diff --git a/src/associators.py b/src/associators.py
index 9ef73a8e7da1684a975792cb8e2f735983197cb5..029849330179a7a7e58f454688b9013fc17835b9 100644
--- a/src/associators.py
+++ b/src/associators.py
@@ -1,3 +1,4 @@
+"""Contains associatiors (Nearest Neighbour etc.)"""
import numpy as np
class NNAssociator:
diff --git a/src/filters.py b/src/filters.py
index 622d3b7abae6ad625e4374a1f60e099a11b55147..a1d0e156b83a77229cc03db41b6c1a6df43d0e0f 100644
--- a/src/filters.py
+++ b/src/filters.py
@@ -1,3 +1,5 @@
+"""Contains filter implementations.
+"""
import numpy as np
import scipy.stats as stats
import jax
@@ -66,9 +68,9 @@ class EKF:
H = self.sensor_model['dhdx'](x.flatten())
Sk = H@P@H.T+self.sensor_model['R']
Kk = P@H.T@np.linalg.inv(Sk)
- x +=(Kk@res)
- P -= Kk@Sk@Kk.T
- return x, P
+ xf = x+Kk@res
+ Pf = P-Kk@Sk@Kk.T
+ return xf, Pf
class IMM:
def __init__(self, filters, sensor_model, transition_prob):
diff --git a/src/gaters.py b/src/gaters.py
index e84211d0e556b0f8a5c0333db2289024c2e3a4ac..01c8e4baf8b7afa06874663dda787af9cdd9aa31 100644
--- a/src/gaters.py
+++ b/src/gaters.py
@@ -1,3 +1,4 @@
+"""Contains gaters that validates measurements to state estimates."""
import numpy as np
class MahalanobisGater:
diff --git a/src/logic.py b/src/logic.py
index d7d12443cfe3aa8d75fdaf8a38ebb5d10935243f..8c6102d73b3653a55f6331b0054dc16955eabf41 100644
--- a/src/logic.py
+++ b/src/logic.py
@@ -1,3 +1,4 @@
+"""Contains track logic."""
import scipy.stats as stats
import numpy as np
diff --git a/src/models.py b/src/models.py
index bedb2647c5b4386c591d437250116374ca7b85a8..dc892fc38ea2a300594ea86bf2cc263483f42804 100644
--- a/src/models.py
+++ b/src/models.py
@@ -1,3 +1,4 @@
+"""Contains both sensor and motion models."""
import numpy as np
import jax.numpy as jnp
import jax
diff --git a/src/plotters.py b/src/plotters.py
index bf93c24b36032c94acece440f9567bac724f50c8..a0f08463474d9d298f90f3c404612ffa0a96f750 100644
--- a/src/plotters.py
+++ b/src/plotters.py
@@ -1,3 +1,4 @@
+"""Plotting functionality for the exercise sessions."""
import numpy as np
import matplotlib.pyplot as plt
@@ -63,26 +64,27 @@ def plot_result_ex2_2(result, trajs):
for key, T in trajs.items():
ax[1].plot(T[0, :], T[1, :], color='k', lw=2)
+ confirmed_id = [track['identity'] for track in result['confirmed_tracks']]
for track in result['tracks']:
x = np.vstack(track['x'])
t = np.hstack(track['t']).flatten()
assoc = np.hstack(track['associations']).flatten()
- if track in result['confirmed_tracks']:
+ if track['identity'] in confirmed_id:
ls = '-'
- l = ax[0].plot(t, track['identity']*np.ones(t.shape), ls=ls, markersize=3)[0]
- ax[0].plot(assoc, track['identity']*np.ones(assoc.shape), ls='', color=l.get_color(), marker='x', markersize=6)
- ax[1].plot(x[:, 0], x[:, 1], ls=ls, color=l.get_color(), lw=3)
+ l = ax[0].plot(t, [str(track['identity'])]*t.size, ls=ls, markersize=3)[0]
+ ax[0].plot(assoc, [str(track['identity'])]*assoc.size, ls='', color=l.get_color(), marker='x', markersize=6)
+ ax[1].plot(x[:, 0], x[:, 1], ls=ls, color=l.get_color(), lw=3, label='Track {}'.format(track['identity']))
else:
ls = '--'
ax[1].plot(x[:, 0], x[:, 1], ls=ls, lw=2)
-
ax[0].set_ylabel('Track identity')
ax[0].set_title('Confirmed tracks over time')
ax[0].set_xlabel('Time index, k')
- ax[1].plot(yx, yy, '.', color='k')
+ ax[1].plot(yx, yy, '.', color='k', label='Measurements')
ax[1].set_xlabel(r'$p_x$')
ax[1].set_ylabel(r'$p_y$')
ax[1].set_title('Measurements and measurement predictions + tracks')
+ ax[1].legend(loc=2)
# Plot the RMSE for the matched trajectories
for track_id, key in result['matches'].items():
@@ -122,23 +124,22 @@ def plot_result_ex2_24(result):
x = np.vstack(track['x'])
t = np.hstack(track['t']).flatten()
assoc = np.hstack(track['associations']).flatten()
-
if track['identity'] in confirmed_id:
ls = '-'
- l = ax[0].plot(t, track['identity']*np.ones(t.shape), ls=ls, markersize=3)[0]
- ax[0].plot(assoc, track['identity']*np.ones(assoc.shape), ls='', color=l.get_color(), marker='x', markersize=6)
- ax[1].plot(x[:, 0], x[:, 1], ls=ls, color=l.get_color(), lw=3)
