Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Loading...
Clear generated features for a DSP block (used in tests).
Project ID
DSP Block ID, use the impulse functions to retrieve the ID
OK
Whether the operation succeeded
Optional error description (set if 'success' was false)
Retrieve the names of the features the DSP block generates
Project ID
DSP Block ID, use the impulse functions to retrieve the ID
OK
Whether the operation succeeded
Optional error description (set if 'success' was false)
"`[ \"accX RMS\", \"accX Peak 1 Freq\" ]`"Get a sample of trained features, this extracts a number of samples and their labels. Used to visualize the current training set.
Project ID
DSP Block ID, use the impulse functions to retrieve the ID
Which of the three acquisition categories to download data from
Feature axis 1
Feature axis 2
Feature axis 3
OK
Whether the operation succeeded
Optional error description (set if 'success' was false)
Total number of windows in the data set
Data by feature index for this window
"`{ 0: 9.81, 11: 0.32, 22: 0.79 }`"Training label index
Training label string
When showing the processed features, skip the first X features. This is used in dimensionality reduction where artificial features are introduced in the response (on the first few positions).
Retrieve the configuration parameters for the DSP block. Use the impulse functions to retrieve all DSP blocks.
Project ID
DSP Block ID, use the impulse functions to retrieve the ID
OK
Whether the operation succeeded
Optional error description (set if 'success' was false)
1"Spectral features"3000"spectral-analysis"Whether this block has generated features
Number of generated features
Names of the features
Classes that the features were generated on
Expected number of windows that would be generated
Axes that this block depends on.
Whether this type of DSP block supports autotuning.
For DSP blocks that support autotuning, this value specifies the minimum block implementation version for which autotuning is supported.
Whether autotune results exist for this DSP block.
Whether this DSP block uses state.
"Scaling""Scale axes""text""Divide axes by this number""scale-axes"Interface section to render parameter in.
Only valid for type "string". Will render a multiline text area.
If set, shows a hint below the input.
Sets the placeholder text on the input element (for types "string", "int", "float" and "secret")
Get slice of raw sample data, but with only the axes selected by the DSP block. E.g. if you have selected only accX and accY as inputs for the DSP block, but the raw sample also contains accZ, accZ is filtered out.
Project ID
DSP Block ID, use the impulse functions to retrieve the ID
Sample ID
Begin index of the slice
End index of the slice. If not given, the sample will be sliced to the same length as the impulse input block window length.
OK
Whether the operation succeeded
Optional error description (set if 'success' was false)
2"idle01.d8Ae"Whether signature validation passed
true"HS256"Either the shared key or the public key that was used to validate the sample
Timestamp when the sample was created on device, or if no accurate time was known on device, the time that the file was processed by the ingestion service.
Timestamp when the sample was last modified.
"training""healthy-machine"Interval between two windows (1000 / frequency). If the data was resampled, then this lists the resampled interval.
16Frequency of the sample. If the data was resampled, then this lists the resampled frequency.
62.5Interval between two windows (1000 / frequency) in the source data (before resampling).
16Frequency of the sample in the source data (before resampling).
62.5Name of the axis
"accX"Type of data on this axis. Needs to comply to SenML units (see https://www.iana.org/assignments/senml/senml.xhtml).
Number of readings in this file
Total length (in ms.) of this file
Timestamp when the sample was added to the current acquisition bucket.
True if the current sample is excluded from use
True if the current sample is still processing (e.g. for video)
Set when sample is processing and a job has picked up the request
Set when processing this sample failed
Error (only set when processing this sample failed)
Whether the sample is cropped from another sample (and has crop start / end info)
Sample free form associated metadata
Unique identifier of the project this sample belongs to
Name of the owner of the project this sample belongs to
Name of the project this sample belongs to
What labeling flow the project this sample belongs to uses
Data sample SHA 256 hash (including CBOR envelope if applicable)
Start index of the label (e.g. 0)
End index of the label (e.g. 3). This value is inclusive, so { startIndex: 0, endIndex: 3 } covers 0, 1, 2, 3.
