oafuncs 0.0.59__py2.py3-none-any.whl

oafuncs/__init__.py

@@ -0,0 +1,26 @@
+#!/usr/bin/env python
+# coding=utf-8
+'''
+Author: Liu Kun && 16031215@qq.com
+Date: 2024-09-17 16:09:20
+LastEditors: Liu Kun && 16031215@qq.com
+LastEditTime: 2024-10-14 17:08:57
+FilePath: \\Python\\My_Funcs\\OAFuncs\\OAFuncs\\__init__.py
+Description:  
+EditPlatform: vscode
+ComputerInfo: XPS 15 9510
+SystemInfo: Windows 11
+Python Version: 3.11
+'''
+
+# 会导致OAFuncs直接导入所有函数，不符合模块化设计
+from .oa_cmap import *
+from .oa_data import *
+from .oa_down import *
+from .oa_draw import *
+from .oa_file import *
+from .oa_help import *
+from .oa_nc import *
+from .oa_python import *
+from .oa_sign import *
+from .oa_tool import *

oafuncs/oa_cmap.py

@@ -0,0 +1,141 @@
+#!/usr/bin/env python
+# coding=utf-8
+'''
+Author: Liu Kun && 16031215@qq.com
+Date: 2024-09-17 16:55:11
+LastEditors: Liu Kun && 16031215@qq.com
+LastEditTime: 2024-10-06 19:34:57
+FilePath: \\Python\\My_Funcs\\OAFuncs\\OAFuncs\\oa_cmap.py
+Description:  
+EditPlatform: vscode
+ComputerInfo: XPS 15 9510
+SystemInfo: Windows 11
+Python Version: 3.11
+'''
+
+
+import matplotlib as mpl
+import matplotlib.pyplot as plt
+import numpy as np
+
+__all__ = ['show', 'extract_colors', 'create_custom',
+           'create_diverging', 'create_5rgb_txt']
+
+# ** 将cmap用填色图可视化（官网摘抄函数）
+
+
+def show(colormaps: list):
+    """
+    Helper function to plot data with associated colormap.
+    example:
+    cmap = ListedColormap(["darkorange", "gold", "lawngreen", "lightseagreen"])
+    show([cmap])
+    """
+    np.random.seed(19680801)
+    data = np.random.randn(30, 30)
+    n = len(colormaps)
+    fig, axs = plt.subplots(1, n, figsize=(n * 2 + 2, 3),
+                            constrained_layout=True, squeeze=False)
+    for [ax, cmap] in zip(axs.flat, colormaps):
+        psm = ax.pcolormesh(data, cmap=cmap, rasterized=True, vmin=-4, vmax=4)
+        fig.colorbar(psm, ax=ax)
+    plt.show()
+
+
+# ** 将cmap转为list，即多个颜色的列表
+def extract_colors(cmap, n=256):
+    '''
+    cmap        : cmap名称
+    n           : 提取颜色数量
+    return      : 提取的颜色列表
+    example     : out_cmap = extract_colors('viridis', 256)
+    '''
+    c_map = mpl.colormaps.get_cmap(cmap)
+    out_cmap = [c_map(i) for i in np.linspace(0, 1, n)]
+    return out_cmap
+
+
+# ** 自制cmap，多色，可带位置
+def create_custom(colors: list, nodes=None):  # 利用颜色快速配色
+    '''
+    func        : 自制cmap，自动确定颜色位置（等比例）
+    description : colors可以是颜色名称，也可以是十六进制颜色代码
+    param        {*} colors 颜色
+    param        {*} nodes 颜色位置，默认不提供，等间距
+    return       {*} c_map
