Add Multiple Model Framework#1257
Open
jswright-dstl wants to merge 29 commits intomainfrom
Open
Conversation
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #1257 +/- ##
==========================================
+ Coverage 91.92% 94.45% +2.53%
==========================================
Files 233 245 +12
Lines 16843 17053 +210
Branches 2372 2397 +25
==========================================
+ Hits 15483 16108 +625
+ Misses 1006 636 -370
+ Partials 354 309 -45
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
sdhiscocks
requested changes
Apr 9, 2026
Member
sdhiscocks
left a comment
sdhiscocks
left a comment
There was a problem hiding this comment.
Thanks @jswright-dstl
Doc strings are required for new classes.
It would be useful to have a brief explanation/summary for each algorithm in the tutorial, and possibly a metric for performance comparison.
Comment on lines
+10
to
+13
| class Augmentor(Base): | ||
| transition_probabilities: TransitionMatrix = Property(doc="TPM") | ||
| transition_models: Sequence[TransitionModel] = Property(doc="List of transition models") | ||
| histories: int = Property(doc="Depth of history to be stored") |
Member
There was a problem hiding this comment.
Could do with complete doc strings
| from ..augmentor.base import Augmentor | ||
|
|
||
|
|
||
| class IdentityAugmentor(Augmentor): |
Comment on lines
+29
to
+33
| try: | ||
| (isinstance(states[0][0], WeightedGaussianState)) | ||
| except TypeError or IndexError: | ||
| print(len(states)) | ||
|
|
Member
There was a problem hiding this comment.
Legacy debug?
Suggested change
| try: | |
| (isinstance(states[0][0], WeightedGaussianState)) | |
| except TypeError or IndexError: | |
| print(len(states)) |
Comment on lines
+42
to
+55
| if isinstance(states[0], ExpandedModelAugmentedWeightedGaussianState): | ||
| pass | ||
| else: | ||
| Likelihood_j = [] | ||
| for i in range(len(states)): | ||
| Likelihood_j.append(mvn.pdf( | ||
| states[i].hypothesis.measurement.state_vector.T, | ||
| states[i].hypothesis.measurement_prediction.mean.ravel(), | ||
| states[i].hypothesis.measurement_prediction.covar)) | ||
| c_j = self.transition_probabilities[states[0]].T @ states.weights | ||
| weights = Likelihood_j * c_j.ravel() | ||
| weights = weights / np.sum(weights) | ||
| for i in range(len(states)): | ||
| states[i].weight = weights[i] |
Member
There was a problem hiding this comment.
Suggested change
| if isinstance(states[0], ExpandedModelAugmentedWeightedGaussianState): | |
| pass | |
| else: | |
| Likelihood_j = [] | |
| for i in range(len(states)): | |
| Likelihood_j.append(mvn.pdf( | |
| states[i].hypothesis.measurement.state_vector.T, | |
| states[i].hypothesis.measurement_prediction.mean.ravel(), | |
| states[i].hypothesis.measurement_prediction.covar)) | |
| c_j = self.transition_probabilities[states[0]].T @ states.weights | |
| weights = Likelihood_j * c_j.ravel() | |
| weights = weights / np.sum(weights) | |
| for i in range(len(states)): | |
| states[i].weight = weights[i] | |
| if not isinstance(states[0], ExpandedModelAugmentedWeightedGaussianState): | |
| Likelihood_j = [] | |
| for i in range(len(states)): | |
| Likelihood_j.append(mvn.pdf( | |
| states[i].hypothesis.measurement.state_vector.T, | |
| states[i].hypothesis.measurement_prediction.mean.ravel(), | |
| states[i].hypothesis.measurement_prediction.covar)) | |
| c_j = self.transition_probabilities[states[0]].T @ states.weights | |
| weights = Likelihood_j * c_j.ravel() | |
| weights = weights / np.sum(weights) | |
| for i in range(len(states)): | |
| states[i].weight = weights[i] |
Comment on lines
+56
to
+58
| else: | ||
| pass | ||
|
|
Comment on lines
+24
to
+26
| def __init__(self, *args, **kwargs): | ||
| super().__init__(*args, **kwargs) | ||
|
|
Member
There was a problem hiding this comment.
Suggested change
| def __init__(self, *args, **kwargs): | |
| super().__init__(*args, **kwargs) |
| elif len(self.transition_probabilities[states[0]].ravel())**2 == len(states.weights): | ||
| m_ij_weights = ( | ||
| self.transition_probabilities.transition_matrices[1].ravel() * | ||
| (states.weights/np.sum(states.weights)))*np.random.rand(len(states.weights)) |
Member
There was a problem hiding this comment.
