# 导入 pandas 读取数据
import pandas as pd
import numpy as np

# 读取数据
data = pd.read_csv("./data/peach_spectra_brix.csv")

# m * n 
print("数据矩阵 data.shape：",data.shape)

# 50个样本， 600个 波段 第一列是 桃子糖度值 需要分离开
X = data.values[:,1:] 

from pretreatment import Pretreatment as pre
p = pre()

# 该方法为快速示例 而编写 
# 测试用例 图片名 波段起始点 波段间距
p.PlotSpectrum(X, '演示', 0, 5).show()
## 为保证数据可以正确绘制 请将 矩阵转化为 numpy.ndarray 格式
type(X)

msc = p.msc(X)
p.PlotSpectrum(msc, 'msc', 0, 5).show()

snv = p.snv(X)
p.PlotSpectrum(snv, 'snv', 0, 5).show()

# 此处参数为参考 具体 后续 详述
sg = p.SG(X, 4*5+1,2*3,2)
p.PlotSpectrum(sg, 'sg', 0, 5).show

move_avg = p.move_avg(X)
p.PlotSpectrum(move_avg, 'move_avg', 0, 5).show

D1 = p.D1(X)
p.PlotSpectrum(D1, 'D1', 0, 5).show

D2 = p.D2(X)
p.PlotSpectrum(D2, 'D2', 0, 5).show

wave = p.wave(X)
p.PlotSpectrum(wave, 'wave', 0, 5).show

mean_centralization = p.mean_centralization(X)
p.PlotSpectrum(mean_centralization, 'mean_centralization', 0, 5).show

standardlize = p.standardlize(X)
p.PlotSpectrum(standardlize, 'standardlize', 0, 5).show

max_min_normalization = p.max_min_normalization(X)
p.PlotSpectrum(max_min_normalization, 'max_min_normalization', 0, 5).show

vector_normalization = p.vector_normalization(X)
p.PlotSpectrum(vector_normalization, 'vector_normalization', 0, 5).show