experimental_tracks

Extract experimental trajectories.

This module contains functions used for the extraction of real experimental trajectories (stored as MDF or CSV file format).

experimental_tracks.extract_1_csv(folder_name, parms, filename='tracks.csv')

Extracts experimental trajectories from a single CSV file and returns a pandas DataFrame which stores the trajectories.

Parameters:

  • folder_name (str) – Name of the folder containing the trajectory file.

  • parms (dict) – Stored parameters containing global variables and instructions.

  • filename (str) – File which stores trajectory coordinates extracted from the tracking. [Defaults: ‘tracks.csv’]

Returns:

All extracted trajectories as a dataframe with keys: x, y, frame, data_folder, track_id.

Return type:

pd.DataFrame

experimental_tracks.extract_1_mdf(folder_name, parms, filename='tracks.simple.mdf')

Extracts experimental trajectories from a single MDF file and returns a pandas DataFrame which stores the trajectories.

Parameters:

  • folder_name (str) – Name of the folder containing the trajectory file.

  • parms (dict) – Stored parameters containing global variables and instructions.

  • filename (str) – File which stores trajectory coordinates extracted from the tracking. [Defaults: ‘tracks.simple.mdf’]

Returns:

All extracted trajectories as a dataframe with keys: x, y, frame, data_folder, track_id.

Return type:

pd.DataFrame

experimental_tracks.extract_all_tracks(parms)

Extracts the trajectories from several MDF (or CSV) files, fills the gaps and return a unique pandas DataFrame containing tracks.

Parameters:

parms (dict) – Stored parameters containing global variables and instructions.

Returns:

All extracted trajectories as a dataframe with keys: x, y, frame, data_folder, track_id.

Return type:

pd.DataFrame

experimental_tracks.fill_gaps(parms, track_df)

Fill the gap in trajectory frames by taking an intermediate point and splits the trajectory into two separate tracks for gaps larger than 1 frame.

Parameters:

  • parms (dict) – Stored parameters containing global variables and instructions.

  • track_df (pd.DataFrame) – Dataframe containing all extracted trajectories with keys: x, y, frame, data_folder, track_id.

Returns:

The updated dataframe with filled gaps or splitted tracks.

Return type:

pd.DataFrame

experimental_tracks.fill_gaps_l1(track)

Fills the 1-length gaps with an intermediate point having an additional bias (to add randomness).

Parameters:

track (pd.DataFrame) – Given track which its 1-length gaps will be filled with an appropriate generated point.

Returns:

(The updated track with 1-length gaps filled, the number of gaps filled for this given track)

Return type:

(pd.DataFrame, int)

experimental_tracks.predict_states(track_df, model, parms)

Predicts the states for each trajectory using a trained model and saves the resulting state prediction as a csv file.

Parameters:

  • track_df (pd.DataFrame) – Name of the folder containing the trajectory file.

  • model (keras model) – Trained neural network.

  • parms (dict) – Stored parameters containing global variables and instructions.