sentence-transformers How to use petkopetkov/e5-large-v2-patent with sentence-transformers:
from sentence_transformers import SentenceTransformer
model = SentenceTransformer("petkopetkov/e5-large-v2-patent")
sentences = [
"query: OPTIMIZED ALLOCATION OF FUNCTIONS IN HYBRID MOTOR CONTROLLER IMPLEMENTATIONS A system for controlling a motor (14) with a plurality of motor control functions including at least a current control loop and a velocity control loop. The system includes one of a hybrid Digital Signal Processor (DSP)-Field Programmable Gate Array (FPGA) architecture having an integral DSP and an integral FPGA or a System on a Chip (SoC) architecture having a Microcontroller Sub-System (MSS) and an FPGA fabric. The current control loop function is assigned to the integral FPGA for the hybrid DSP-FPGA architecture, and at least the velocity control loop function is assigned to the DSP the hybrid DSP-FPGA architecture. Alternatively, the current control loop function is assigned the FPGA fabric of the SoC architecture, and at least the velocity control loop function is assigned to the MSS of the SoC architecture. A method of allocating motor control functions for a hybrid Digital Signal Processor (DSP)-Field Programmable Gate Array (FPGA) architecture having an integral DSP and an integral FPGA or a System on a Chip (SoC) architecture having a Microcontroller Sub-System (MSS) and an FPGA fabric, the method comprising: identifying a plurality of motor control functions to be allocated the plurality of motor control functions including at least a current control loop and a velocity control loop; selecting at least one functional requirement for each motor control function of the plurality of motor control functions; determining how the at least one functional requirement would be performed most efficiently by either the integral DSP or integral FPGA of the hybrid DSP-FPGA architecture or the MSS and an FPGA fabric of the SoC architecture; and assigning the motor control function to only one of the integral DSP or integral FPGA of the hybrid DSP-FPGA architecture or the MSS and an FPGA fabric of the SoC architecture, wherein at least the current control loop function is assigned to the integral FPGA for the hybrid DSP-FPGA architecture, and at least the velocity control loop function is assigned to the DSP the hybrid DSP-FPGA architecture; or wherein at least the current control loop function is assigned the FPGA fabric of the SoC architecture, and at least the velocity control loop function is assigned to the MSS of the SoC architecture. The method of claim 1, further including identifying another of the motor control functions of the plurality of motor control functions as motor position sensing and allocating the position sensing to the FPGA the hybrid DSP-FPGA architecture or the FPGA fabric of the SoC architecture. --> The method of claim 1 or 2, further including identifying another of the motor control functions of the plurality of motor control functions as motor position control loop and allocating the motor position control loop to the DSP of the hybrid DSP-FPGA architecture or the MSS of the SoC architecture. The method of any preceding claim, further including identifying another of the motor control functions of the plurality of motor control functions as Continuous Built in Test (CBIT) and allocating a portion of the CBIT to the DSP of the hybrid DSP-FPGA architecture or the MSS of the SoC architecture; and optionally further including identifying another of the motor control functions of the plurality of motor control functions as Continuous Built in Test function (CBIT) and allocating a portion of the CBIT function to the FPGA of the hybrid DSP-FPGA architecture or the FPGA fabric of the SoC architecture; and optionally wherein the portion of the CBIT to the FPGA of the hybrid DSP-FPGA architecture or the FPGA fabric of the SoC architecture is based on a processing rate of the CBIT function. The method of any preceding claim, further including identifying another of the motor control functions of the plurality of motor control functions as current sensor processing and allocating the current sensor processing to the FPGA of the hybrid DSP-FPGA architecture or the FPGA fabric of the SoC architecture. The method of any preceding claim, further including identifying another of the motor control functions of the plurality of motor control functions as PWM processing and allocating the PWM processing to the DSP of the hybrid DSP-FPGA architecture or the MSS of the SoC architecture. The method of any preceding claim, further including identifying another of the motor control functions of the plurality of motor control functions as DC Bus processing and allocating the DC bus processing to the FPGA of the hybrid DSP-FPGA architecture or the FPGA fabric of the SoC architecture. The method of any preceding claim, further including identifying another of the motor control functions of the plurality of motor control functions all --> communications processing and allocating the communications processing to the DSP of the hybrid DSP-FPGA architecture or the MSS of the SoC architecture. The method of any preceding claim, further including identifying another of the motor control functions of the plurality of motor control functions as the system state machine and allocating the system state machine functionality processing to the DSP of the hybrid DSP-FPGA architecture or the MSS of