Abstract: A graphics processing unit (GPU) for clustering of machine learning (ML) functional components, including: a plurality of compute units; a plurality of ML clusters, wherein each of the ML clusters comprises at least one arithmetic logic unit (ALU), and wherein each of the ML clusters is associated with a respective subset of the compute units; and a plurality of memory modules each positioned on the GPU adjacent to a respective ML cluster of the plurality of ML clusters, wherein each ML cluster is configured to directly access one or more adjacent memory modules.

Abstract: Systems, apparatuses, and methods for achieving higher cache hit rates for machine learning models are disclosed. When a processor executes a given layer of a machine learning model, the processor generates and stores activation data in a cache subsystem a forward or reverse manner. Typically, the entirety of the activation data does not fit in the cache subsystem. The processor records the order in which activation data is generated for the given layer. Next, when the processor initiates execution of a subsequent layer of the machine learning model, the processor processes the previous layer's activation data in a reverse order from how the activation data was generated. In this way, the processor alternates how the layers of the machine learning model process data by either starting from the front end or starting from the back end of the array.

Abstract: A processing system includes a central processing unit (CPU) and a graphics processing unit (GPU) that has a plurality of compute units. The GPU receives an image from the CPU and determines a total result area in a virtual-matrix-multiplication space of a virtual matrix-multiplication output matrix based on convolutional parameters associated with the image in an image space. The GPU partitions the total result area of the virtual matrix-multiplication output matrix into a plurality of virtual segments. The GPU allocates convolution operations to the plurality of compute units based on each virtual segment of the plurality of virtual segments.

Abstract: A processing system includes a central processing unit (CPU) and a graphics processing unit (GPU) that has a plurality of compute units. The GPU receives an image from the CPU and determines a total result area in a virtual-matrix-multiplication space of a virtual matrix-multiplication output matrix based on convolutional parameters associated with the image in an image space. The GPU partitions the total result area of the virtual matrix-multiplication output matrix into a plurality of virtual segments. The GPU allocates convolution operations to the plurality of compute units based on each virtual segment of the plurality of virtual segments.

Abstract: Processing of non-real-time and real-time workloads is performed using discrete pipelines. A first pipeline includes a first shader and one or more fixed function hardware blocks. A second pipeline includes a second shader that is configured to emulate the at least one fixed function hardware block. First and second memory elements store first state information for the first pipeline and second state information for the second pipeline, respectively. A non-real-time workload executing in the first pipeline is preempted at a primitive boundary in response to a real-time workload being dispatched for execution in the second pipeline. The first memory element retains the first state information in response to preemption of the non-real-time workload. The first pipeline is configured to resume processing the subsequent primitive on the basis of the first state information stored in the first memory element.

Abstract: Processing of non-real-time and real-time workloads is performed using discrete pipelines. A first pipeline includes a first shader and one or more fixed function hardware blocks. A second pipeline includes a second shader that is configured to emulate the at least one fixed function hardware block. First and second memory elements store first state information for the first pipeline and second state information for the second pipeline, respectively. A non-real-time workload executing in the first pipeline is preempted at a primitive boundary in response to a real-time workload being dispatched for execution in the second pipeline. The first memory element retains the first state information in response to preemption of the non-real-time workload. The first pipeline is configured to resume processing the subsequent primitive on the basis of the first state information stored in the first memory element.

Abstract: Processing of non-real-time and real-time workloads is performed using discrete pipelines. A first pipeline includes a first shader and one or more fixed function hardware blocks. A second pipeline includes a second shader that is configured to emulate the at least one fixed function hardware block. First and second memory elements store first state information for the first pipeline and second state information for the second pipeline, respectively. A non-real-time workload executing in the first pipeline is preempted at a primitive boundary in response to a real-time workload being dispatched for execution in the second pipeline. The first memory element retains the first state information in response to preemption of the non-real-time workload. The first pipeline is configured to resume processing the subsequent primitive on the basis of the first state information stored in the first memory element.

