Ask Your Question

Revision history [back]

click to hide/show revision 1
initial version

asked 2018-04-25 18:34:38 -0600

sjhalayka gravatar image

Python ANN, training parameters

I have the following code for the ANN in Python. How do I set the NO_INPUT_SCALE parameter? Thanks for any help that you can provide.

 # https://www.safaribooksonline.com/library/view/learning-opencv-3/9781785283840/ch09s03.html

import cv2
import numpy as np

ann = cv2.ml.ANN_MLP_create()
ann.setLayerSizes(np.array([2, 8, 1], dtype=np.uint8))
ann.setTrainMethod(cv2.ml.ANN_MLP_BACKPROP)
ann.setActivationFunction(cv2.ml.ANN_MLP_SIGMOID_SYM)
ann.setTermCriteria((cv2.TERM_CRITERIA_COUNT | cv2.TERM_CRITERIA_EPS, 10000, 0.00001 ))

input_array = np.array([[0.0, 0.0], [0.0, 1.0], [1.0, 0.0], [1.0, 1.0]], dtype=np.float32)
output_array = np.array([0.0, 0.0, 0.0, 1.0], dtype=np.float32)

ann.train(input_array, cv2.ml.ROW_SAMPLE, output_array)

print(ann.predict(input_array))

Python ANN, training parameters

I have the following code for the ANN in Python. How do I set the NO_INPUT_SCALE parameter? Thanks for any help that you can provide.

EDIT: I put up the newest code, which is pretty close to C++. It's still not working:

 # https://www.safaribooksonline.com/library/view/learning-opencv-3/9781785283840/ch09s03.html

import cv2
import numpy as np

ann = cv2.ml.ANN_MLP_create()
ann.setLayerSizes(np.array([2, 8, 1], dtype=np.uint8))
ann.setTrainMethod(cv2.ml.ANN_MLP_BACKPROP)
ann.setTrainMethod(cv2.ml.ANN_MLP_BACKPROP, 0.0001)
ann.setActivationFunction(cv2.ml.ANN_MLP_SIGMOID_SYM)
ann.setTermCriteria((cv2.TERM_CRITERIA_COUNT | cv2.TERM_CRITERIA_EPS, 10000, 0.00001 1, 0.001 ))

input_array = np.array([[0.0, 0.0], [0.0, 1.0], [1.0, 0.0], [1.0, np.array([[1.0, 1.0]], dtype=np.float32)
output_array = np.array([0.0, 0.0, 0.0, 1.0], np.array([[1.0]], dtype=np.float32)

ann.train(input_array, td = cv2.ml.TrainData_create(input_array, cv2.ml.ROW_SAMPLE, output_array)

for i in range(0, 10000):
    ann.train(td, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)

print(ann.predict(input_array))

Python ANN, training parameters

I have the following code for the ANN in Python. How do I set the NO_INPUT_SCALE parameter? Thanks for any help that you can provide.

EDIT: I put up the newest code, which is pretty close to C++. It's still not working:sort of working?

import cv2
import numpy as np

ann = cv2.ml.ANN_MLP_create()
ann.setLayerSizes(np.array([2, 8, 1], dtype=np.uint8))
ann.setTrainMethod(cv2.ml.ANN_MLP_BACKPROP, 0.0001)
ann.setActivationFunction(cv2.ml.ANN_MLP_SIGMOID_SYM)
ann.setTermCriteria((cv2.TERM_CRITERIA_COUNT | cv2.TERM_CRITERIA_EPS, 1, 0.001 ))

input_array = np.array([[1.0, 1.0]], dtype=np.float32)
output_array = np.array([[1.0]], dtype=np.float32)
td = cv2.ml.TrainData_create(input_array, cv2.ml.ROW_SAMPLE, output_array)

ann.train(td, cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)

for i in range(0, 10000):
    ann.train(td, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)

print(ann.predict(input_array))

Python ANN, training parameters

I have the following code for the ANN in Python. How do I set the NO_INPUT_SCALE parameter? Thanks for any help that you can provide.