The label for this section.
If this sample was created by a synthetic data job, it's referenced here.
Sensor readings and metadata
Unique identifier for this device. Only set this when the device has a globally unique identifier (e.g. MAC address).
"ac:87:a3:0a:2d:1b"Device type, for example the exact model of the device. Should be the same for all similar devices.
"DISCO-L475VG-IOT01A"Array with sensor axes
Name of the axis
"accX"Type of data on this axis. Needs to comply to SenML units (see https://www.iana.org/assignments/senml/senml.xhtml).
Array of sensor values. One array item per interval, and as many items in this array as there are sensor axes. This type is returned if there are multiple axes.
New start index of the cropped sample
0New end index of the cropped sample
128Total number of payload values
Download labels for a DSP block over all data in the training set, already sliced in windows.
Project ID
DSP Block ID, use the impulse functions to retrieve the ID
Which of the three acquisition categories to download data from
Numpy binary file
Download output from a DSP block over all data in the training set, already sliced in windows. In Numpy binary format.
Project ID
DSP Block ID, use the impulse functions to retrieve the ID
Which of the three acquisition categories to download data from
Whether to download raw data or processed data. Processed data is the default.
Numpy binary file
Set configuration parameters for the DSP block. Only values set in the body will be overwritten.
Project ID
DSP Block ID, use the impulse functions to retrieve the ID
OK
Whether the operation succeeded
Optional error description (set if 'success' was false)
Get a set of parameters, found as a result of running the DSP autotuner.
Project ID
DSP Block ID, use the impulse functions to retrieve the ID
OK
Whether the operation succeeded
Optional error description (set if 'success' was false)
Get estimated performance (latency and RAM) for the DSP block, for all supported project latency devices.
Project ID
DSP Block ID, use the impulse functions to retrieve the ID
OK
Whether the operation succeeded
Optional error description (set if 'success' was false)
List of performance estimates for each supported MCU.
Latency estimate, in ms
RAM estimate, in bytes
Retrieve the feature importance for a DSP block (only available for blocks where dimensionalityReduction is not enabled)
Project ID
DSP Block ID, use the impulse functions to retrieve the ID
OK
Whether the operation succeeded
Optional error description (set if 'success' was false)
Returns performance characteristics for a custom DSP block (needs hasTfliteImplementation). Updates are streamed over the websocket API (or can be retrieved through the /stdout endpoint). Use getProfileTfliteJobResult to get the results when the job is completed.
Project ID
DSP Block ID, use the impulse functions to retrieve the ID
DSP parameters with values
OK
Whether the operation succeeded
Optional error description (set if 'success' was false)
Job identifier. Status updates will include this identifier.
12873488112Runs the DSP block against a sample. This will move the sliding window (dependent on the sliding window length and the sliding window increase parameters in the impulse) over the complete file, and run the DSP function for every window that is extracted.
Project ID
DSP Block ID, use the impulse functions to retrieve the ID
Sample ID
OK
Whether the operation succeeded
Optional error description (set if 'success' was false)
Total number of windows in the data set
Feature data for this window
"`{ 9.81, 0.32, 0.79 }`"Training label index
Training label string
When showing the processed features, skip the first X features. This is used in dimensionality reduction where artificial features are introduced in the response (on the first few positions).
Retrieve the metadata from a generated DSP block.
Project ID
DSP Block ID, use the impulse functions to retrieve the ID
Whether to exclude 'includedSamples' in the response (as these can slow down requests significantly).
OK
Whether the operation succeeded
Optional error description (set if 'success' was false)
Date when the features were created
Labels in the dataset when generator ran
Names of the generated features. Only set if axes have explicit labels.
Number of features for this DSP block
The included samples in this DSP block. Note that these are sorted in the same way as the npy files are laid out. So with the windowCount parameter you can exactly search back to see which file contributed to which windows there.
Length of the sliding window when generating features.
Increase of the sliding window when generating features.
Whether data was zero-padded when generating features.
Frequency of the original data in Hz.