+    example     : c_map = mk_cmap(['#C2B7F3','#B3BBF2','#B0CBF1','#ACDCF0','#A8EEED'])
+                c_map = mk_cmap(['aliceblue','skyblue','deepskyblue'],[0.0,0.5,1.0])
+    '''
+    if nodes is None:  # 采取自动分配比例
+        cmap_color = mpl.colors.LinearSegmentedColormap.from_list(
+            'mycmap', colors)
+    else:  # 按照提供比例分配
+        cmap_color = mpl.colors.LinearSegmentedColormap.from_list(
+            "mycmap", list(zip(nodes, colors)))
+    return cmap_color
+
+# ** 自制diverging型cmap，默认中间为白色
+
+
+def create_diverging(colors: list):
+    '''
+    func        : 自制cmap，双色，中间默认为白色；如果输入偶数个颜色，则中间为白，如果奇数个颜色，则中间色为中间色
+    description : colors可以是颜色名称，也可以是十六进制颜色代码
+    param        {*} colors
+    return       {*}
+    example     : diverging_cmap = mk_cmap_diverging(["#00c0ff", "#a1d3ff", "#DCDCDC", "#FFD39B", "#FF8247"])
+    '''
+    # 自定义颜色位置
+    n = len(colors)
+    nodes = np.linspace(0.0, 1.0, n + 1 if n % 2 == 0 else n)
+    newcolors = colors
+    if n % 2 == 0:
+        newcolors.insert(int(n / 2), '#ffffff')  # 偶数个颜色，中间为白色
+    cmap_color = mpl.colors.LinearSegmentedColormap.from_list(
+        "mycmap", list(zip(nodes, newcolors)))
+    return cmap_color
+
+# ** 根据RGB的txt文档制作色卡（利用Grads调色盘）
+
+
+def create_5rgb_txt(rgb_txt_filepath: str):  # 根据RGB的txt文档制作色卡/根据rgb值制作
+    '''
+    func        : 根据RGB的txt文档制作色卡
+    description : rgb_txt_filepath='E:/python/colorbar/test.txt'
+    param        {*} rgb_txt_filepath txt文件路径
+    return       {*} camp
+    example     : cmap_color=dcmap(path)
+    '''
+    with open(rgb_txt_filepath) as fid:
+        data = fid.readlines()
+    n = len(data)
+    rgb = np.zeros((n, 3))
+    for i in np.arange(n):
+        rgb[i][0] = data[i].split(',')[0]
+        rgb[i][1] = data[i].split(',')[1]
+        rgb[i][2] = data[i].split(',')[2]
+    max_rgb = np.max(rgb)
+    if max_rgb > 2:  # 如果rgb值大于2，则认为是0-255的值，需要归一化
+        rgb = rgb / 255.0
+    icmap = mpl.colors.ListedColormap(rgb, name='my_color')
+    return icmap
+
+
+if __name__ == '__main__':
+    # ** 测试自制cmap
+    colors = ['#C2B7F3', '#B3BBF2', '#B0CBF1', '#ACDCF0', '#A8EEED']
+    nodes = [0.0, 0.2, 0.4, 0.6, 1.0]
+    c_map = create_custom(colors, nodes)
+    show([c_map])
+
+    # ** 测试自制diverging型cmap
+    diverging_cmap = create_diverging(["#4e00b3", "#0000FF", "#00c0ff",
+                                       "#a1d3ff", "#DCDCDC", "#FFD39B", "#FF8247", "#FF0000", "#FF5F9E"])
+    show([diverging_cmap])
+
+    # ** 测试根据RGB的txt文档制作色卡
+    file_path = 'E:/python/colorbar/test.txt'
+    cmap_color = create_5rgb_txt(file_path)
+
+    # ** 测试将cmap转为list
+    out_cmap = extract_colors('viridis', 256)

