from enum import unique

import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
import operator
import seaborn as sns
from scipy import stats
from mpl_toolkits.mplot3d import Axes3D
from os import path
from matplotlib.colors import LogNorm


from collections import Counter
import sys
import math
from sklearn.preprocessing import minmax_scale
# from matplotlib.colors import LogNorm
from sklearn import preprocessing
import argparse
import joblib
import os



headers = {"C-100003", "2018-07-27", "FOD Turbine",
"C-100005", "2018-06-25", "FOD Compressor",
"C-100004", "2018-06-07", "FOD Compressor",
"C-100002", "2018-06-08", "FOD Compressor",
"C-100001", "2018-07-18", "FOD Compressor"}

data = pd.read_csv("Multi_P1FWM_turbo_failure_deviation_ano.csv", header=0)

##############################################################

parser = argparse.ArgumentParser()
parser.add_argument('-a', "--anonymous", action='store_true', default=False, help='Hide axes')
parser.add_argument('-t', "--truck", type=str, default=None, help='PLot one truck ID')
parser.add_argument('-f', "--filename", type=str, default="Multi_P1FWM_turbo_failure_deviation_ano.csv", help='CSV filename')
parser.add_argument('-n', "--normalise", action='store_true', default=False, help='Normalise')
parser.add_argument("--max_pics", type=int, default=16, help='Max histograms on a plot')
parser.add_argument("--cols", type=int, default=4, help='Number of histogram columns in plot')
parser.add_argument("--rows", type=int, default=None, help='Number of rows to read from CSV file')
parser.add_argument("--top", type=int, default=None, help='Top n')
parser.add_argument("--start", type=int, default=0, help='Start histogram')
parser.add_argument("--trucks", type=str, default=None, help='Multiple chassis IDs, comma seperated')


args = parser.parse_args()
print("Reading data...", end="", flush=True)
if not os.path.exists("df_train.pickle"):
    lines = []  # list(range(1, 100000))
    df_train = pd.read_csv(args.filename,
                           sep=";", dtype={'VEHICL_ID': str},
                           # skiprows=lines,
                           nrows=args.rows
                           )
    print( "pickling...", end="", flush=True )
    joblib.dump( df_train,"df_train.pickle" )
    df_train.to_pickle("df_train.pickle")
else:
    print("from pickle...", end="", flush=True)
    df_train = pd.read_pickle("df_train.pickle")
print("Ready")
print(df_train.shape)
print(df_train.head(2))
print(data.shape)



c = data[data['VEHICLE_ID'] == 1]
d = data[data['VEHICLE_ID'] == 2]
e = data[data['VEHICLE_ID'] == 3]
f = data[data['VEHICLE_ID'] == 4]
g = data[data['VEHICLE_ID'] == 5]

m1 = c.SEND_DATETIME
m2 = d.SEND_DATETIME
m3 = e.SEND_DATETIME
m4 = f.SEND_DATETIME
m5 = g.SEND_DATETIME

m1 = pd.to_datetime(m1)
m2 = pd.to_datetime(m2)
m3 = pd.to_datetime(m3)
m4 = pd.to_datetime(m4)
m5 = pd.to_datetime(m5)


n=m1.dt.year
print(m1)
A=g.sample(n=5)
B=g.drop(['VEHICLE_ID','SEND_DATETIME','PARAMETER_CODE','T_CHASSIS','Y_INDEX_2_X_INDEX_13_VALUE'] ,axis=1)
print(B.shape)
fig, axes = plt.subplots(nrows=1, ncols=g.shape[0])
fig.set_figheight(20)
fig.set_figwidth(20)
num_pics = data.shape[0]
if num_pics > args.max_pics:
      st = num_pics - args.max_pics
      num_pics = args.max_pics
print(num_pics)

lable="SEND_DATETIME"
cols = args.cols
rows = int(num_pics/cols)+ 1*((num_pics % cols)!=0)
sp=1
pc=0
fig = plt.figure(figsize=(cols*3, rows*3)) #plt.figure(figsize=(sz,sz))
plt.subplots_adjust( hspace=0.7, wspace=0.5 )

for index, row in B.iterrows():
    ax = fig.add_subplot(rows, cols, sp)

    print(type(row))
    xx = row.iloc[:19]
    yy = row.iloc[19:]

    drop_reshape1 = row.values.reshape(20, 20)
    im = drop_reshape1.astype(int)
    im = np.flipud(im)  # rot90(im) #flipud(im) #rot90()

    if args.normalise:
        _min = 0
        _max = 1
        im += -(np.min(im))
        im /= np.max(im) / (_max - _min)
    #ax.append( fig.add_subplot(20, 20, index+1) )

    im_masked = np.ma.masked_where(im == 0, im)
    plt.imshow(im_masked, interpolation='none')

