distrib

import os
import pandas as pd
import numpy as np

from matplotlib import pyplot as plt, gridspec
from matplotlib.backends.backend_pdf import PdfPages
from collections import OrderedDict
from datetime import datetime, timedelta


from IPython.core.interactiveshell import InteractiveShell
InteractiveShell.ast_node_interactivity = 'all'

wkdir = 'C:/Users/MAR8RNG/BOSCH-PROJECTS/1-Humidity/python/condensed_CSVs/'
bse_prm = 'basic/'

os.chdir(wkdir+bse_prm)

csv_list = os.listdir(wkdir+bse_prm)

###there are issues when resampling data as the name of the months are removed, consider
###that part to be part of the work after the CSV has been made instead 'edit RawDataCond

for csv in csv_list:    
    imei=csv[:-4]
    imei
    with open(csv) as f:
        
        df=pd.read_csv(f, parse_dates=['DateTime'], index_col='DateTime')
        
        #duplicate df['dVWC] column
        df['Evap']=df['dVWC']
        
        #resampling requires index to be of type 'datetime'--done at import
        #Hourly based resampling with 'Evap' column summed and the rest mean-ed
        df=df.resample('H').agg({'Voltage':'mean', 'IMEI':'mean', 
                              'DigiTemp':'mean','HM1':'mean', 
                              'T1':'mean','VWC':'mean','dVWC':'mean',
                              'Evap':'sum'})
        
        #Unsure why this step has to be taken: reset index, re-establish to_datetime
        #in order to insert the 'string' columns removed by the resampling
        
        df.reset_index(inplace=True)
        df['DateTime'] = pd.to_datetime(df['DateTime'])
        
        #These are to simplify indexing
        df.insert(1, column="Month", value = df['DateTime'].dt.strftime('%b'))
        df.insert(2, column="Week", value = df['DateTime'].dt.week)
        df.insert(3, column="Day", value = df['DateTime'].dt.strftime('%a-%d'))
        
        #For iterating over months and obtaining weeks
        mon_list = df['Month'].unique()
        df.set_index(['Month'], drop=False, inplace=True)
        
        mon_week = OrderedDict()
        for m in mon_list:
            if m not in mon_week:
                mon_week[m] = df.loc[m]['Week'].unique()
        
        #Month-Week (e.g. Dec-52) Selection
        df.set_index(['Month', 'Week'], inplace=True)
        df.sort_index(inplace=True)
        
        ##Dataframe
        'df.head()'
        df.head()
        
        #dailydf
        #resample at hourly interval so that that 'Evap' is the hourly difference
        #at each observation
        
        #take the daily averages of evaporation
        
        #and make a monthly plot for this average
        
        #df.plot[x='Datetime', y='Evap']
        
        #Save figs as PDFs

        with PdfPages(wkdir+'figures/'+imei+'.pdf') as pdf:
            
            #Montly-Weekly plots
            for m in mon_list:
                
                #create figure
                fig = plt.figure()
                
                #number of weekly plots in fig
                nweeks = len(mon_week[m])

                #create grid in for figure
                #gs = gridspec.GridSpec(1, 2, width_ratios=[3, 1]) 
                grid = gridspec.GridSpec(3, 3, height_ratios=[5,3,3])
                
                #plot monthly and position in grid
                month=plt.subplot(grid[0,:])
                
               
                df.loc[m].groupby('Day', sort=False).VWC.mean().plot(figsize=(16,9),ax=month,rot=45,legend=True,style='r')
                df.loc[m].groupby('Day', sort=False).VWC.max().plot(figsize=(16,9),ax=month,rot=45,legend=True, style='ro')
                df.loc[m].groupby('Day', sort=False).Evap.mean().plot(figsize=(16,9),ax=month,rot=45,legend=True,style='b')
                df.loc[m].groupby('Day', sort=False).Evap.max().plot(figsize=(16,9),ax=month,rot=45,legend=True, style='bo')
                #df.loc[m].groupby('Day', sort=False).Evap.min().plot(figsize=(16,9),ax=month, rot=45, legend=True)
                #df.loc[m].groupby('Day', sort=False).Evap.max().plot(figsize=(16,9),ax=month, rot=45, legend=True)
                month.legend(['VWC-mean', 'VWC-max', 'Evap-mean', 'Evap-max'])
                n=5
                i=1
                
        
                for w in mon_week[m]:
                    #plot weekly and position in grid
                    week=plt.subplot(grid[i,5%n])
                    
                    df.loc[m,w].groupby('Day', sort=False).Evap.mean().plot(ax=week,
                                                                           rot=45, grid=True)
                    #df.hist()
                    #sets up week grid positions
                    n-=1
                    if n==2:
                        i=2
                        n=5
       
                ####don't forget to add labels####
                ####graph the moving average, median, i.e smoothing####
                ####graph all using the same y-tick window####
                
                plt.suptitle(imei, y=1)
                grid.tight_layout(fig)
                pdf.savefig(fig)
                plt.close('all')
        break