【gcForest源码解读】win_utils.py

最新推荐文章于 2022-09-30 22:09:43 发布

褚骏逸

最新推荐文章于 2022-09-30 22:09:43 发布

阅读量236

点赞数 1

分类专栏： python # gcForest machine_learning # deep_learning

本文链接：https://blog.csdn.net/weixin_41171061/article/details/84338413

版权

python 同时被 3 个专栏收录

30 篇文章 1 订阅

订阅专栏

machine_learning

29 篇文章 3 订阅

订阅专栏

deep_learning

21 篇文章 0 订阅

订阅专栏

一.序列数据

样本\序列	1	2	3	4	5	6	7	8	9	10	11
Sample_1	A	B	E	C	D	E	F	G	A	B	C
Sample_2	B	A	B	C	D	E	F	G	A	B	C
Sample_2	A	B	C	D	E	F	G	A	B	C	D
Sample_2	A	D	C	D	E	F	G	A	B	C	M
Sample_5	N	A	B	C	D	E	F	G	A	B	C

原维度：shape =（#sample,#feature）= (5,11)

reshape：shape = (#sample,#feature,seq_len,height_y) = (n,c,h,w) = (5，1，11，1) #序列数据?height_y = 1（固定）

二.代码注释

以下代码只提取了需要理解的部分，且为修改后的代码。

def get_windows(X, win_x, win_y, stride_x=1, stride_y=1, pad_x=0, pad_y=0):
    """
    parallizing get_windows
    Arguments:
        X (ndarray): n x c x h x w
    Return:
        X_win (ndarray): n x nh x nw x nc
    """
    n, c, h, w = X.shape
    nc = win_y * win_x * c
    nh = (h - win_x) / stride_x + 1
    nw = (w - win_y) / stride_y + 1
    X_win = np.empty(( nc, n * nh * nw), dtype=np.float32)

    Parallel(n_jobs=-1)(delayed(get_windows_channel)
(X, X_win, des_id, nw, nh, win_x, win_y, stride_x, stride_y) 
for des_id in range(c * win_x * win_y))
    
    X_win = X_win.reshape((n, nh, nw, nc))
    return X_win

X_win.shape = (nc,n*nh*nw) → (n,nh,nw,nc) 即

#windows/per sample（见上图）： $nc = win_y\ast win_x\ast c$

#在x轴方向上窗口移动后的样本的向量长度 / #pooled_feature_x： $nh = \frac{h-win_x}{stride_x} +1$

#在y轴方向上窗口移动后的样本的向量长度 / #pooled_feature_y： $nw = \frac{w-win_y}{stride_y} +1$ （序列数据：w = 1 , stride_y = 1）

注意：其中的Parallel()(delayed(get_windows_channel)...)代码，是指在此时要运行如下代码：

def get_windows_channel(X, X_win, des_id, nw, nh, win_x, win_y, stride_x, stride_y):
    """
    X: N x C x H x W
    X_win: N x nc x nh x nw
    (k, di, dj) in range(X.channle, win_y, win_x)
    """
    #des_id = (k * win_y + di) * win_x + dj
    dj = des_id % win_x
    di = des_id / win_x % win_y
    k = des_id / win_x / win_y
    src = X[:, k, dj:dj+nh*stride_x:stride_x, di:di+nw*stride_y:stride_y].ravel()
    des = X_win[des_id, :]
    np.copyto(des, src)

实际上就是：

for des_id in range(c * win_x * win_y) = range(nc):
    for k in range(c) = range(#feature) = range(1) = 0:
    for di in range(win_y):
    for dj in range(win_x):

des.shape = (n*nh*nw)

src.shape = [n * (dj:dj+nh*stride_x:stride_x) * (di:di+nw*stride_y:stride_y)] #双冒号为切片(start:end:step)

src和des的维度应该是一致的。