EDIT: I put up the newest code, which is pretty close to C++. It's sort of working?

import cv2
import numpy as np

ann = cv2.ml.ANN_MLP_create()
ann.setLayerSizes(np.array([2, 8, 1], dtype=np.uint8))
ann.setTrainMethod(cv2.ml.ANN_MLP_BACKPROP, 0.0001)
ann.setActivationFunction(cv2.ml.ANN_MLP_SIGMOID_SYM)
ann.setTermCriteria((cv2.TERM_CRITERIA_COUNT | cv2.TERM_CRITERIA_EPS, 1, 0.001 ))

input_array = np.array([[1.0, 1.0]], dtype=np.float32)
output_array = np.array([[1.0]], dtype=np.float32)
td = cv2.ml.TrainData_create(input_array, cv2.ml.ROW_SAMPLE, output_array)

ann.train(td, cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)

for i in range(0, 10000):
    ann.train(td, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)

print(ann.predict(input_array))

Thanks to @berak for the help!!!

Python ANN, training parameters

I have the following code for the ANN in Python. How do I set the NO_INPUT_SCALE parameter? Thanks for any help that you can provide.

EDIT: I put up the newest code, which is pretty close to C++. It's sort of working?working.

import cv2
import numpy as np

ann = cv2.ml.ANN_MLP_create()
ann.setLayerSizes(np.array([2, 8, 1], dtype=np.uint8))
ann.setTrainMethod(cv2.ml.ANN_MLP_BACKPROP, 0.0001)
ann.setActivationFunction(cv2.ml.ANN_MLP_SIGMOID_SYM)
ann.setTermCriteria((cv2.TERM_CRITERIA_COUNT | cv2.TERM_CRITERIA_EPS, 1, 0.001 ))

input_array = np.array([[1.0, 1.0]], dtype=np.float32)
output_array = np.array([[1.0]], dtype=np.float32)
td = cv2.ml.TrainData_create(input_array, cv2.ml.ROW_SAMPLE, output_array)

ann.train(td, cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)

for i in range(0, 10000):
    ann.train(td, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)

print(ann.predict(input_array))

Thanks to @berak for the help!!!

Python ANN, training parameters

EDIT: I put up the newest code, which is pretty close to C++. It's working.

import cv2
import numpy as np

ann = cv2.ml.ANN_MLP_create()
ann.setLayerSizes(np.array([2, 8, 1], dtype=np.uint8))
ann.setTrainMethod(cv2.ml.ANN_MLP_BACKPROP, 0.0001)
ann.setActivationFunction(cv2.ml.ANN_MLP_SIGMOID_SYM)
ann.setTermCriteria((cv2.TERM_CRITERIA_COUNT | cv2.TERM_CRITERIA_EPS, 1, 0.001 ))

input_array = np.array([[1.0, 1.0]], dtype=np.float32)
output_array = np.array([[1.0]], dtype=np.float32)
td = cv2.ml.TrainData_create(input_array, cv2.ml.ROW_SAMPLE, output_array)

ann.train(td, cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)

for i in range(0, 10000):
    ann.train(td, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)

print(ann.predict(input_array))

Thanks to @berak for the help!!!

I put the code into a repository: https://github.com/sjhalayka/python_opencv_xor

Python ANN, training parameters

EDIT: I put up the newest code, which is pretty close to C++. It's working.

import cv2
import numpy as np

ann = cv2.ml.ANN_MLP_create()
ann.setLayerSizes(np.array([2, 8, 1], dtype=np.uint8))
ann.setTrainMethod(cv2.ml.ANN_MLP_BACKPROP, 0.0001)
ann.setActivationFunction(cv2.ml.ANN_MLP_SIGMOID_SYM)
ann.setTermCriteria((cv2.TERM_CRITERIA_COUNT | cv2.TERM_CRITERIA_EPS, 1, 0.001 0.000001 ))

input_array input_array0 = np.array([[0.0, 0.0]], dtype=np.float32)
output_array0 = np.array([[0.0]], dtype=np.float32)
input_array1 = np.array([[1.0, 0.0]], dtype=np.float32)
output_array1 = np.array([[0.0]], dtype=np.float32)
input_array2 = np.array([[0.0, 1.0]], dtype=np.float32)
output_array2 = np.array([[0.0]], dtype=np.float32)
input_array3 = np.array([[1.0, 1.0]], dtype=np.float32)
output_array output_array3 = np.array([[1.0]], dtype=np.float32)
td 
td0 = cv2.ml.TrainData_create(input_array, cv2.ml.TrainData_create(input_array0, cv2.ml.ROW_SAMPLE, output_array)
output_array0)
td1 = cv2.ml.TrainData_create(input_array1, cv2.ml.ROW_SAMPLE, output_array1)
td2 = cv2.ml.TrainData_create(input_array2, cv2.ml.ROW_SAMPLE, output_array2)
td3 = cv2.ml.TrainData_create(input_array3, cv2.ml.ROW_SAMPLE, output_array3)
 ann.train(td, ann.train(td0, cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)

