This repository provides files and notebooks for training simple models for predicting hiking times. More precisely, they predict the Moving time, Stopped time and duration of a hike. The moving time does not include stopping or pausing. When using own records of GPX tracks for training, the models can estimate personalized hiking times.

The estimates are based on simple statistics about GPX tracks, GPX tracks themselves or a combination of the two. The statistical features include the length of tracks, the total elevation gain and the total elevation loss. In order to be able to get good predictions based on a relatively small number of GPX tracks, the latter are split into smaller pieces which are used as inputs for the model. The same preprocessing steps are applied during inference. A dataset of 40-50 GPX tracks allows to train models that predict hiking times better than standard formulas.

The repository includes files for extracting data from tracks, training models and making predictions.

Run

conda env create -n hikingTimePredictor

and activate environment hikingTimePredictor by typing

source activate hikingTimePredictor.

Install dependencies via

pip install -r requirements.txt.

Run

python prepare_data.py base_folder filter_key

where base_folder is the folder in which the program searches recursively for GPX tracks (with file names ending on gpx). It filters for files that contain filter_key in their path, which could for example be Hiking.

Open the notebook hikingTimeRegression_v1.ipynb (or v2, v3) and follow the steps described there.

Run

python inference.py input_file model_type

where input_file is the name of the GPX track for which hiking times should be estimated. model_type is 'simple' for v1, 'recurrent' for v2 and 'mixed' for v3.

MIT, see LICENSE for more information