julian_to_gregorian¶
-
hydrostats.data.julian_to_gregorian(dataframe, frequency=None, inplace=False)¶ Converts the index of the merged dataframe from julian float values to gregorian datetime values.
Parameters: - dataframe (Pandas DataFrame) – A DataFrame with an index of type float
- frequency (string) – Optional. Sometimes when converting from julian to gregorian there will be rounding errors due to the inability of computers to store floats as perfect decimals. Providing the frequency will automatically attempt to round the dates. A list of all the frequencies pandas provides is found here. Common frequencies include daily (“D”) and hourly (“H”).
- inplace (bool) – Default False. If True, will modify the index of the dataframe in place rather than creating a copy and returning the copy. Use when the time series are very long and making a copy would take a large amount of memory
Returns: A pandas DataFrame with gregorian index.
Return type: Pandas DataFrame
Examples
>>> import pandas as pd >>> import hydrostats.data as hd >>> import numpy as np
>>> # The julian dates in an array >>> julian_dates = np.array([2444239.5, 2444239.5416666665, 2444239.5833333335, 2444239.625, >>> 2444239.6666666665, 2444239.7083333335, 2444239.75, >>> 2444239.7916666665, 2444239.8333333335, 2444239.875]) >>> # Creating a test dataframe >>> test_df = pd.DataFrame(data=np.random.rand(10, 2), # Random data in the columns >>> columns=("Simulated Data", "Observed Data"), >>> index=julian_dates) >>> test_df Simulated Data Observed Data 2.444240e+06 0.764719 0.126610 2.444240e+06 0.372736 0.141392 2.444240e+06 0.008645 0.686477 2.444240e+06 0.656825 0.480444 2.444240e+06 0.555247 0.869409 2.444240e+06 0.643896 0.549590 2.444240e+06 0.242720 0.799617 2.444240e+06 0.432421 0.185760 2.444240e+06 0.694631 0.136986 2.444240e+06 0.700422 0.390415
>>> # Making a new df with gregorian index >>> test_df_gregorian = hd.julian_to_gregorian(test_df) >>> test_df_gregorian Simulated Data Observed Data 1980-01-01 00:00:00.000000 0.585454 0.457238 1980-01-01 01:00:00.028800 0.524764 0.083464 1980-01-01 01:59:59.971200 0.516821 0.416683 1980-01-01 03:00:00.000000 0.948483 0.553874 1980-01-01 04:00:00.028800 0.492280 0.232901 1980-01-01 04:59:59.971200 0.527967 0.296395 1980-01-01 06:00:00.000000 0.650018 0.212802 1980-01-01 07:00:00.028800 0.585592 0.802971 1980-01-01 07:59:59.971200 0.448243 0.665814 1980-01-01 09:00:00.000000 0.137395 0.201721
>>> # Rounding can be applied due to floating point inaccuracy >>> test_df_gregorian_rounded = julian_to_gregorian(test_df, frequency="H") # Hourly Rounding Frequency >>> test_df_gregorian_rounded Simulated Data Observed Data 1980-01-01 00:00:00 0.309527 0.938991 1980-01-01 01:00:00 0.872284 0.497708 1980-01-01 02:00:00 0.168046 0.225845 1980-01-01 03:00:00 0.954494 0.275607 1980-01-01 04:00:00 0.875885 0.194380 1980-01-01 05:00:00 0.236849 0.992770 1980-01-01 06:00:00 0.639346 0.029808 1980-01-01 07:00:00 0.855828 0.903927 1980-01-01 08:00:00 0.638805 0.916124 1980-01-01 09:00:00 0.273430 0.443980
>>> # The DataFrame can also be modified in place, increasing efficiency with large time series >>> julian_to_gregorian(test_df, inplace=True, frequency="H") >>> test_df Simulated Data Observed Data 1980-01-01 00:00:00 0.309527 0.938991 1980-01-01 01:00:00 0.872284 0.497708 1980-01-01 02:00:00 0.168046 0.225845 1980-01-01 03:00:00 0.954494 0.275607 1980-01-01 04:00:00 0.875885 0.194380 1980-01-01 05:00:00 0.236849 0.992770 1980-01-01 06:00:00 0.639346 0.029808 1980-01-01 07:00:00 0.855828 0.903927 1980-01-01 08:00:00 0.638805 0.916124 1980-01-01 09:00:00 0.273430 0.443980