Abstract: A stage of a graphics pipeline in a graphics processing unit (GPU) detects an interrupt concurrently with the stage processing primitives in a first bin that represents a first portion of a first frame generated by a first application. The stage forwards a completed portion of the primitives to a subsequent stage of the graphics pipeline in response to the interrupt. The stage diverts a second bin that represents a second portion of the first frame from the stage to a memory in response to the interrupt. The stage processes primitives in a third bin that represents a portion of a second frame generated by a second application subsequent to diverting the second bin to the memory. The stage can then retrieve the second bin from the memory in response to the stage completing processing of the primitives in the third bin for additional processing.

Abstract: A method for preventing damage to a document by a document transport apparatus provides a member for receiving the document as a stack of sheets to be serial fed to a feeding station. At least two spaced-apart microphones are disposed at the feeding station and responsive to audio to produce signals representing audio energy received by each microphone respectively. The energy received from each microphone is compared to determine if it is ambient noise or if it indicates that two attached sheets are being fed or a single sheet is being damaged. The document transport apparatus is shut off to prevent damage to documents when it has been determined that two attached sheets are being fed or that a single sheet is being damaged.

Abstract: A document handling apparatus having a document transport path for moving a document therethrough. A detector proximate the transport detects a multifeed indication in the transport path or a misfeed indication in the transport path. A processing system processes the indications and issues a termination signal if a multifeed or a misfeed, or both, is determined.

Abstract: A method for feeding sheets through a sheet transport path. Ultrasonic energy is directed toward a sheet in the transport path while an audio receiver detects audio data generated by the ultrasonic source and the mechanisms that transport the sheet. The audio data is processed to determine whether a multifeed or a misfeed condition exists in the transport path.

Abstract: A document handling apparatus having a document transport path for moving a document therethrough. A single detector proximate the transport detects a misfeed indication in the transport path. A processing system processes the indication and issues a termination signal if a misfeed is determined.

Abstract: A method for feeding sheets through a sheet transport path. Ultrasonic energy is directed toward a sheet in the transport path while an audio receiver detects audio data generated by the ultrasonic source and the mechanisms that transport the sheet. The audio data is processed to determine whether a multifeed or a misfeed condition exists in the transport path.

Abstract: A document handling apparatus having a document transport path for moving a document therethrough. A detector proximate the transport detects a multifeed indication in the transport path or a misfeed indication in the transport path. A processing system processes the indications and issues a termination signal if a multifeed or a misfeed, or both, is determined.

Abstract: A method for preventing damage to a document by a document transport apparatus provides a member for receiving the document as a stack of sheets to be serial fed to a feeding station. At least two spaced-apart microphones are disposed at the feeding station and responsive to audio to produce signals representing audio energy received by each microphone respectively. The energy received from each microphone is compared to determine if it is ambient noise or if it indicates that two attached sheets are being fed or a single sheet is being damaged. The document transport apparatus is shut off to prevent damage to documents when it has been determined that two attached sheets are being fed or that a single sheet is being damaged.

Abstract: A method of determining document characteristics prior to processing the document in a document scanner (10) includes capturing an input image of documents in an input tray (20); transmitting the images to a processor (65); determining characteristics of the documents; and processing the document based on the characteristics.

Abstract: A document scanner (10) includes an input tray (20) for holding documents (11); an input image capture device (30) for capturing images of documents in the input tray; an image processor for determining characteristics of documents; and the documents are processed based on the characteristics.

Abstract: An apparatus for classifying documents (5) based on sound includes a document transport (30) for transporting a document; an audio transducer (20) for detecting a sonic profile produced by the document as it is transported; and a controller for determining document characteristics based on the sonic profile.

Abstract: A method to identify and classify a document (5) by weight or thickness based on the sound the document makes while moving through a document transport (30). Using an audio transducer (20), the sound of the document is captured and compared to previously saved and stored characteristics of various weighted documents and then classified when matched to a specific set of characteristics.