the SoC architecture. A system for controlling a motor with a plurality of motor control functions including at least a current control loop and a velocity control loop, the system comprising:one of: a hybrid Digital Signal Processor (DSP)-Field Programmable Gate Array (FPGA) architecture having an integral DSP and an integral FPGA; or a System on a Chip (SoC) architecture having a Microcontroller Sub-System (MSS) and an FPGA fabric; wherein at least the current control loop function is assigned to the integral FPGA for the hybrid DSP-FPGA architecture, and at least the velocity control loop function is assigned to the DSP the hybrid DSP-FPGA architecture; or wherein at least the current control loop function is assigned the FPGA fabric of the SoC architecture, and at least the velocity control loop function is assigned to the MSS of the SoC architecture. The system for controlling a motor of claim 10, further including motor position sensing as another motor control function of the plurality of motor control functions and allocating the motor position sensing to the FPGA the hybrid DSP-FPGA architecture or the FPGA fabric of the SoC architecture; and/or further including a motor position control loop function as another motor control functions of the plurality of motor control functions and allocating the motor position control loop function to the DSP of the hybrid DSP-FPGA architecture or the MSS of the SoC architecture; and/or further --> including a Continuous Built in Test (CBIT) function as another motor control functions of the plurality of motor control functions and allocating a at least a portion of the CBIT to the DSP of the hybrid DSP-FPGA architecture or the MSS of the SoC architecture; and/or further including allocating at least portion of the CBIT function to the FPGA of the hybrid DSP-FPGA architecture or the FPGA fabric of the SoC architecture, wherein the portion of the CBIT to the FPGA of the hybrid DSP-FPGA architecture or the FPGA fabric of the SoC architecture is based on a processing rate of the CBIT function; and/or further including a current sensor processing function as another motor control functions of the plurality of motor control functions and allocating the current sensor processing to the FPGA of the hybrid DSP-FPGA architecture or the FPGA fabric of the SoC architecture; and/or further including a PWM processing function as another motor control functions of the plurality of motor control functions and allocating the PWM processing to the DSP of the hybrid DSP-FPGA architecture or the MSS of the SoC architecture; and/or further including a system state machine function as another motor control functions of the plurality of motor control functions and allocating the system state machine function processing to the DSP of the hybrid DSP-FPGA architecture or the MSS of the SoC architecture. A motor drive system, the motor drive system comprising: a power source (12); a drive (20) operably connected to the power source, the drive including a controller; a motor (14) operably connected to the drive, the motor having a plurality of motor control functions including at least a current control loop and a velocity control loop, the system comprising:wherein the controller includes one of: a hybrid Digital Signal Processor (DSP)-Field Programmable Gate Array (FPGA) architecture having an integral DSP and an integral FPGA --> a System on a Chip (SoC) architecture having a Microcontroller Sub-System (MSS) and an FPGA fabric; wherein at least the current control loop function is assigned to the integral FPGA for the hybrid DSP-FPGA architecture, and at least the velocity control loop function is assigned to the DSP the hybrid DSP-FPGA architecture; or wherein at least the current control loop function is assigned the FPGA fabric of the SoC architecture, and at least the velocity control loop function is assigned to the MSS fabric of the SoC architecture.",
"passage: TELEHANDLER According to one embodiment, the application relates to a telehandler (100), which comprises a chassis section (110) and a boom assembly (120) supported on the chassis section and capable of being fitted with a liftable attachment (130). The telehandler further comprises a control unit (112), including chassis control means (114) for controlling the chassis section and/or boom assembly control means (116) for controlling the boom assembly. The control unit is adapted to be operated from outside the unmanned telehandler. A telehandler (100), comprisinga chassis section (110),a boom assembly (120) supported on the chassis section and capable of being fitted with a liftable attachment (130), anda control unit (112), including chassis control means (114) for controlling the chassis section and/or boom assembly control means (116) for controlling the boom assembly,characterized in thatthe control unit is adapted to be operated from outside the unmanned telehandler. A telehandler according to claim 1, wherein the control unit comprises a portable control device (118), which is in telecommunication (119) with the control unit and enables the chassis section and/or the boom assembly to be controlled from a certain distance outside the chassis section. A telehandler according to any of the preceding claims, wherein the boom assembly includes a telescopic boom (122) and a linkage (124), which is supported on a boom assembly slewing mechanism and which is adapted to operate the boom. A telehandler