for i in range(0, 10000):
100000):
 ann.train(td, ann.train(td0, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)
    ann.train(td1, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)
    ann.train(td2, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)
    ann.train(td3, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)

print(ann.predict(input_array))
print(ann.predict(input_array0))
print(ann.predict(input_array1))
print(ann.predict(input_array2))
print(ann.predict(input_array3))

Thanks to @berak for the help!!!

I put the code into a repository: https://github.com/sjhalayka/python_opencv_xor

Python ANN, training parameters

EDIT: I put up the newest code, which is pretty close to C++. It's working.sort of working?

import cv2
import numpy as np

ann = cv2.ml.ANN_MLP_create()
ann.setLayerSizes(np.array([2, 8, 1], dtype=np.uint8))
ann.setTrainMethod(cv2.ml.ANN_MLP_BACKPROP, 0.0001)
ann.setActivationFunction(cv2.ml.ANN_MLP_SIGMOID_SYM)
ann.setTermCriteria((cv2.TERM_CRITERIA_COUNT | cv2.TERM_CRITERIA_EPS, 1, 0.000001 ))

input_array0 = np.array([[0.0, 0.0]], dtype=np.float32)
output_array0 = np.array([[0.0]], dtype=np.float32)
input_array1 = np.array([[1.0, 0.0]], dtype=np.float32)
output_array1 = np.array([[0.0]], dtype=np.float32)
input_array2 = np.array([[0.0, 1.0]], dtype=np.float32)
output_array2 = np.array([[0.0]], dtype=np.float32)
input_array3 = np.array([[1.0, 1.0]], dtype=np.float32)
output_array3 = np.array([[1.0]], dtype=np.float32)

td0 = cv2.ml.TrainData_create(input_array0, cv2.ml.ROW_SAMPLE, output_array0)
td1 = cv2.ml.TrainData_create(input_array1, cv2.ml.ROW_SAMPLE, output_array1)
td2 = cv2.ml.TrainData_create(input_array2, cv2.ml.ROW_SAMPLE, output_array2)
td3 = cv2.ml.TrainData_create(input_array3, cv2.ml.ROW_SAMPLE, output_array3)

ann.train(td0, cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)

for i in range(0, 100000):
    ann.train(td0, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)
    ann.train(td1, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)
    ann.train(td2, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)
    ann.train(td3, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)

print(ann.predict(input_array0))
print(ann.predict(input_array1))
print(ann.predict(input_array2))
print(ann.predict(input_array3))

Thanks to @berak for the help!!!

I put the code into a repository: https://github.com/sjhalayka/python_opencv_xor

Python ANN, training parameters

EDIT: I put up the newest code, which is pretty close to C++. It's sort of working?working.

import cv2
import numpy as np

ann = cv2.ml.ANN_MLP_create()
ann.setLayerSizes(np.array([2, 8, 1], dtype=np.uint8))
ann.setTrainMethod(cv2.ml.ANN_MLP_BACKPROP, 0.0001)
0.1)
ann.setActivationFunction(cv2.ml.ANN_MLP_SIGMOID_SYM)
ann.setTermCriteria((cv2.TERM_CRITERIA_COUNT | cv2.TERM_CRITERIA_EPS, 1, 0.000001 ))

input_array0 = np.array([[0.0, 0.0]], dtype=np.float32)
output_array0 = np.array([[0.0]], dtype=np.float32)
input_array1 = np.array([[1.0, 0.0]], dtype=np.float32)
output_array1 = np.array([[0.0]], dtype=np.float32)
input_array2 = np.array([[0.0, 1.0]], dtype=np.float32)
output_array2 = np.array([[0.0]], dtype=np.float32)
input_array3 = np.array([[1.0, 1.0]], dtype=np.float32)
output_array3 = np.array([[1.0]], dtype=np.float32)