oafuncs/oa_data.py

@@ -0,0 +1,205 @@
+#!/usr/bin/env python
+# coding=utf-8
+'''
+Author: Liu Kun && 16031215@qq.com
+Date: 2024-09-17 17:12:47
+LastEditors: Liu Kun && 16031215@qq.com
+LastEditTime: 2024-10-06 19:05:38
+FilePath: \\Python\\My_Funcs\\OAFuncs\\OAFuncs\\oa_data.py
+Description:  
+EditPlatform: vscode
+ComputerInfo: XPS 15 9510
+SystemInfo: Windows 11
+Python Version: 3.11
+'''
+
+
+
+import multiprocessing as mp
+from concurrent.futures import ThreadPoolExecutor
+
+import numpy as np
+from scipy.interpolate import griddata
+
+__all__ = ['interp_2d', 'interp_2d_parallel']
+
+# ** 高维插值函数，插值最后两个维度
+def interp_2d(target_x, target_y, origin_x, origin_y, data, method='linear'):
+    """
+    高维插值函数，默认插值最后两个维度，传输数据前请确保数据的维度正确
+    参数:
+    target_y (array-like): 目标经度网格 1D 或 2D
+    target_x (array-like): 目标纬度网格 1D 或 2D
+    origin_y (array-like): 初始经度网格 1D 或 2D
+    origin_x (array-like): 初始纬度网格 1D 或 2D
+    data (array-like): 数据 (*, lat, lon) 2D, 3D, 4D
+    method (str, optional): 插值方法，可选 'linear', 'nearest', 'cubic' 等，默认为 'linear'
+    返回:
+    array-like: 插值结果
+    """
+
+    # 确保目标网格和初始网格都是二维的
+    if len(target_y.shape) == 1:
+        target_x, target_y = np.meshgrid(target_x, target_y)
+    if len(origin_y.shape) == 1:
+        origin_x, origin_y = np.meshgrid(origin_x, origin_y)
+
+    dims = data.shape
+    len_dims = len(dims)
+    # print(dims[-2:])
+    # 根据经纬度网格判断输入数据的形状是否匹配
+
+    if origin_x.shape != dims[-2:] or origin_y.shape != dims[-2:]:
+        print(origin_x.shape, dims[-2:])
+        raise ValueError(
+            'Shape of data does not match shape of origin_x or origin_y.')
+
+    # 将目标网格展平成一维数组
+    target_points = np.column_stack((np.ravel(target_y), np.ravel(target_x)))
+
+    # 将初始网格展平成一维数组
+    origin_points = np.column_stack((np.ravel(origin_y), np.ravel(origin_x)))
+
+    # 进行插值
+    if len_dims == 2:
+        interpolated_data = griddata(origin_points, np.ravel(
+            data), target_points, method=method)
+        interpolated_data = np.reshape(interpolated_data, target_y.shape)
+    elif len_dims == 3:
+        interpolated_data = []
+        for i in range(dims[0]):
+            dt = griddata(origin_points, np.ravel(
+                data[i, :, :]), target_points, method=method)
+            interpolated_data.append(np.reshape(dt, target_y.shape))
+            print(f'Interpolating {i+1}/{dims[0]}...')
+        interpolated_data = np.array(interpolated_data)
+    elif len_dims == 4:
+        interpolated_data = []
+        for i in range(dims[0]):
+            interpolated_data.append([])
+            for j in range(dims[1]):
+                dt = griddata(origin_points, np.ravel(
+                    data[i, j, :, :]), target_points, method=method)
+                interpolated_data[i].append(np.reshape(dt, target_y.shape))
+                print(
+                    f'\rInterpolating {i*dims[1]+j+1}/{dims[0]*dims[1]}...', end='')
+        print('\n')
+        interpolated_data = np.array(interpolated_data)
+
+    return interpolated_data
+
+
+# ** 高维插值函数，插值最后两个维度，使用多线程进行插值
+# 在本地电脑上可以提速三倍左右，超算上暂时无法加速
+def interp_2d_parallel(target_x, target_y, origin_x, origin_y, data, method='linear'):
+    '''
+    param        {*} target_x 目标经度网格 1D 或 2D
+    param        {*} target_y 目标纬度网格 1D 或 2D
+    param        {*} origin_x 初始经度网格 1D 或 2D
+    param        {*} origin_y 初始纬度网格 1D 或 2D
+    param        {*} data 数据 (*, lat, lon) 2D, 3D, 4D
+    param        {*} method 插值方法，可选 'linear', 'nearest', 'cubic' 等，默认为 'linear'
+    return       {*} 插值结果
+    description : 高维插值函数，默认插值最后两个维度，传输数据前请确保数据的维度正确
+    example     : interpolated_data = interp_2d_parallel(target_x, target_y, origin_x, origin_y, data, method='linear')
+    '''
+    def interp_single2d(target_y, target_x, origin_y, origin_x, data, method='linear'):
+        target_points = np.column_stack(
+            (np.ravel(target_y), np.ravel(target_x)))
+        origin_points = np.column_stack(
+            (np.ravel(origin_y), np.ravel(origin_x)))
+
+        dt = griddata(origin_points, np.ravel(
+            data[:, :]), target_points, method=method)
+        return np.reshape(dt, target_y.shape)
+
+    def interp_single3d(i, target_y, target_x, origin_y, origin_x, data, method='linear'):
