MacroModule¶| genre | ML_Inference_PatchBased |
| author | Jan-Martin Kuhnigk |
| package | FMEstable/ReleaseMeVis |
| definition | ApplyTileProcessorsPageWiseOrtho.def |
| see also | TensorFlowServingTileProcessor, ApplyTileProcessorPageWise |
| keywords | classif, predic, deep, learning, ApplyTileClassifiers, Ortho |
Applies up to three different processors, one for each orthogonal orientation (sagittal, coronal, transversal).
Connect image and processors and update.
See ApplyTileProcessorPageWise for details on tile size computation, border filling, and ROI application.
Processor Hot Switching
In certain situations, e.g. when switching between different models for sagittal, coronal, etc. via a single remote server connection (cf. TensorFlowServingTileProcessor), it may be necessary to set a new processor for each orientation during processing. This is supported by ignoring input processor field updates when updating the field Currently Active Processor for the next orientation to process. The user can attach a listener to the field and provide a new processor depending on its value. An example for this use case is provided in the Test Suite (search for “hot switching”).
outParameterInfo or outCombinedParameterInfo.Currently Active Processor field, which is set before using a different processor.
name: inTileProcessorSagittal, type: MLBase, deprecated name: inTileClassifierSagittal¶Processor used for sagittal 2d processing
name: outParameterInfo, type: MLBase¶Parameter information with one separate entry per orientation, comprising information about processor application (top level) and the processor itself (below the ‘processor’ key).
name: outCombinedParameterInfo, type: MLBase¶Parameter information combined across orientations, comprising information about processor application (top level) and the processor itself (below the ‘processor’ key).
The combination works as follows: In the toplevel key COMBINED_INFO_IDS you can see the enabled orientations in the order used also in the following. For all other keys:
name: onInputChangeBehavior, type: Enum, default: Clear, deprecated name: shouldAutoUpdate,shouldUpdateAutomatically¶Declares how the module should react if a value of an input field changes.
Values:
| Title | Name | Deprecated Name |
|---|---|---|
| Update | Update | TRUE |
| Clear | Clear | FALSE |
name: updateDone, type: Trigger, persistent: no¶Notifies that an update was performed (Check status interface fields to identify success or failure).
name: hasValidOutput, type: Bool, persistent: no¶Indicates validity of output field values (success of computation).
name: statusCode, type: Enum, persistent: no¶Reflects module’s status (successful or failed computations) as one of some predefined enumeration values.
Values:
| Title | Name |
|---|---|
| Ok | Ok |
| Invalid input object | Invalid input object |
| Invalid input parameter | Invalid input parameter |
| Internal error | Internal error |
name: statusMessage, type: String, persistent: no¶Gives additional, detailed information about status code as human-readable message.
name: inUpdateModelParameterInfoFromSingleProcessorConnection, type: Trigger, deprecated name: inUpdateModelParameterInfoFromSingleClassifierConnection¶When using a single, shared processor connection (cf. Use a Single, Shared Processor Input Connection), you may still want to update the (combined) processor parameter infos before actually running the processing, which is what this trigger is for.
name: inEnableTransversal, type: Bool, default: TRUE¶If enabled, transversal processing is performed (a corresponding processor must be connected)
name: inEnableSagittal, type: Bool, default: TRUE¶If enabled, sagittal processing is performed (a corresponding processor must be connected)
name: inEnableCoronal, type: Bool, default: TRUE¶If enabled, coronal processing is performed (a corresponding processor must be connected)
name: inUseProcessingRoi, type: Bool, default: FALSE, deprecated name: inUseClassificationRoi¶If enabled, processing (and hence all output images) are restricted to Processing ROI Voxel Box.
name: inProcessingRoi_voxel, type: String, default: 0 0 0 0 0 0 -1 -1 -1 -1 -1 -1, deprecated name: inClassificationRoi\_voxel¶If Restrict processing to ROI (as opposed to running on the whole image) is enabled, this box is used to clamp the output to prevent unnecessary computation efforts. The box is to be given in integer voxel coordinates relative to the input image.
name: inUseCustomOutputTileSizeTransversal, type: Bool, default: FALSE¶If enabled, Custom Output Tile Size is used instead of the proposition from the connected processor.
name: inUseCustomOutputTileSizeSagittal, type: Bool, default: FALSE¶If enabled, Custom Output Tile Size is used instead of the proposition from the connected processor.
name: inUseCustomOutputTileSizeCoronal, type: Bool, default: FALSE¶If enabled, Custom Output Tile Size is used instead of the proposition from the connected processor.
name: inCustomFillMode, type: Enum, default: FillValue, deprecated name: inFillMode¶If Use is enabled, this overrides the processor’s proposal for the fill mode, which defines how undefined input regions are to be filled.
Values:
| Title | Name | Description |
|---|---|---|
| Fill Value | FillValue | Fill with a fixed value |
| Reflect | Reflect | Reflect valid region (but only within padding range). See ApplyTileProcessorPageWise ‘Details’ section for details. |
name: inUseCustomFillValue, type: Bool, default: FALSE¶If enabled, Fill Value is used instead of the proposition from the connected processor.
name: inExternalDimensionForChannel1, type: Enum, default: C¶Defines which (external, i.e. in input and output image) dimension shall be used by the processor as primary CHANNEL dimension (if it needs one).
Values:
| Title | Name |
|---|---|
| X | X |
| Y | Y |
| Z | Z |
| C | C |
| T | T |
| U | U |
name: inExternalDimensionForChannel2, type: Enum, default: U¶Defines which (external, i.e. in input and output image) dimension shall be used by the processor as secondary CHANNEL dimension (if it needs one).
Values:
| Title | Name |
|---|---|
| X | X |
| Y | Y |
| Z | Z |
| C | C |
| T | T |
| U | U |
name: inUseCustomExternalToProcessorDimensionMapping, type: Bool, default: FALSE, deprecated name: inUseCustomExternalToClassifierDimensionMapping¶If enabled, Expected Dimensions (in ML Order, largest stride last) is used instead of the one provided by the processors. Should usually not be necessary, try to reconfigure your processors instead!
name: inCustomExternalToProcessorDimensionMapping, type: String, default: X, Y, CHANNEL, BATCH, deprecated name: inCustomExternalToClassifierDimensionMapping¶Comma-separated list describing the order in which the input data dimensions are expected by the processor, in ML order, i.e. smallest stride (X) first. On top of the data (typically spatial) dimensions provided to the processor, you can specify where the input channel and batch dimensions are located. Valid elemental values are X, Y, Z, C, T, U, as well as CHANNEL (or CHANNEL1), CHANNEL2, BATCH and UNUSED. UNUSED can be used for all dimensions not required by the processor (but it can also be omitted).
Typical examples are
X, Y, CHANNEL, BATCH
for 2D processors with a channel and batch dimension, or
X, Y, Z, CHANNEL, BATCH
for 3D processors with a channel and batch dimension.
** Normally, you should not have to edit this here, as the processor should already propose the configuration it needs, so that Use can be left disabled. **
name: outSagittalInputImageExtent, type: Vector6, persistent: no¶name: outCoronalInputImageExtent, type: Vector6, persistent: no¶name: outTransversalInputImageExtent, type: Vector6, persistent: no¶name: outCurrentlyActiveProcessor, type: Enum, persistent: no, deprecated name: outCurrentlyActiveClassifier¶Indicates which processor is currently used. Set before using the processor, so that it is possible for a user to react to this.
Values:
| Title | Name |
|---|---|
| None | None |
| Sagittal | Sagittal |
| Coronal | Coronal |
| Transversal | Transversal |