td0 = cv2.ml.TrainData_create(input_array0, cv2.ml.ROW_SAMPLE, output_array0)
td1 = cv2.ml.TrainData_create(input_array1, cv2.ml.ROW_SAMPLE, output_array1)
td2 = cv2.ml.TrainData_create(input_array2, cv2.ml.ROW_SAMPLE, output_array2)
td3 = cv2.ml.TrainData_create(input_array3, cv2.ml.ROW_SAMPLE, output_array3)

ann.train(td0, cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)

for i in range(0, 100000):
    ann.train(td0, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)
    ann.train(td1, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)
    ann.train(td2, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)
    ann.train(td3, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)

print(ann.predict(input_array0))
print(ann.predict(input_array1))
print(ann.predict(input_array2))
print(ann.predict(input_array3))

Thanks to @berak for the help!!!

I put the code into a repository: https://github.com/sjhalayka/python_opencv_xor

Python ANN, training parameters

EDIT: I put had a question on how to use the training parameters. I ended up writing the newest Python code, which is pretty close to C++. C++ in this case. It's working.working:

import cv2
import numpy as np

ann = cv2.ml.ANN_MLP_create()
ann.setLayerSizes(np.array([2, 8, 1], dtype=np.uint8))
ann.setTrainMethod(cv2.ml.ANN_MLP_BACKPROP, 0.1)
ann.setActivationFunction(cv2.ml.ANN_MLP_SIGMOID_SYM)
ann.setTermCriteria((cv2.TERM_CRITERIA_COUNT | cv2.TERM_CRITERIA_EPS, 1, 0.000001 ))

input_array0 = np.array([[0.0, 0.0]], dtype=np.float32)
output_array0 = np.array([[0.0]], dtype=np.float32)
input_array1 = np.array([[1.0, 0.0]], dtype=np.float32)
output_array1 = np.array([[0.0]], dtype=np.float32)
input_array2 = np.array([[0.0, 1.0]], dtype=np.float32)
output_array2 = np.array([[0.0]], dtype=np.float32)
input_array3 = np.array([[1.0, 1.0]], dtype=np.float32)
output_array3 = np.array([[1.0]], dtype=np.float32)

td0 = cv2.ml.TrainData_create(input_array0, cv2.ml.ROW_SAMPLE, output_array0)
td1 = cv2.ml.TrainData_create(input_array1, cv2.ml.ROW_SAMPLE, output_array1)
td2 = cv2.ml.TrainData_create(input_array2, cv2.ml.ROW_SAMPLE, output_array2)
td3 = cv2.ml.TrainData_create(input_array3, cv2.ml.ROW_SAMPLE, output_array3)

ann.train(td0, cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)

for i in range(0, 100000):
    ann.train(td0, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)
    ann.train(td1, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)
    ann.train(td2, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)
    ann.train(td3, cv2.ml.ANN_MLP_UPDATE_WEIGHTS | cv2.ml.ANN_MLP_NO_INPUT_SCALE | cv2.ml.ANN_MLP_NO_OUTPUT_SCALE)

print(ann.predict(input_array0))
print(ann.predict(input_array1))
print(ann.predict(input_array2))
print(ann.predict(input_array3))

https://github.com/sjhalayka/python_opencv_xor/blob/master/opencv_xor.py

Thanks to @berak for the help!!!

I put the code into a repository: https://github.com/sjhalayka/python_opencv_xor

Python ANN, training parameters

EDIT: I had a question on how to use the ANN training parameters. I ended up writing the Python code, which is pretty close to C++ in this case. It's working:

https://github.com/sjhalayka/python_opencv_xor/blob/master/opencv_xor.py

Thanks to @berak for the help!!!

Python ANN, training parameters

EDIT: I had a question on how to use the ANN training parameters. I ended up writing the Python code, which is pretty close to C++ in this case. It's working:The network learns the XOR problem:

https://github.com/sjhalayka/python_opencv_xor/blob/master/opencv_xor.py

Thanks to @berak for the help!!!

Copyright OpenCV foundation, 2012-2018. Content on this site is licensed under a Creative Commons Attribution Share Alike 3.0 license.
about | faq | help | privacy policy | terms of service
Powered by Askbot version 0.10.2