Information about the output of the DSP block
Output type of the DSP block
The shape of the block output
Available on all types. Denotes the width of an 'image' or 'spectrogram', or the number of features in a 'flat' block.
Only available for type 'image' and 'spectrogram'
Only available for type 'image'
Number of frames, only available for type 'image'
The version number of the resampling algorithm used (for resampled time series data only)
Takes in a features array and runs it through the DSP block. This data should have the same frequency as set in the input block in your impulse.
Project ID
DSP Block ID, use the impulse functions to retrieve the ID
Array of features. If you have multiple axes the data should be interleaved (e.g. [ax0_val0, ax1_val0, ax2_val0, ax0_val1, ax1_val1, ax2_val1]).
DSP parameters with values
Whether to generate graphs (will take longer)
Whether to request performance info (will take longer unless cached)
OK
Whether the operation succeeded
Optional error description (set if 'success' was false)
Array of processed features. Laid out according to the names in 'labels'
Graphs to plot to give an insight in how the DSP process ran
Name of the graph
"Frequency domain"Base64 encoded image, only present if type is 'image'
Mime type of the Base64 encoded image, only present if type is 'image'
Values on the x-axis per plot. Key is the name of the raw feature. Present if type is 'logarithmic' or 'linear'.
Values of the y-axis. Present if type is 'logarithmic' or 'linear'.
Suggested minimum value of x-axis
Suggested maxium value of x-axis
Suggested minimum value of y-axis
Suggested maximum value of y-axis
Type of graph (either logarithmic, linear or image)
Width of the graph line (if type is logarithmic or linear). Default 3.
Whether to apply smoothing to the graph.
Labels for the graph x and y axes.
Indices of points to highlight, per axis.
Labels of the feature axes
String representation of the DSP state returned
Get slice of sample data, and run it through the DSP block. This only the axes selected by the DSP block. E.g. if you have selected only accX and accY as inputs for the DSP block, but the raw sample also contains accZ, accZ is filtered out.
Project ID
DSP Block ID, use the impulse functions to retrieve the ID
Sample ID
Begin index of the slice
End index of the slice. If not given, the sample will be sliced to the same length as the impulse input block window length.
OK
Whether the operation succeeded
Optional error description (set if 'success' was false)
Array of processed features. Laid out according to the names in 'labels'
Graphs to plot to give an insight in how the DSP process ran
Name of the graph
"Frequency domain"Base64 encoded image, only present if type is 'image'
Mime type of the Base64 encoded image, only present if type is 'image'
Values on the x-axis per plot. Key is the name of the raw feature. Present if type is 'logarithmic' or 'linear'.
Values of the y-axis. Present if type is 'logarithmic' or 'linear'.
Suggested minimum value of x-axis
Suggested maxium value of x-axis
Suggested minimum value of y-axis
Suggested maximum value of y-axis
Type of graph (either logarithmic, linear or image)
Width of the graph line (if type is logarithmic or linear). Default 3.
Whether to apply smoothing to the graph.
Labels for the graph x and y axes.
Indices of points to highlight, per axis.
Labels of the feature axes
String representation of the DSP state returned
Label for the window (only present for time-series data)
2"idle01.d8Ae"Whether signature validation passed
true"HS256"Either the shared key or the public key that was used to validate the sample
Timestamp when the sample was created on device, or if no accurate time was known on device, the time that the file was processed by the ingestion service.
Timestamp when the sample was last modified.
"training""healthy-machine"Interval between two windows (1000 / frequency). If the data was resampled, then this lists the resampled interval.
16Frequency of the sample. If the data was resampled, then this lists the resampled frequency.
62.5Interval between two windows (1000 / frequency) in the source data (before resampling).
16Frequency of the sample in the source data (before resampling).
62.5Name of the axis
"accX"Type of data on this axis. Needs to comply to SenML units (see https://www.iana.org/assignments/senml/senml.xhtml).
Number of readings in this file
Total length (in ms.) of this file
Timestamp when the sample was added to the current acquisition bucket.