Would be good to be able to seed random for this, like models, etc..
Comment on lines
+31
to
+32
| self.transition_probabilities[states[0]] @ | ||
| (states.weights/np.sum(states.weights))) |
Member
There was a problem hiding this comment.
Suggested change
| self.transition_probabilities[states[0]] @ | |
| (states.weights/np.sum(states.weights))) | |
| self.transition_probabilities[states[0]] | |
| @ (states.weights/np.sum(states.weights))) |
| from ..base import Base, Property | ||
|
|
||
|
|
||
| class TransitionMatrix(Base): |
Comment on lines
+34
to
+36
| from stonesoup.models.transition.linear import (CombinedLinearGaussianTransitionModel as CLGTM, | ||
| ConstantVelocity as CV, | ||
| KnownTurnRate as CT) |
Contributor
There was a problem hiding this comment.
These are re-imported a couple lines down
Comment on lines
+66
to
+67
| segment_lengths = [150, 200, 40, 165, 45] | ||
| segment_lengths = [int(x) for x in segment_lengths] |
Contributor
There was a problem hiding this comment.
Suggested change
| segment_lengths = [150, 200, 40, 165, 45] | |
| segment_lengths = [int(x) for x in segment_lengths] | |
| segment_lengths = [150, 200, 40, 165, 45] |
The second line will return the same list right?
| print(len(states)) | ||
|
|
||
| if isinstance(states[0], list): | ||
| if isinstance(states[0][0], WeightedGaussianState): |
Contributor
There was a problem hiding this comment.
what happens if this if is false? is it correct that states will remain as the original list
Comment on lines
+46
to
+50
| for i in range(len(states)): | ||
| Likelihood_j.append(mvn.pdf( | ||
| states[i].hypothesis.measurement.state_vector.T, | ||
| states[i].hypothesis.measurement_prediction.mean.ravel(), | ||
| states[i].hypothesis.measurement_prediction.covar)) |
Contributor
There was a problem hiding this comment.
Suggested change
| for i in range(len(states)): | |
| Likelihood_j.append(mvn.pdf( | |
| states[i].hypothesis.measurement.state_vector.T, | |
| states[i].hypothesis.measurement_prediction.mean.ravel(), | |
| states[i].hypothesis.measurement_prediction.covar)) | |
| for state in states: | |
| Likelihood_j.append(mvn.pdf( | |
| state.hypothesis.measurement.state_vector.T, | |
| state.hypothesis.measurement_prediction.mean.ravel(), | |
| state.hypothesis.measurement_prediction.covar)) |
Comment on lines
+44
to
+45
| if str(history_length) not in [x for x in self.transition_matrices.keys()]: | ||
| history_length = max([x for x in self.transition_matrices.keys()]) + 1 |
Contributor
There was a problem hiding this comment.
Suggested change
| if str(history_length) not in [x for x in self.transition_matrices.keys()]: | |
| history_length = max([x for x in self.transition_matrices.keys()]) + 1 | |
| if history_length not in self.transition_matrices.keys(): | |
| history_length = max(self.transition_matrices.keys()) + 1 |
Comment on lines
+1276
to
+1279
| if self.model_histories is None: | ||
| self.model_histories = [] | ||
| if self.existence is None: | ||
| self.existence = Probability(1) |
Contributor
There was a problem hiding this comment.
Can move this into the "default factory" way
Comment on lines
+129
to
+132
| gtplotter = Plotter() | ||
| ms = [m for _, m in all_measurements] | ||
| gtplotter.plot_ground_truths(truths, [0, 2], linewidth=2) | ||
| gtplotter.fig |
Contributor
There was a problem hiding this comment.
Suggested change
| gtplotter = Plotter() | |
| ms = [m for _, m in all_measurements] | |
| gtplotter.plot_ground_truths(truths, [0, 2], linewidth=2) | |
| gtplotter.fig | |
| plotter = Plotter() | |
| ms = [m for _, m in all_measurements] | |
| plotter.plot_ground_truths(truths, [0, 2], linewidth=2) | |
| plotter.fig |
Makes sense to have the truth on the same plot as the tracks
| measurement_reducer, priors, all_measurements) | ||
|
|
||
| # %% | ||
| plotter = Plotter() |
Contributor
There was a problem hiding this comment.
Suggested change
| plotter = Plotter() |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR adds the components for the Multiple model Framework and provides a tutorial introducing the KF, GPB1, GPB2 and IMM implemented through the Multiple Model Framework.