according to claim 3, wherein the boom assembly further includes a jib (125), which is connected to the boom and which is adapted to operate a liftable attachment in such a way that the liftable attachment rotates with respect to the vertical axis of its connection point (126). A telehandler according to claim 4, wherein the liftable attachment is connected to the jib and electrically coupled with the control unit by means of the boom assembly for controlling the liftable attachment with the boom assembly control means. A telehandler according to any of the preceding claims, wherein the liftable attachment is a lifting fork (130). A telehandler according to claim 6, which is further provided with a man basket (150) which is releasably attachable to the lifting fork and electrically --> connectible to the control unit by way of the boom assembly and by means of a connection unit (127) of the jib. A telehandler according to any of the preceding claims, wherein the basket comprises fastening elements (151 a, 151b) capable of having the lifting fork attached thereto, and locking elements (132) by means of which the lifting fork is capable being locked securely to the fastening elements and the locking status of which are electrically observed with monitoring elements. A telehandler according to any of the preceding claims, wherein the basket is provided with a basket control unit (152), including chassis control means (154) for controlling the chassis section and boom assembly control means (156) for controlling the boom assembly and the lifting fork connected thereto, said basket control unit being electrically connected to the control unit. A man basket (150) for attachment to a telehandler (100) according to any of claims 1-7, said basket comprisingfastening elements (151 a, 151 b) capable of having a lifting fork (130) attached thereto,locking elements (132) for locking the lifting fork securely to the fastening elements, andobservation elements for electrically monitoring the locking status of the locking elements.",
"passage: TRANSMISSION DEVICE, TRANSMISSION METHOD, RECEPTION DEVICE, AND RECEPTION METHOD A normal frame rate of image data and a high frame rate of image data are favorably transported.A base stream including, as an access unit, encoded image data per picture in a base frame rate of image data acquired by performing blending processing in units of temporally successive two pictures in the high frame rate of image data, is acquired and additionally an enhanced stream including, as an access unit, encoded image data per picture in the high frame rate of image data, is acquired. A container in a predetermined format is transmitted, the container including the base stream and the enhanced stream. A transmission device comprising: an image encoding unit configured to acquire a base stream including, as an access unit, encoded image data per picture in a base frame rate of image data acquired by performing blending processing in units of temporally successive two pictures in a high frame rate of image data, the image encoding unit being configured to acquire an enhanced stream including, as an access unit, encoded image data per picture in the high frame rate of image data; and a transmission unit configured to transmit a container in a predetermined format, the container including the base stream and the enhanced stream. The transmission device according to claim 1, further comprising:an information inserting unit configured to insert blending ratio information in the blending processing, into a layer of the enhanced stream. The transmission device according to claim 2,wherein the base stream and the enhanced stream each have a NAL unit structure, andthe information inserting unit inserts a SEI NAL unit having the blending ratio information, into the enhanced stream. The transmission device according to claim 2,wherein the base stream and the enhanced stream each have a NAL unit structure, and -->the information inserting unit inserts the blending ratio information into a PPS NAL unit of the enhanced stream. The transmission device according to claim 1, further comprising:an information inserting unit configured to insert, into each access unit of the enhanced stream, phase information indicating to which of the temporally successive two pictures the access unit corresponds. The transmission device according to claim 1, further comprising:an information inserting unit configured to insert, into a layer of the container, identification information indicating that the image data included in the base stream includes the image data acquired by the performance of the blending processing. The transmission device according to claim 1, wherein the image encoding unit performs prediction encoding processing for the base frame rate of image data, to the base frame rate of image data, so as to acquire the base stream, the image encoding unit being configured to perform, with the high frame rate of image data, processing inverse to the blending processing, to the base frame rate of image data, so as to acquire image data as after-blend-compensation image data, the image data including, when the high frame rate of image data includes image data of one-side pictures in the units of temporally successive two pictures, image data of the other-side pictures, the image encoding unit being configured to perform prediction encoding processing with the --> after-blend-compensation image