True if the current sample is excluded from use
True if the current sample is still processing (e.g. for video)
Set when sample is processing and a job has picked up the request
Set when processing this sample failed
Error (only set when processing this sample failed)
Whether the sample is cropped from another sample (and has crop start / end info)
Sample free form associated metadata
Unique identifier of the project this sample belongs to
Name of the owner of the project this sample belongs to
Name of the project this sample belongs to
What labeling flow the project this sample belongs to uses
Data sample SHA 256 hash (including CBOR envelope if applicable)
Start index of the label (e.g. 0)
End index of the label (e.g. 3). This value is inclusive, so { startIndex: 0, endIndex: 3 } covers 0, 1, 2, 3.
The label for this section.
If this sample was created by a synthetic data job, it's referenced here.
Sensor readings and metadata
Unique identifier for this device. Only set this when the device has a globally unique identifier (e.g. MAC address).
"ac:87:a3:0a:2d:1b"Device type, for example the exact model of the device. Should be the same for all similar devices.
"DISCO-L475VG-IOT01A"Array with sensor axes
Name of the axis
"accX"Type of data on this axis. Needs to comply to SenML units (see https://www.iana.org/assignments/senml/senml.xhtml).
Array of sensor values. One array item per interval, and as many items in this array as there are sensor axes. This type is returned if there are multiple axes.
New start index of the cropped sample
0New end index of the cropped sample
128Total number of payload values
Get raw sample data, but with only the axes selected by the DSP block. E.g. if you have selected only accX and accY as inputs for the DSP block, but the raw sample also contains accZ, accZ is filtered out. If you pass dspId = 0 this will return a raw graph without any processing.
Project ID
DSP Block ID, use the impulse functions to retrieve the ID
Sample ID
Limit the number of payload values in the response
OK
Whether the operation succeeded
Optional error description (set if 'success' was false)
2"idle01.d8Ae"Whether signature validation passed
true"HS256"Either the shared key or the public key that was used to validate the sample
Timestamp when the sample was created on device, or if no accurate time was known on device, the time that the file was processed by the ingestion service.
Timestamp when the sample was last modified.
"training""healthy-machine"Interval between two windows (1000 / frequency). If the data was resampled, then this lists the resampled interval.
16Frequency of the sample. If the data was resampled, then this lists the resampled frequency.
62.5Interval between two windows (1000 / frequency) in the source data (before resampling).
16Frequency of the sample in the source data (before resampling).
62.5Name of the axis
"accX"Type of data on this axis. Needs to comply to SenML units (see https://www.iana.org/assignments/senml/senml.xhtml).
Number of readings in this file
Total length (in ms.) of this file
Timestamp when the sample was added to the current acquisition bucket.
True if the current sample is excluded from use
True if the current sample is still processing (e.g. for video)
Set when sample is processing and a job has picked up the request
Set when processing this sample failed
Error (only set when processing this sample failed)
Whether the sample is cropped from another sample (and has crop start / end info)
Sample free form associated metadata
Unique identifier of the project this sample belongs to
Name of the owner of the project this sample belongs to
Name of the project this sample belongs to
What labeling flow the project this sample belongs to uses
Data sample SHA 256 hash (including CBOR envelope if applicable)
Start index of the label (e.g. 0)
End index of the label (e.g. 3). This value is inclusive, so { startIndex: 0, endIndex: 3 } covers 0, 1, 2, 3.
The label for this section.
If this sample was created by a synthetic data job, it's referenced here.
Sensor readings and metadata
Unique identifier for this device. Only set this when the device has a globally unique identifier (e.g. MAC address).
"ac:87:a3:0a:2d:1b"Device type, for example the exact model of the device. Should be the same for all similar devices.
"DISCO-L475VG-IOT01A"Array with sensor axes
Name of the axis
"accX"Type of data on this axis. Needs to comply to SenML units (see https://www.iana.org/assignments/senml/senml.xhtml).
Array of sensor values. One array item per interval, and as many items in this array as there are sensor axes. This type is returned if there are multiple axes.
New start index of the cropped sample
0New end index of the cropped sample
128Total number of payload values