    ax.set_aspect('equal')
    ax.get_xaxis().set_ticks([0, 19])
    ax.get_yaxis().set_ticks([])
    #
    ax.set_xticks(np.arange(-.5, 19, 1), minor=True);
    ax.set_yticks(np.arange(-.5, 19, 1), minor=True);
    ax.grid(which='minor', color='w', linestyle='-', linewidth=1)
    #
    if not args.anonymous:
        ax.set_xlabel("engine speed")
        ax.set_ylabel("engine torque")
        plt.colorbar(orientation='vertical', ax=ax, format='%.1f', fraction=0.0408, pad=0.04)
        # plt.clim(0, 10);
    else:
        plt.colorbar(orientation='vertical', ax=ax, ticks=[])
    sp += 1
    if not args.anonymous:
        fig.suptitle('VEHICLE_ID'+ ": " + str('SEND_DATETIME') + ' P1FWM')  # Training data '+str(st)+" +"+str(nn) )
    fn = "training_" + str('SEND_DATETIME') + "_" + str(index) + "+" + str(pc)
    if args.normalise:
        fn += "_N"
    fn += ".png"
    print("Saving", fn)
    if os.path.exists(fn):
        os.remove(fn)
    fig.savefig(fn, dpi=288)

plt.show()
    #ax.set_aspect('equal')
    #ax.get_xaxis().set_ticks([0, 19])
    #ax.get_yaxis().set_ticks([])
    #
    #ax.set_xticks(np.arange(-.5, 19, 1), minor=True);
    #ax.set_yticks(np.arange(-.5, 19, 1), minor=True);
    #ax.grid(which='minor', color='w', linestyle='-', linewidth=1)
    #
    #if not args.anonymous:
        #ax.set_xlabel("engine speed")
        #ax.set_ylabel("engine torque")
        #plt.colorbar(orientation='vertical', ax=ax, format='%.1f', fraction=0.0408, pad=0.04)
        # plt.clim(0, 10);
    #else:
        #plt.colorbar(orientation='vertical', ax=ax, ticks=[])

    #print(drop_reshape1.shape)
    #x = drop_reshape1[:,19]
    #y = drop_reshape1[19,:]
    #fig2 = plt.figure()
    #plt.hist2d(x, y, bins=100)
    #plt.xlabel('x')
    #plt.ylabel('y')
    #cbar = plt.colorbar()
    #cbar.ax.set_ylabel('Counts')
#heat plots

    #df = pd.DataFrame(drop_reshape1)
    #df.columns = df.columns + 1
    #df.index = df.index + 1
    #f, ax = plt.subplots(figsize=(10, 6))
    #corr = df.corr()
    #hm = sns.heatmap(round(corr, 2), annot=True, ax=ax, cmap="coolwarm", fmt='.2f',linewidths=.05)
    #f.subplots_adjust(top=0.93)
    #t = f.suptitle('Wine Attributes Correlation Heatmap', fontsize=14)

#img plots

#histogram plots
    #plt.hist(drop_reshape1)
    #plt.title(m5[index])
    #plt.show()
    #fig.savefig("dataname_"+str(index)+".png")


#drop = m.drop(['VEHICLE_ID','SEND_DATETIME','PARAMETER_CODE','T_CHASSIS','Y_INDEX_2_X_INDEX_13_VALUE'] ,axis=1)
#drop_array=np.array(B)
#drop_reshape = drop_array.reshape(20,20)
#plt.title('2015-06-04 19:42:43')
#plt.hist(drop_reshape)
#plt.show()

#a = np.array(m)
#b = np.array(l)
#d = np.array(data)
#sort = sorted(data,key=operator.itemgetter(1))

#s.plot()
#plt.savefig('vehicle id ')
#print(l)
#print(m)
#v=data.ix[0:,['SEND_DATETIME','Y_INDEX_1_X_INDEX_1_VALUE']]
#print (grouped)
#data = data[data['VEHICLE_ID'] == 1]
#data = data.set_index(['SEND_DATETIME'])
#data.plot()
#plt.show()
# Assign colors for each airline and the names
# Make the histogram using a list of lists
# Normalize the flights and assign colors and names
#plt.hist([x1, x2, x3, x4, x5], bins=int(180 / 15), normed=True,
         #color=colors, label=names)
# Plot formatting
#g= np.reshape(a, (np.product(a.shape),))
#k= np.reshape(m, (np.product(m.shape),))

#print(g.shape)
#plt.ploy(a,)
#plt.show()

#BY FAR BEST PLOTS
# use the function regplot to make a scatterplot
#sns.regplot(x=data["VEHICLE_ID"], y=data["Y_INDEX_1_X_INDEX_1_VALUE"])
#plt.show()

# Without regression fit:
#sns.regplot(x=data["VEHICLE_ID"], y=data["Y_INDEX_1_X_INDEX_1_VALUE"], fit_reg=False)
#plt.show()
#BY FAR BEST PLOTS

#fig = plt.figure()
#ax = fig.add_subplot(111, projection='3d')
#ax.scatter(data['VEHICLE_ID'], data['Y_INDEX_1_X_INDEX_1_VALUE'], data['C-100002'], c='skyblue', s=60)
#ax.view_init(30, 185)