saveImage(img_dirs, "_remove_margin", img_remove_margin)
# 扩充为正方形并缩小为256x256
img_new = normalization(img_remove_margin)
saveImage(img_dirs, "_new", img_new)
# 打印信息到输出台
printToConsole(start_time, f, count, total, 5)
success += 1
except Exception as e:
# 错误情况
saveError(e, out_dir, f)
fail += 1
end_time = datetime.datetime.now()
expend = end_time - start_time
print(
"\n\ntotal: %d\nsuccessful: %d\nskip: %d\nfailed: %d\nExpend time: %s"
% (total, success, skip, fail, expend))
os.startfile(out_dir)
if __name__ == '__main__':
file_path = "C:\\Study\\test\\old"
out_dir = "C:\\Study\\test\\old_no_norm_no_sucessxxx"
data = "new_data.txt"
feature, label = loadData(data)
# 直方图归一化
# feature = handleHistogram(feature)
w0 = ridgeRegression(feature, label, 0.5)
handle(file_path, out_dir, (45, -45, 45, -45), w0)
def test():
inputNodes = 256
hiddenNodes = 1000
outputNodes = 150 # 输出的值个数
learningRate = 0.1
net = neuralNetwork(inputNodes, hiddenNodes, outputNodes, learningRate)
# 读取训练数据集
data, labels = loadData("new_data.txt")
data = handleHistogram(data)
m, n = numpy.shape(data)
######## 训练 #########
iteration = 50 # 迭代次数
for it in range(iteration):
for i in range(m):
inputs = data[i, :] + 0.01
targets = numpy.zeros(outputNodes) + 0.0001 # 初始化输出节点，避免0
targets[int(labels[i,
0])] = 0.99 # 将数据的每个标签的值置为0.99(避免1)， 数据标签为0-155
# 避免0和1是因为激活函数生成0和1是不可能
net.train(inputs, targets)
# print(targets, int(labels[i, 0]), inputs)
# break
# break
print("完成第 ", it, "次迭代")
############# 测试 ##############
predict_y = []
result = [] # 存储test结果
for i in range(m):
inputs = data[i, :] + 0.01
outputs = net.test(inputs)
label = numpy.argmax(outputs) # 找出最大索引，若训练正确则索引即为标签值
predict_y.append(int(label))
# print(label, labels[i, 0])
# print(outputs)
if label == labels[i, 0]:
print("第 ", i, "成功！")
result.append(1)
else:
print("第 ", i, "失败！")
result.append(0)
# break
# 最终准确率
result_array = numpy.asarray(result) # asarray不会深度拷贝数组，占用内存少
print("识别率 = ", result_array.sum() / result_array.size)
actual_x = [] # 绘制直线的x轴坐标
predict_x = [] # 绘制预测值的x坐标
for i in labels:
actual_x.append(int(i[0]))
predict_x.append(int(i[0]))
actual_y = actual_x # 直线的y坐标
# 得到预测值
color = numpy.arctan2(predict_y, predict_x)
# 绘制散点图
plt.scatter(predict_x, predict_y, s=10, c=color, alpha=1)
# 设置坐标轴范围
plt.xlim([0, 150])
plt.ylim([0, 150])
plt.xlabel("actual value")
plt.ylabel("prediction")
plt.plot(actual_x, actual_y)
plt.savefig("bp")
plt.show()
# coding:UTF-8
# 2018-10-25
# https://blog.csdn.net/zchshhh/article/details/78215087
from api import handleHistogram, plotScatter, loadData
from regression import ridgeRegression
if __name__ == '__main__':
print("loading data ...")
feature, label = loadData("data.txt")
feature = handleHistogram(feature, alpha=20000, is_total=True)
# feature = handleHistogram(feature)
w0 = ridgeRegression(feature, label, 0.5)
plotScatter(feature, label, w0, [(0, 150), (0, 150)], "regression")
end_time = datetime.datetime.now()
expend = end_time - start_time
print("\n\ntotal: %d\nsuccessful: %d\nskip: %d\nfailed: %d\nExpend time: %s" %(total, success, skip, fail, expend))
os.startfile(out_dir)
if __name__ == '__main__':
file_path = r"C:\Study\test\bone\2"
out_dir = "C:\\Study\\test\\softmax_norm_test"
# weights_file = "weights.npy"
# w0 = loadWeights(weights_file)
# w0 = np.load(weights_file)
# print(np.shape(w0))
# print(w0)
inputfile = "data.txt"
feature, label = loadData(inputfile)
feature = handleHistogram(feature)
# print(np.shape(feature), np.shape(label))
# print(feature)
k = 256
w0 = train(feature, label, k, 100000, 0.1)
fp = open("pickle_weight_test.dat", "wb")
pickle.dump(w0, fp)
fp.close()
handle(file_path, out_dir, (45,-45,45,-45), w0)
# # 分割评估
# file_path = "C:\\Study\\test\\100-gt" # 标准分割图像目录路径
# out_dir = "C:\\Study\\test\\est_results" #结果保存目录
# coding:UTF-8
# 2018-11-7
# 绘制像素均值-最佳阈值直方图
from api import loadData, getHistogramMean
import numpy as np
import matplotlib.pyplot as plt
import matplotlib
matplotlib.use('Agg')
if __name__ == '__main__':
print("loading data ...")
feature, label = loadData("new_data.txt")
# feature, label = dataFilter(feature, label)
line_x, line_y = [], []
x, y = [], []
m, n = np.shape(feature)
for i in range(m):
mean = getHistogramMean(feature[i, :])
x.append(int(mean))
y.append(int(label[i, :]))
print(x, y)
color = np.arctan2(y, x)
# 绘制散点图
plt.scatter(x, y, s=10, c=color, alpha=1)
# 设置坐标轴范围
plt.xlim([0, 150])
plt.ylim([0, 150])
line_x = x
line_y = x
plt.xlabel("mean value")