Keras information

Get information about a Keras block, such as its dependencies. Use the impulse blocks to find the learnId.

GEThttps://studio.edgeimpulse.com/v1/api/{projectId}/training/keras/{learnId}

Authorization

Path parameters

projectId*integer

Project ID

learnId*integer

Learn Block ID, use the impulse functions to retrieve the ID

Response

Body

success*boolean

Whether the operation succeeded

errorstring

Optional error description (set if 'success' was false)

dependencies*object

classes*array of string

itemsstring

blockNames*array of string

itemsstring

featureCount*integer

sampleCount*integer

trained*boolean

Whether the block is trained

name*string

typeenum

The type of learning block (anomaly, keras, keras-transfer-image, keras-transfer-kws, keras-object-detection, keras-regression). Each behaves differently.

anomalyanomaly-gmmkeraskeras-transfer-imagekeras-transfer-kwskeras-object-detectionkeras-regressionkeras-akidakeras-akida-transfer-imagekeras-akida-object-detectionkeras-visual-anomaly

script*string

The Keras script. This script might be empty if the mode is visual.

minimumConfidenceRating*number

Minimum confidence rating required for the neural network. Scores below this confidence are tagged as uncertain.

selectedModelType*enum

The model type that is currently selected.

int8float32akidarequiresRetrain

mode*enum

The mode (visual or expert) to use for editing this network.

visualexpert

visualLayers*array of object

The visual layers (if in visual mode) for the neural network. This will be an empty array when in expert mode.

type*enum

denseconv1dconv2dreshapeflattendropoutbatchNormalizationtransfer_mobilenetv2_a35transfer_mobilenetv2_a1transfer_mobilenetv2_a05transfer_mobilenetv2_160_a1transfer_mobilenetv2_160_a75transfer_mobilenetv2_160_a5transfer_mobilenetv2_160_a35transfer_mobilenetv1_a25_d100transfer_mobilenetv1_a2_d100transfer_mobilenetv1_a1_d100transfer_kws_mobilenetv1_a1_d100transfer_kws_mobilenetv2_a35_d100transfer_kws_syntiant_ndp10xtransfer_kws_conv2d_tinyobject_ssd_mobilenet_v2_fpnlite_320x320fomo_mobilenet_v2_a01fomo_mobilenet_v2_a35transfer_organizationtransfer_akidanet_imagenet_160_a100transfer_akidanet_imagenet_160_a50transfer_akidanet_imagenet_160_a25transfer_akidanet_imagenet_224_a100transfer_akidanet_imagenet_224_a50transfer_akidanet_imagenet_224_a25fomo_akidanet_a50fomo_ad_gmmfomo_ad_patchcore

neuronsinteger

Number of neurons or filters in this layer (only for dense, conv1d, conv2d) or in the final conv2d layer (only for transfer layers)

kernelSizeinteger

Kernel size for the convolutional layers (only for conv1d, conv2d)

dropoutRatenumber

Fraction of input units to drop (only for dropout) or in the final layer dropout (only for transfer layers)

columnsinteger

Number of columns for the reshape operation (only for reshape)

stackinteger

Number of convolutional layers before the pooling layer (only for conv1d, conv2d)

enabledboolean

organizationModelIdinteger

Custom transfer learning model ID (when type is set to transfer_organization)

trainingCycles*integer

Number of training cycles. If in expert mode this will be 0.

learningRate*number

Learning rate (between 0 and 1). If in expert mode this will be 0.

batchSizeinteger

The batch size used during training.

defaultBatchSize*integer

The default batch size if a value is not configured.

shape*string

Python-formatted tuple of input axes

trainTestSplitnumber

Train/test split (between 0 and 1)

autoClassWeightsboolean

Whether to automatically balance class weights, use this for skewed datasets.

useLearnedOptimizerboolean

Use learned optimizer and ignore learning rate.

augmentationPolicyImage*enum

The data augmentation policy to use with image input

noneall

augmentationPolicySpectrogramobject

enabled*boolean

True if spectrogram augmentation is enabled. Other properties will be ignored if this is false.