data, to the high frame rate of image data, so as to acquire the enhanced stream. The transmission device according to claim 7, wherein the image encoding unit acquires, per predicted block in the high frame rate of image data, image data over a range of more than the predicted block, as the after-blend-compensation image data. A transmission method comprising: an image encoding step of acquiring a base stream including, as an access unit, encoded image data per picture in a base frame rate of image data acquired by performing blending processing in units of temporally successive two pictures in a high frame rate of image data, and acquiring an enhanced stream including, as an access unit, encoded image data per picture in the high frame rate of image data; and a transmission step of transmitting a container in a predetermined format by a transmission unit, the container including the base stream and the enhanced stream. A reception device comprising: a reception unit configured to receive a container in a predetermined format, the container including a base stream and an enhanced stream, the base stream being acquired by performing prediction encoding processing for a base frame rate of image data, to the base frame rate of image data acquired by performing blending processing in units of temporally successive two pictures in a high frame rate of image, the enhanced stream being acquired by performing prediction encoding processing with --> after-blend-compensation image data, to the high frame rate of image data, the after-blend-compensation image data being acquired by performing, with the high frame rate of image data, processing inverse to the blending processing, to the base frame rate of image data, the after-blend-compensation image data including, when the high frame rate of image data includes image data of one-side pictures in the units of temporally successive two pictures, image data of the other-side pictures; and a processing unit configured to process only the base stream so as to acquire the base frame rate of image data or both of the base stream and the enhanced stream so as to acquire the high frame rate of image data, wherein, when performing decoding processing to the enhanced stream, the processing unit performs, with the high frame rate of image data acquired by the processing of the enhanced stream, the processing inverse to the blending processing, to the base frame rate of image data acquired by the processing of the base stream, so as to acquire the after-blend-compensation image data including, when the high frame rate of image data includes the image data of the one-side pictures in the units of temporally successive two pictures, the image data of the other-side pictures, the processing unit being configured to use the after-blend-compensation image data as reference image data. The reception device according to claim 10,wherein a layer of the enhanced stream includes blending ratio information in the blending processing, inserted, andthe processing unit uses the blending ratio information --> in performing the processing inverse to the blending processing. The reception device according to claim 10,wherein each access unit in the enhanced stream includes phase information indicating to which of the temporally successive two pictures the access unit corresponds, inserted, andthe processing unit uses the phase information in performing the processing inverse to the blending processing. A reception method comprising: a reception step of receiving a container in a predetermined format by a reception unit, the container including a base stream and an enhanced stream, the base stream being acquired by performing prediction encoding processing for a base frame rate of image data, to the base frame rate of image data acquired by performing blending processing in units of temporally successive two pictures in a high frame rate of image, the enhanced stream being acquired by performing prediction encoding processing with after-blend-compensation image data, to the high frame rate of image data, the after-blend-compensation image data being acquired by performing, with the high frame rate of image data, processing inverse to the blending processing, to the base frame rate of image data, the after-blend-compensation image data including, when the high frame rate of image data includes image data of one-side pictures in the units of temporally successive two pictures, image data of the other-side pictures; and a processing step of processing only the base stream --> so as to acquire the base frame rate of image data or both of the base stream and the enhanced stream so as to acquire the high frame rate of image data, wherein, in the processing step, when decoding processing is performed to the enhanced stream, with the high frame rate of image data acquired by the processing of the enhanced stream, the processing inverse to the blending processing is performed to the base frame rate of image data acquired by the processing of the base stream, so as to acquire the after-blend-compensation image data including, when the high frame rate of image data includes the image data of the one-side pictures in the units of temporally successive two pictures, the image data of the other-side pictures, and the after-blend-compensation image data is used as reference image data. A reception device comprising: a reception unit configured to receive a container in a predetermined format, the container including a base stream and an enhanced stream, the base stream being acquired by performing encoding processing to a base frame rate of image data acquired by performing blending processing in units of temporally successive two pictures in a high frame rate of image data, the enhanced stream being acquired by performing encoding processing to the high frame rate of image data; and a processing unit configured to process only the base stream so as to acquire the base frame rate of image data or both of the base stream and the enhanced stream so as to acquire the high frame rate of image data.",