+        target_points = np.column_stack(
+            (np.ravel(target_y), np.ravel(target_x)))
+        origin_points = np.column_stack(
+            (np.ravel(origin_y), np.ravel(origin_x)))
+
+        dt = griddata(origin_points, np.ravel(
+            data[i, :, :]), target_points, method=method)
+        return np.reshape(dt, target_y.shape)
+
+    def interp_single4d(i, j, target_y, target_x, origin_y, origin_x, data, method='linear'):
+        target_points = np.column_stack(
+            (np.ravel(target_y), np.ravel(target_x)))
+        origin_points = np.column_stack(
+            (np.ravel(origin_y), np.ravel(origin_x)))
+
+        dt = griddata(origin_points, np.ravel(
+            data[i, j, :, :]), target_points, method=method)
+        return np.reshape(dt, target_y.shape)
+
+    if len(target_y.shape) == 1:
+        target_x, target_y = np.meshgrid(target_x, target_y)
+    if len(origin_y.shape) == 1:
+        origin_x, origin_y = np.meshgrid(origin_x, origin_y)
+
+    dims = data.shape
+    len_dims = len(dims)
+
+    if origin_x.shape != dims[-2:] or origin_y.shape != dims[-2:]:
+        raise ValueError('数据形状与 origin_x 或 origin_y 的形状不匹配.')
+
+    interpolated_data = []
+
+    # 使用多线程进行插值
+    with ThreadPoolExecutor(max_workers=mp.cpu_count()-2) as executor:
+        print(f'Using {mp.cpu_count()-2} threads...')
+        if len_dims == 2:
+            interpolated_data = list(executor.map(interp_single2d, [target_y], [
+                                     target_x], [origin_y], [origin_x], [data], [method]))
+        elif len_dims == 3:
+            interpolated_data = list(executor.map(interp_single3d, [i for i in range(dims[0])], [
+                                     target_y]*dims[0], [target_x]*dims[0], [origin_y]*dims[0], [origin_x]*dims[0], [data]*dims[0], [method]*dims[0]))
+        elif len_dims == 4:
+            interpolated_data = list(executor.map(interp_single4d, [i for i in range(dims[0]) for j in range(dims[1])], [j for i in range(dims[0]) for j in range(dims[1])], [
+                                     target_y]*dims[0]*dims[1], [target_x]*dims[0]*dims[1], [origin_y]*dims[0]*dims[1], [origin_x]*dims[0]*dims[1], [data]*dims[0]*dims[1], [method]*dims[0]*dims[1]))
+            interpolated_data = np.array(interpolated_data).reshape(
+                dims[0], dims[1], target_y.shape[0], target_x.shape[1])
+
+    interpolated_data = np.array(interpolated_data)
+
+    return interpolated_data
+
+
+if __name__ == '__main__':
+    import time
+
+    import matplotlib.pyplot as plt
+
+    # 测试数据
+    origin_x = np.linspace(0, 10, 11)
+    origin_y = np.linspace(0, 10, 11)
+    target_x = np.linspace(0, 10, 101)
+    target_y = np.linspace(0, 10, 101)
+    data = np.random.rand(11, 11)
+
+    # 高维插值
+    origin_x = np.linspace(0, 10, 11)
+    origin_y = np.linspace(0, 10, 11)
+    target_x = np.linspace(0, 10, 101)
+    target_y = np.linspace(0, 10, 101)
+    data = np.random.rand(10, 10, 11, 11)
+
+    start = time.time()
+    interpolated_data = interp_2d(target_x, target_y, origin_x, origin_y, data)
+    print(f'Interpolation time: {time.time()-start:.2f}s')
+
+    print(interpolated_data.shape)
+
+    # 高维插值多线程
+    start = time.time()
+    interpolated_data = interp_2d_parallel(
+        target_x, target_y, origin_x, origin_y, data)
+    print(f'Interpolation time: {time.time()-start:.2f}s')
+
+    print(interpolated_data.shape)
+    print(interpolated_data[0, 0, :, :].shape)
+    plt.figure()
+    plt.contourf(target_x, target_y, interpolated_data[0, 0, :, :])
+    plt.colorbar()
+    plt.show()

oafuncs/oa_down/__init__.py

@@ -0,0 +1,20 @@
+#!/usr/bin/env python
+# coding=utf-8
+'''
+Author: Liu Kun && 16031215@qq.com
+Date: 2024-09-17 16:09:20
+LastEditors: Liu Kun && 16031215@qq.com
+LastEditTime: 2024-10-14 18:12:12
+FilePath: \\Python\\My_Funcs\\OAFuncs\\OAFuncs\\oa_down\\__init__.py
+Description:  
+EditPlatform: vscode
+ComputerInfo: XPS 15 9510
+SystemInfo: Windows 11
+Python Version: 3.11
+'''
+
+# from .love_ocean import sign as love_ocean
+# from .meteorological_home import sign as meteorological_home
+
+from .hycom_3hourly import *
+from .refs_pdf import *