 0.2],
[ 4.8, 3. , 1.4, 0.1],
[ 4.3, 3. , 1.1, 0.1],
[ 5.8, 4. , 1.2, 0.2],
[ 5.7, 4.4, 1.5, 0.4],
[ 5.4, 3.9, 1.3, 0.4],
[ 5.1, 3.5, 1.4, 0.3 .................
There are various things, but for the time being, we need data and target.
>>> iris.data
array([[ 5.1, 3.5, 1.4, 0.2],
[ 4.9, 3. , 1.4, 0.2],
[ 4.7, 3.2, 1.3, 0.2],
[ 4.6, 3.1, 1.5, 0.2],
[ 5. , 3.6, 1.4, 0.2],
[ 5.4, 3.9, 1.7, 0.4],
[ 4.6, 3.4, 1.4, 0.3],
[ 5. , 3.4, 1.5, 0.2],
[ 4.4, 2.9, 1.4, 0.2],
[ 4.9, 3.1, 1.5, 0.1],
[ 5.4, 3.7, 1.5, 0.2],
[ 4.8, 3.4, 1.6, 0.2],
[ 4.8, 3. , 1.4, 0.1],
[ 4.3, 3. , 1.1, 0.1],
[ 5.8, 4. , 1.2, 0.2],...........
>>> iris.target
array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2])
data is a 4-dimensional vector, each -Petal length -Petal width -The length of the sepal -Width of sepals is.
The corresponding target is the type of Iris. Each
-setosa
-versicolor
-virginica
is.
I think I got the image somehow.
Teacher data and test data
Machine learning requires teacher data to teach the classifier the characteristics of the data and test data to measure the accuracy of the completed classifier. Dividing the data prepared in this way into teacher data and test data and verifying the validity is called cross-validation.
This is a state in which the types of 4-dimensional iris vectors are associated with labels of 0, 1, and 2.
>>>iris.data
array([[ 5.1, 3.5, 1.4, 0.2],
[ 4.9, 3. , 1.4, 0.2],
[ 4.7, 3.2, 1.3, 0.2],......
>>>iris.target
array([0, 0, 0,.....
Since it is necessary to vectorize the target in order to train the neural network, we will process it. That is, if iris is setosa [1, 0, 0] will do. If iris is versicolor It is [0, 1, 0]. Put 1 in the dimension of the correct answer.
>>>n = y.size
>>>y.zeros(n * 3).reshape(n, 3).astype(np.float32)
array([[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.],
[ 0., 0., 0.]])
Now that the container is ready, we will add the value.
>>>for i in range(n):
y2[i, y[i]] = 1.0
>>> y2
array([[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 1., 0., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 1., 0.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.],
[ 0., 0., 1.]], dtype=float32)
Since there are 150 elements for both data and target, prepare an index.
>>>index = np.arange(n)
array([ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12,
13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25,
26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38,
39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51,
52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64,
65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77,
78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90,
91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103,
104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116,
117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129,
130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142,
143, 144, 145, 146, 147, 148, 149])
The odd elements of the index array are the teacher data, and the even numbers are the test data.
>>>xtrain = x[index[index % 2 != 0], :]
>>>ytrain = y2[index[index % 2 != 0], :]
>>>xtest = x[index[index % 2 == 0], :]
>>>ytest = y2[index[index % 2 == 0], :]
In this way, the data is prepared as corresponding data. Up to here for this time
reference
Takayoshi Yamashita Deep learning Kodansha that can be seen in the illustration Hiroyuki Shinno Practical deep learning with Chainer-How to implement complex NN-Ohmsha
Recommended Posts