"passage: RESOURCE CONFIGURATION SYSTEM, RESOURCE CONFIGURATION METHOD AND RESOURCE CONFIGURATION PROGRAM The present invention provides a cloud service achieving high processing performance specialized in particular processing, image processing, or parallel processing. A resource selection apparatus 1 selects a computational resource from a plurality of computational resources including at least an FPGA or a GPU and a provisioning method from a plurality of provisioning methods, based on whether a performance requirement and a functional requirement from a user require that particular computational processing, image processing, or parallel processing be performed with processing performance of a certain level or higher. A resource configuration system comprising: a resource selection apparatus that selects a resource on a cloud; and a resource reconfiguration apparatus that configures a resource or reconfigures a configuration of the resource, wherein the resource selection apparatus includes a reception unit that receives a requirement for the resource from a user, and a selection unit that selects a computational resource from a plurality of computational resources including at least an FPGA or a GPU and a provisioning method from a plurality of provisioning methods, based on whether the requirement requires that any of particular processing, image processing, and parallel processing be performed with processing performance of a certain level or higher. The resource configuration system according to claim 1, whereinwhen the requirement requires that the particular processing be performed with the processing performance of the certain level or higher, the selection unit selects the FPGA as the computational resource and selects bare-metal provisioning as the provisioning method for the computational resource. --> The resource configuration system according to claim 1, whereinwhen the requirement requires that the image processing or the parallel processing be performed with the processing performance of the certain level or higher, the selection unit selects the GPU as the computational resource and selects bare-metal provisioning or container provisioning as the provisioning method for the computational resource, based on the level of the processing performance to be achieved by the computational resource or whether OS customization is necessary. The resource configuration system according to claim 1, whereinthe selection unit selects a block-based provisioning method or an object-based provisioning method based on the requirement, and selects a storage resource according to a characteristic of an application program to run. The resource configuration system according to claim 1, whereinthe resource reconfiguration apparatus includes a configuration unit that sets, in a computational resource with an FPGA, a computation logic suitable for the requirement from the user, and causes the computational resource to use the computation logic. --> The resource configuration system according to claim 1, whereinthe resource reconfiguration apparatus includes a collection unit that collects usage frequency for each of various kinds of computation processing that the user is using on an already-configured computational resource, and a reconfiguration unit that changes the configuration of the already-configured computational resource as suited for particular computation processing the usage frequency of which has increased. A resource configuration method performed by a resource selection apparatus that selects a resource on a cloud and a resource reconfiguration apparatus that configures a resource or reconfigures a configuration of the resource, the method comprising, by the resource selection apparatus: receiving a requirement for the resource from a user; and selecting a computational resource from a plurality of computational resources including at least an FPGA or a GPU and a provisioning method from a plurality of provisioning methods, based on whether the requirement requires that any of particular processing, image processing, and parallel processing be performed with processing performance of a certain level or higher. The resource configuration method according to claim 7, comprising, by the resource reconfiguration apparatus: --> collecting usage frequency for each of various kinds of computation processing that the user is using on an already-configured computational resource; and changing the configuration of the already-configured computational resource as suited for particular computation processing the usage frequency of which has increased. A resource configuration program that causes a computer to function as the resource configuration system according to claim 1."
]
embeddings = model.encode(sentences)
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [4, 4] 
