#!/usr/bin/env python
# coding: utf-8

# ### How does the imported datasets look like?
# Here we show how the impoted predictors and predictand datasets look like. We show for one example of the weather station (stuttgart) and predictor (tp: total precipitation)
# 1. read the csv file loaded as Weather station object using pandas 
# 2. demonstrate the structure of the ERA5 datasets used predictor sets for this exercise
# 

# In[3]:


# set paths to the datasets
stationdir = "C:/Users/dboateng/Desktop/Python_scripts/ESD_Package/examples/tutorials/data"
predictordir = "C:/Users/dboateng/Desktop/Datasets/ERA5/monthly_1950_2021/"
name_station= "stuttgart.csv"
filename_predictor = "tp_monthly.nc"


# In[2]:


import os
import pandas as pd
import xarray as xr


# In[4]:


# read the station data
df = pd.read_csv(os.path.join(stationdir, name_station))


# In[5]:


df.head(10)


# You can see that the data structure is very simple, first 4 rows contain the station information, and the rest is the Time and precipitation measurement

# In[6]:


#read the predictor
data = xr.open_dataset(os.path.join(predictordir, filename_predictor))


# In[7]:


data


# Here, the ERA5 dataset we loaded is a global datasets with the tp as the variable name, which is provided in the predictors list. We can even plot the global datasets for on time step

# In[9]:


data.tp.sel(time="1950-01-01").plot()