glimpse.glab.api

DEFAULT_LAYER

Default model layer to use for evaluation.

Reset()

Clear all experimental settings and results.

SetParams(params=None, **kw)

Choose model parameters.

SetParamsWithGui(params=None, **kw)

Choose model parameters using a graphical interface.

This presents a display similar to that shown in Figure 1.

Figure 1: A screenshot of the editor for model parameters.

LoadParams(path)

Read model parameters from disk.

SetLayer(layer)

Set the model layer to use for evaluation.

StoreExperiment(path)

Store settings and results for experiment to disk.

LoadExperiment(path)

Load settings and results for experiment from disk.

GetExperiment()

Get the current experiment object.

This is an advanced function. In general, the user should not modify the experiment object directly.

Verbose(flag=True)

Set the verbosity of log output.

Parameters:flag (bool) – Whether to enable verbose logging.

GetModel()

Get the Glimpse model used for this experiment.

This is an advanced function. In general, the user should not need to interact with the model directly.

SetCorpus(corpus_dir, balance=False)

Read images from the corpus directory.

This function assumes that each sub-directory contains images for exactly one object class, with a different object class for each sub-directory. Training and testing subsets are chosen automatically.

Parameters:

• root_dir (str) – Path to corpus directory.
• balance (bool) – Ensure an equal number of images from each class (by random selection).
• reader – Filesystem reader.

See also

SetCorpusSubdirs(), SetCorpusSplit()

SetCorpusSubdirs(corpus_subdirs, balance=False)

Read images from per-class corpus sub-directories.

This function assumes that each sub-directory contains images for exactly one object class, with a different object class for each sub-directory. Training and testing subsets are chosen automatically.

Parameters:

• subdirs (iterable of str) – Path of each corpus sub-directory.
• balance (bool) – Ensure an equal number of images from each class (by random selection).
• reader – Filesystem reader.

See also

SetCorpus(), SetCorpusSplit()

SetCorpusSplit(train_dir, test_dir)

Read images and training information from the corpus directory.

This function assumes that the train_dir and test_dir have the same set of sub-directories. Each sub-directory shoudl contain images for exactly one object class, with a different object class for each sub-directory.

Parameters:

• train_dir (str) – Path to corpus of training images.
• test_dir (str) – Path to corpus of test images.
• reader – Filesystem reader.

See also

SetCorpus(), SetCorpusSubdirs()

SetCorpusByName(name)

Use a sample image corpus for this experiment.

Parameters:name (str) – Corpus name. One of ‘easy’, ‘moderate’, or ‘hard’.

This provides access to a small set of images for demonstration purposes, which are composed of simple shapes on various background patterns.

SetS2Prototypes(prototypes)

Manually specify the set of S2 prototypes.

Parameters:prototypes (str or list of array of float) – Path to prototypes on disk, or prototypes array to set. Return type:list of array of float Returns:Set of model prototypes.

ImprintS2Prototypes(num_prototypes)

Create a set of S2 prototypes by “imprinting” from training images.

Patches are drawn from all classes of the training data.

Parameters:num_prototypes (int) – Number of prototypes to create.

MakeUniformRandomS2Prototypes(num_prototypes, low=None, high=None)

Create a set of random S2 prototypes drawn from the uniform distribution.

Each element of every prototype is drawn independently from an uniform distribution with the same parameters.

Parameters:

• num_prototypes (int) – Number of prototypes to create.
• low (float) – Minimum value in uniform range.
• high (float) – Maximum value in uniform range.

MakeShuffledRandomS2Prototypes(num_prototypes)

Create a set of “imprinted” S2 prototypes that have been shuffled.

Each prototype has its contents randomly permuted across location and orientation band.

Parameters:num_prototypes (int) – Number of prototypes to create.

MakeHistogramRandomS2Prototypes(num_prototypes)

Create a set of S2 prototypes drawn from a 1D histogram of C1 activity.

The set is created by drawing elements from a distribution that is estimated from a set of imprinted prototypes. Each entry is drawn independently of the others.

Parameters:num_prototypes (int) – Number of prototypes to create.

MakeNormalRandomS2Prototypes(num_prototypes)

Create a set of random S2 prototypes drawn from the normal distribution.

Each element of every prototype is drawn independently from a normal distribution with the same parameters.

Parameters:num_prototypes (int) – Number of prototypes to create.

MakeKmeansS2Prototypes(num_prototypes, num_patches=None)

Create a set of S2 prototypes by clustering C1 samples with k-Means.

Parameters:

• num_prototypes (int) – Number of prototypes to create.
• num_patches (int) – Number of sample patches passed to k-Means.

ComputeActivation(save_all=False)

Compute the model activity for all images in the experiment.

Parameters:

• layers (str or list of str) – One or more model layers to compute.
• pool – Worker pool to use for parallel computation.
• save_all (bool) – Whether to save activation for all model layers, rather than just those in layers.
• progress – Handler for incremental progress updates.