+ l = ax[0].plot(t, [str(track['identity'])]*t.size, ls=ls, markersize=3)[0]
+ ax[0].plot(assoc, [str(track['identity'])]*assoc.size, ls='', color=l.get_color(), marker='x', markersize=6)
+ ax[1].plot(x[:, 0], x[:, 1], ls=ls, color=l.get_color(), lw=3, label='Track {}'.format(track['identity']))
else:
ls = '--'
ax[1].plot(x[:, 0], x[:, 1], ls=ls, lw=2)
-
ax[0].set_ylabel('Track identity')
ax[0].set_title('Confirmed tracks over time')
ax[0].set_xlabel('Time index, k')
- ax[1].plot(yx, yy, '.', color='k')
+ ax[1].plot(yx, yy, '.', color='k', label='Measurements')
ax[1].set_xlabel(r'$p_x$')
ax[1].set_ylabel(r'$p_y$')
ax[1].set_title('Measurements and measurement predictions + tracks')
+ ax[1].legend(loc=2)
ax[1].set_xlim([-2000, 2000])
ax[1].set_ylim([-21000, -17000])
return fig
diff --git a/src/sim.py b/src/sim.py
index be98ad38ffd4e3de61985b53e93f531ae14bcfd3..89735570255a79b63482b05e81e37316a2b0651e 100644
--- a/src/sim.py
+++ b/src/sim.py
@@ -1,3 +1,4 @@
+"""Simulation functionality (data generation)"""
import numpy as np
def generate_data(trajectories, sensor_model, clutter_model, rng=None):
diff --git a/src/trajectories.py b/src/trajectories.py
index f9c13739a5a3a0ed549907debb37317a62045523..f9b9781d14d726ebb9be717a63bda6c66eb8c0a5 100644
--- a/src/trajectories.py
+++ b/src/trajectories.py
@@ -1,3 +1,4 @@
+"""Contains exercise trajectories (deterministic)"""
import numpy as np
def get_ex1_trajectories():
diff --git a/src/utility.py b/src/utility.py
index dae556d95075e0380d7315b687aef75c79e1588c..1d969c0f93884fc8dacfb2f4a5b8cd090a8d458e 100644
--- a/src/utility.py
+++ b/src/utility.py
@@ -1,3 +1,4 @@
+"""Contains utility functionality."""
import copy
import numpy as np
import murty as murty_
@@ -44,6 +45,29 @@ def match_tracks_to_ground_truth(tracks, ground_truth):
return matches
def recreate_trajectories(hypothesis, marginalize=True):
+ """Recreates target trajectories from an MHT output (see src.trackers).
+
+ It takes a dict of hypothesis where each key corresponds to a certain time
+ step. Each such value then is a list of plausible hypothesis for that
+ particular time instance.
+
+ Parameters
+ ----------
+ hypothesis : dict
+ A dict with keys corresponding to specific time steps. Values are lists
+ of varying length of plausible hypothesis in each time step.
+ marginalize : bool
+ If True, it marginalizes over the hypothesis in each time step. Otherwise
+ returns the MAP estimate.
+
+ Returns
+ -------
+ list
+ A list of all tracks, tentative, confirmed and deleted.
+ list
+ A list of only confirmed tracks, even if they have died.
+
+ """
confirmed_tracks = dict()
tracks = dict()
for t, hyp_t in hypothesis.items():
@@ -109,10 +133,14 @@ def recreate_trajectories(hypothesis, marginalize=True):
if tracks[track_identity]['stage'] == 'confirmed':
if track_identity not in confirmed_tracks.keys():
confirmed_tracks[track_identity] = tracks[track_identity]
- return tracks.values(), confirmed_tracks.values()
+ return list(tracks.values()), list(confirmed_tracks.values())
def murty(C):
- """Algorithm due to Murty."""
+ """Algorithm due to Murty.
+
+ This particular implementation is taken from Jonatan Olofsson, see
+ https://github.com/jonatanolofsson/mht.
+ """
mgen = murty_.Murty(C)
while True:
ok, cost, sol = mgen.draw()
@@ -120,13 +148,38 @@ def murty(C):
return None
yield cost, sol
-# Save result
def save_result(filename, result):
+ """Save a tracking result.
+
+ Parameters
+ ----------
+ filename : str
+ A filename to save the result to
+ result : dict
+ A dictionary to save
+
+ """
for track in result['tracks']:
track.pop('filt', None) # Pop the filter object so it is only data that is saved.
np.save(filename, result)
def load_result(filename):
+ """Load a tracking result.
+
+ NOTE: THIS IS NOT SAFE LOADING. PICKLED OBJECTS MAY EXECUTE MALICIOUS CODE.
+ MAKE SURE TO VERIFY THAT YOU ARE LOADING SPECIFICALLY WHAT YOU WANT.
+
+ Parameters
+ ----------
+ filename : str
+ A filename to load
+
+ Returns
+ ----------
+ -
+ Returns whatever data is stored in the file.
+
+ """
if '.npy' not in filename:
filename += '.npy'
result = np.load(filename, allow_pickle=True)