warpingboolean

True if warping along the time axis is enabled.

freqMaskingenum

The amount of frequency masking to apply.

nonelowhigh

timeMaskingenum

The amount of time masking to apply.

nonelowhigh

gaussianNoiseenum

The amount of Gaussian noise to add.

nonelowhigh

transferLearningModels*array of object

name*string

shortName*string

abbreviatedNamestring

description*string

hasNeurons*boolean

hasDropout*boolean

defaultNeuronsinteger

defaultDropoutnumber

defaultLearningRatenumber

defaultTrainingCyclesnumber

hasImageAugmentationboolean

type*enum

learnBlockTypeenum

The type of learning block (anomaly, keras, keras-transfer-image, keras-transfer-kws, keras-object-detection, keras-regression). Each behaves differently.

anomalyanomaly-gmmkeraskeras-transfer-imagekeras-transfer-kwskeras-object-detectionkeras-regressionkeras-akidakeras-akida-transfer-imagekeras-akida-object-detectionkeras-visual-anomaly

organizationModelIdinteger

implementationVersioninteger

repositoryUrlstring

URL to the source code of this custom learn block.

author*string

blockType*enum

officialpersonalenterprisepro-or-enterprisecommunity

customParametersarray of object

name*string

Example: "Scale axes"

valuestring

defaultValue*string

type*string

Example: "text"

helpstring

Example: "Divide axes by this number"

param*string

Example: "scale-axes"

selectOptionsarray of object

valuestring

selectedboolean

optionLabelstring

readonly*boolean

shouldShow*boolean

showIfobject

parameter*string

operator*enum

eqneq

value*string

invalidTextstring

sectionenum

Interface section to render parameter in.

advancedaugmentationmodelProfiling

multilineboolean

Only valid for type "string". Will render a multiline text area.

required*boolean

hintstring

If set, shows a hint below the input.

placeholderstring

Sets the placeholder text on the input element (for types "string", "int", "float" and "secret")

displayCategoryenum

Category to display this block in the UI.

classicaltao

profileInt8*boolean

Whether to profile the i8 model (might take a very long time)

skipEmbeddingsAndMemory*boolean

If set, skips creating embeddings and measuring memory (used in tests)

akidaEdgeLearningConfigobject

enabled*boolean

True if Akida Edge Learning model creation is enabled. Other properties will be ignored if this is false.

additionalClassesnumber

Number of additional classes that will be added to the Edge Learning model.

neuronsPerClassnumber

Number of neurons in each class on the last layer in the Edge Learning model.

customValidationMetadataKeystring

This metadata key is used to prevent group data leakage between train and validation datasets.

showAdvancedTrainingSettings*boolean

Whether the 'Advanced training settings' UI element should be expanded.

showAugmentationTrainingSettings*boolean

Whether the 'Augmentation training settings' UI element should be expanded.

customParametersobject

Training parameters, this list depends on the list of parameters that the model exposes.

Other propertiesnullable string

anomalyCapacityenum

Capacity level for visual anomaly detection (GMM). Determines which set of default configurations to use. The higher capacity, the higher number of (Gaussian) components, and the more adapted the model becomes to the original distribution

lowmediumhigh

lastShownModelVariantenum

Last shown variant on the Keras screen. Used to keep the same view after refreshing.

int8float32akida

lastShownModelEngineenum

Last shown model engine on the Keras screen. Used to keep the same view after refreshing.

tflite-eontflite-eon-ram-optimizedtflite

blockParametersone of

Training parameters specific to the type of the learn block. Parameters may be adjusted depending on the model defined in the visual layers. Used for our built-in blocks.

backbonestring

The backbone to use for feature extraction

numLayersinteger

The number of layers in the feature extractor (1-3)

poolSizeinteger

The pool size for the feature extractor

samplingRationumber

The sampling ratio for the coreset, used for anomaly scoring

numNearestNeighborsinteger

The number of nearest neighbors to consider, used for anomaly scoring

Request

Response