|
|
import numpy as np
|
|
|
|
|
|
import lqmtest_x3_2_load_data
|
|
|
import lqmtest_x3_4_conv_proc
|
|
|
|
|
|
class ModelObj: # 网络对象
|
|
|
def __init__(self, ObjID, ObjType, ObjLable, ParaString, ObjX, ObjY):
|
|
|
self.ObjID = ObjID # 图元号
|
|
|
self.ObjType = ObjType # 图元类别
|
|
|
self.ObjLable = ObjLable # 对象标签
|
|
|
self.ParaString = ParaString # 参数字符串
|
|
|
self.ObjX = ObjX # 对象位置x坐标
|
|
|
self.ObjY = ObjY # 对象位置y坐标
|
|
|
class Pool_Class(ModelObj): # 池化对象
|
|
|
def __init__(self, ObjID, ObjType, ObjLable, ParaString, ObjX, ObjY):
|
|
|
super().__init__(ObjID, ObjType, ObjLable, ParaString, ObjX, ObjY)
|
|
|
self.MaxPoolProc = self.pool_proc # 基本操作函数
|
|
|
self.SetPollPara = self.setpool_para # 参数设置函数
|
|
|
def pool_proc(self, image, PoolPara):
|
|
|
pool_mode = PoolPara["pool_mode"]
|
|
|
pool_size = PoolPara["pool_size"]
|
|
|
stride = PoolPara["stride"]
|
|
|
c, h, w = image.shape # 获取输入特征图的高度和宽度
|
|
|
out_h = int((h - pool_size) / stride) + 1 # 计算输出特征图的高度
|
|
|
out_w = int((w - pool_size) / stride) + 1 # 计算输出特征图的宽度
|
|
|
out = np.zeros((c, out_h, out_w)) # 初始化输出特征图为全零数组
|
|
|
for k in range(c): # 对于输出的每一个位置上计算：
|
|
|
for i in range(out_h):
|
|
|
for j in range(out_w):
|
|
|
window = image[k, i * stride:i * stride + pool_size,
|
|
|
j * stride:j * stride + pool_size]
|
|
|
if pool_mode == "max": # 最大池化
|
|
|
out[k][i][j] = np.max(window)
|
|
|
elif pool_mode == "avg": # 平均池化
|
|
|
out[k][i][j] = np.mean(window)
|
|
|
elif pool_mode == "min": # 最小池化
|
|
|
out[k][i][j] = np.min(window)
|
|
|
else: # 无效的池化类型
|
|
|
raise ValueError("Invalid pooling mode")
|
|
|
return out # 返回特征图。
|
|
|
def setpool_para(self): # 定义设置池化参数的函数
|
|
|
pool_mode = input("请输入池化模式（max/avg/min）: ") # 用户输入池化模式
|
|
|
pool_size = int(input("请输入池化大小: ")) # 用户输入池化大小
|
|
|
stride = int(input("请输入步长: ")) # 用户输入步长
|
|
|
PoolPara = {"pool_mode": pool_mode, "pool_size": pool_size,
|
|
|
"stride": stride} # 返回PoolPara参数，这里用字典来存储
|
|
|
return PoolPara # 返回PoolPara参数
|
|
|
if __name__ == '__main__':
|
|
|
DataSet = lqmtest_x3_2_load_data.Data_Class("DataSet1", 1, "数据集1", [], 120, 330)
|
|
|
# setload_data()函数，获取加载数据集的参数
|
|
|
DataPara = DataSet.SetDataPara()
|
|
|
train_images, test_images = DataSet.LoadData(DataPara)
|
|
|
|
|
|
Conv = lqmtest_x3_4_conv_proc.Conv_Class("Conv1", 2, "卷积1", [], 250, 330)
|
|
|
ConvPara = Conv.SetConvPara()
|
|
|
for i in range(len(train_images) // 32):
|
|
|
images = train_images[i * 32:(i + 1) * 32]
|
|
|
conv_images = [] # 存储卷积处理后的图片的列表
|
|
|
for image in images: # 获取训练集的图片数据
|
|
|
dim = len(image.shape) # 获取矩阵的维度
|
|
|
if dim == 2: # 如果是二维矩阵,则转化为三维矩阵
|
|
|
image_h, image_w = image.shape
|
|
|
image = np.reshape(image, (1, image_h, image_w))
|
|
|
# 调用ConvProc()函数，根据ConvPara参数完成卷积计算
|
|
|
output = Conv.ConvProc(image, ConvPara)
|
|
|
conv_images.append(output) # 将卷积结果存储到列表
|
|
|
elif dim == 3: # 若为三维矩阵，则保持不变直接卷积处理
|
|
|
output = Conv.ConvProc(image, ConvPara)
|
|
|
conv_images.append(output)
|
|
|
# 将卷积处理后的图片列表转换为数组形式，方便后续处理
|
|
|
conv_images = np.array(conv_images)
|
|
|
|
|
|
Pool = Pool_Class("Pool1", 3, "最大池化1", [], 380, 330)
|
|
|
PoolPara = Pool.SetPollPara()
|
|
|
pool_images = [] # 存储池化处理后的图片的列表
|
|
|
for image in conv_images: # 获取卷积后的图片数据
|
|
|
output = Pool.MaxPoolProc(image, PoolPara)
|
|
|
pool_images.append(output) # 将池化结果存储到列表
|
|
|
# 将池化处理后的图片列表转换为数组形式，方便后续处理
|
|
|
pool_images = np.array(pool_images)
|
