Python package for manipulating gpx files and easily convert gpx to other different formats.
- Free software: MIT license
- Documentation: https://gpx-converter.readthedocs.io.
- You need to convert GPX to other formats
- You need to convert other formats like csv to GPX
- You want to interpolate the GPX coordinates
- High level of abstraction
- Stable API
- easy to use and to extend
I decided to create this project because I had gpx data that I needed to manipulate. I searched for a python package for this but I did not find what I was looking for, therefore I created the gpx-converter package to make gpx files manipulation very easy. Furthermore, the package contains methods for applying interpolation on the gpx data. This feature was very helpful for me since I also needed to interpolate the gpx data and convert it to csv. Feel free to contribute or to give me feedback anytime :)
- Convert gpx files to other formats such as csv, numpy arrays, dataframes, excel and json
- Convert csv files to gpx
- Apply interpolation on the gpx data
$ pip install gpx-converter
from gpx_converter import Converter
Just read the gpx to dictionary
dic = Converter(input_file='your_input.gpx').gpx_to_dictionary(latitude_key='latitude', longitude_key='longitude')
# now you have a dictionary and can access the longitudes and latitudes values from the keys
latitudes = dic['latitude']
longitudes = dic['longitude']
Convert GPX to other formats
- Convert from gpx to csv:
Converter(input_file='your_input.gpx').gpx_to_csv(output_file='your_output.csv')
- Convert from gpx to excel sheets:
Converter(input_file='your_input.gpx').gpx_to_excel(output_file='your_output.xlsx')
- Convert from gpx to json:
Converter(input_file='your_input.gpx').gpx_to_json(output_file='your_output.json)
- Convert gpx file to dataframe:
df = Converter(input_file='your_input.gpx').gpx_to_dataframe()
- Convert gpx file to numpy array:
np_array = Converter(input_file='your_input.gpx').gpx_to_numpy_array()
Now convert other formats to GPX
- csv to gpx
Converter(input_file='your_input.csv').csv_to_gpx(lats_colname='column_name_of_latitudes',
longs_colname='column_name_of_longitudes',
output_file='your_output.gpx')
- excel to gpx
Converter(input_file='your_input.xlsx').excel_to_gpx(lats_colname='column_name_of_latitudes',
longs_colname='column_name_of_longitudes',
output_file='your_output.gpx')
- dataframe to gpx (notice that the method is static)
Converter.dataframe_to_gpx(input_df=your_df,
lats_colname='column_name_of_latitudes',
longs_colname='column_name_of_longitudes',
output_file='your_output.gpx')
- json to gpx
Converter(input_file='your_input.json').json_to_gpx(input_df=your_df,
lats_colname='column_name_of_latitudes',
longs_colname='column_name_of_longitudes',
output_file='your_output.gpx')
- Automate the conversion of multiple csv file to gpx:
Converter.convert_multi_csv_to_gpx(dirpath='your_directory/')
- Apply spline interpolation on gpx file (you need to install scipy for this to work):
interpolated_coordinates = Converter.spline_interpolation(cv=your_array_of_control_vertices)
Alternatively you can use the gpx_converter directly from terminal. You just need to pass the function, input file and output file as arguments.
- function: the conversion method you want to use. For example "gpx_to_csv"
- input file: path to your input file
- output file: path to your output file
$ gpx_converter --function "gpx_to_csv" --input_file "home/your_input.gpx" --output_file "home/your_output.csv"
or maybe you prefer the short version
$ gpx_converter -func "gpx_to_csv" -in "home/your_input.gpx" -out "home/your_output.csv"
Check this article to know more about gpx files and how to use the gpx-converter package. https://medium.com/p/57da00bd36fc/edit
Contributions are always welcome. Make sure you check the guidlines first https://gpx-converter.readthedocs.io/en/latest/contributing.html