EvaluateClassifier(cross_validate=False, algorithm=None, train_size=None, num_folds=None, score_func=None)

Apply a classifier to the image features in the experiment.

Parameters:

• cross_validate (bool) – Whether to use cross-validation. The default will use a fixed training and testing split.
• algorithm – Learning algorithm, which is fit to features. This should be a scikit-learn classifier object. If not set, a LinearSVC object is used.
• train_size (float or int) – Size of training split, specified as a fraction (between 0 and 1) of total instances or as a number of instances (1 to N, where N is the number of available instances).
• num_folds (int) – Number of folds to use for cross-validation. Default is 10.
• score_func (str) – Name of the scoring function to use, as specified by ResolveScoreFunction().

Return type:

ExperimentData

Returns:

Results of evaluation.

GetFeatures()

Get the feature vectors for all images in the experiment.

GetImagePaths()

Returns the filename for each image in the corpus.

GetLabelNames()

Returns the class name for each image in the corpus.

GetParams()

Returns the model parameters for the experiment.

GetNumPrototypes(kwidth=0)

Return the number of S2 prototypes in the model.

Parameters:kwidth (int) – Index of kernel shape.

GetPrototype(prototype=0, kwidth=0)

Return an S2 prototype from the experiment.

Parameters:

• prototype (int) – Index of S2 prototype.
• kwidth (int) – Index of kernel shape.

GetImprintLocation(prototype=0, kwidth=0)

Return the image location from which a prototype was imprinted.

This requires that the prototypes were learned by imprinting.

Parameters:

• prototype (int) – Index of S2 prototype.
• kwidth (int) – Index of kernel shape.

Returns:

Location information in the format (image index, scale, y-offset, x-offset), where scale and y- and x-offsets identify the S2 unit from which the prototype was “imprinted”.

Return type:

4 element array of int

GetEvaluationLayers(evaluation=0)

Returns the model layers from which features were extracted.

Parameters:evaluation (int) – Index of the evaluation record to use. Return type:list of str Returns:Names of layers used for evaluation.

GetEvaluationResults(evaluation=0)

Returns the results of a model evaluation.

Parameters:evaluation (int) – Index of the evaluation record to use. Return type:glimpse.util.data.Data Returns:Result data, with attributes that depend on the method of evaluation. In general, the feature_builder, score, score_func attributes will be available.

GetPredictions(training=False, evaluation=0)

Get information about classifier predictions.

Parameters:

• exp – Experiment data.
• training (bool) – Return information about training images. Otherwise, information about the test set is returned.
• evaluation (int) – Index of the evaluation record to use.

Return type:

list of 3-tuple of str

Returns:

filename, true label, and predicted label for each image in the set

ShowS2Activity(image=0, scale=0, prototype=0, kwidth=0)

Plot the S2 activity for a given image.

Parameters:

• image – Path to image on disk, or index of image in experiment.
• scale (int) – Index of scale band to use.
• prototype (int) – Index of S2 prototype to use.
• kwidth (int) – Index of kernel shape.

ShowPrototype(prototype=0, kwidth=0)

Plot the prototype activation.

There is one plot for each orientation band.

Parameters:

• prototype (int) – Index of S2 prototype to use.
• kwidth (int) – Index of kernel shape.
• colorbar (bool) – Add a colorbar to the plot.

AnnotateImprintedPrototype(prototype=0, kwidth=0)

Plot the image region used to construct a given imprinted prototype.

This shows the image in the background, with a red box over the imprinted region.

Parameters:

• prototype (int) – Index of S2 prototype to use.
• kwidth (int) – Index of kernel shape.

AnnotateS2Activity(image=0, scale=0, prototype=0, kwidth=0)

Plot the S2 activity and image data for a given image.

This shows the image in the background, with the S2 activity on top.

Parameters:

• image – Path to image on disk, or index of image in experiment.
• scale (int) – Index of scale band to use.
• prototype (int) – Index of S2 prototype to use.
• kwidth (int) – Index of kernel shape.

AnnotateC1Activity(image=0, scale=0)

Plot the C1 activation for a given image.

This shows the image in the background, with the activation plotted on top. There is one plot for each orientation band.

Parameters:

• image – Path to image on disk, or index of image in experiment.
• scale (int) – Index of scale band to use.

AnnotateS1Activity(image=0, scale=0)

Plot the S1 activation for a given image.

This shows the image in the background, with the activation plotted on top. There is one plot for each orientation band.

Parameters:

• image – Path to image on disk, or index of image in experiment.
• scale (int) – Index of scale band to use.
• colorbar (bool) – Whether to show colorbar with plot.

ShowS1Kernels()

Plot the S1 Gabor kernels.

There is one plot for each orientation band.