Abstract: Some embodiments provide a neural network inference circuit (NNIC) for executing a network having multiple layers. The NNIC includes multiple circuit sets. Each circuit set includes a dot product circuit to compute dot products between weight values and activation values for at least a subset of a first set of the layers, a math function circuit to compute values based on computations using activation values for at least a subset of a second set of the layers, and a post-processing circuit to receive (i) values output by the dot product circuit and (ii) values output by the math function circuit and to perform post-processing operations on the received values. The NNIC includes a set of accumulation circuits. Each accumulation circuit is to accumulate outputs of math function circuits for layers of the second set of layers that perform matrix multiplication of sets of activation values output by previous layers.

Abstract: Some embodiments provide a method for executing a machine-trained (MT) network that includes multiple layers. For an input set for the network divided into at least two blocks of input data, the method propagates each respective block of the input data separately through a first set of the layers of the MT network to generate respective blocks of intermediate data. The method combines the blocks of intermediate data into a set of intermediate data. The method propagates the set of intermediate data together through a second set of the layers of the MT network to generate output data for the input set.

Abstract: A vehicle can include one or more processing circuits. The one or more processing circuits can determine that one or more operations include a number of vehicle backups performed by the vehicle, provide a user interface that includes an indication of the number of vehicle backups, present a prompt for input to identify one or more factors that contributed to an occurrence of at least one vehicle backup of the number of vehicle backups, receive at least one response to the prompt that identifies the one or more factors, determine that an opportunity to avoid subsequent occurrences of the at least one vehicle backup exists, and update the user interface to display an indication of the opportunity to avoid the subsequent occurrences of the at least one vehicle backup.

Abstract: A system for monitoring operation of a vehicle. The system includes a device configured to collect data pertaining to operations of the vehicle and one or more processing circuits in communication with the device. The one or more processing circuits configured to receive, from the device responsive to a first operation of the vehicle, data pertaining to the first operation of the vehicle, determine, using the data pertaining to the first operation of the vehicle, that the first operation of the vehicle includes a vehicle backup, and provide, to a user device, a user interface that displays an indication of the vehicle backup.

Abstract: A method is provided for the synthesis of (-)-trans-?9-tetrahydrocannabivarin (?9-THCV) and analogs thereof such that in the reaction product, the molar ratio of the ?9 isomer to incidentally formed ?8 isomers is greater than 4:1. Synthesis is carried out by combining a selected cannabinoid reactant, e.g., cannabidivarin (CBDV) or an analog thereof, with an acid in a solvent for the cannabinoid reactant, wherein the acid comprises (i) a Lewis acid having an acid softness index value in the range of ?10 ?G0f, Mn+ to ?150 ?G0f, Mn+, (ii) a Brønsted acid having a pKa in the range of ?4.0 to +4.0, or (iii) a combination of (i) and (ii), under reaction conditions comprising a reaction temperature in the range of ?0° C. to 25° C. and a reaction time in the range of 1 hour to 24 hours. The reaction is thereafter quenched with base and the solvent removed, wherein the crude reaction product so provided may be purified, e.g., chromatographically purified.

Abstract: A refuse vehicle includes a first controller area network (CAN) bus for body functions of the refuse vehicle. The refuse vehicle also includes multiple controllable elements of a body of the refuse vehicle communicably coupled with the first CAN bus. The refuse vehicle includes a second CAN bus for chassis functions of the refuse vehicle. The refuse vehicle includes multiple controllable elements of a chassis of the refuse vehicle communicably coupled with the second CAN bus. The refuse vehicle includes a telematics module communicably coupled with both the first CAN bus and the second CAN bus. The telematics module is configured to monitor communications on both the first CAN bus and the second CAN bus and transmit the communications to a cloud computing system. The first CAN bus and the second CAN bus are communicatively separate and do not communicate with each other directly.

Abstract: Pigment compositions and fibers may include a single pigment component comprising a cobalt titanium oxide complex or a tin titanium zinc oxide complex. The pigment compositions may additionally include one or more additives having compatibilizer, adhesion promoting, and/or dispersion aiding capabilities, and a carrier resin.

Abstract: An airbag inflator assembly includes a cylindrical housing and internal components disposed in the cylindrical housing. The internal components include an initiator assembly, a generant, an internal nozzle, and a canister downstream of the internal nozzle. A center vent is secured to a downstream end of the cylindrical housing and includes a central passageway adjacent the canister for products of combustion and an exit passage that directs the products of combustion out of the assembly. The internal components and the center vent define a tortuous path for the products of combustion.

Abstract: An airbag inflator assembly includes a cylindrical housing, an initiator assembly secured in an upstream end of the cylindrical housing, and a generant disposed in the cylindrical housing adjacent the initiator assembly. Internal components define a tortuous path for products of combustion and define a plurality of debris collection areas. A rupture disk assembly is secured over an exit aperture, and a diffuser region downstream of the rupture disk assembly includes an output opening through which the products of combustion are vented into the airbag.

Abstract: A refuse vehicle includes a first controller area network (CAN) bus for body functions of the refuse vehicle. The refuse vehicle also includes multiple controllable elements of a body of the refuse vehicle communicably coupled with the first CAN bus. The refuse vehicle includes a second CAN bus for chassis functions of the refuse vehicle. The refuse vehicle includes multiple controllable elements of a chassis of the refuse vehicle communicably coupled with the second CAN bus. The refuse vehicle includes a telematics module communicably coupled with both the first CAN bus and the second CAN bus. The telematics module is configured to monitor communications on both the first CAN bus and the second CAN bus and transmit the communications to a cloud computing system. The first CAN bus and the second CAN bus are communicatively separate and do not communicate with each other directly.

Abstract: A system for verifying refuse collection includes a camera system coupled to a refuse vehicle and configured to detect an object on one or more sides of the refuse vehicle and one or more memory devices storing instructions thereon that, when executed by one or more processors, cause the one or more processors to receive image data from the camera system; transmit the image data to an object detection system; determine whether the object is present at a location, whether the object is expected to be present for collection at the location, and whether the object is collected; and instruct the camera system to store an image of the location on the one or more memory devices.

Abstract: A vehicle control system for a refuse vehicle includes one or more vehicle sensors configured to obtain vehicle sensor data relating to the refuse vehicle, one or more wearable devices communicatively coupled to the one or more vehicle sensors and configured to obtain wearable data, and processing circuitry configured to obtain the vehicle sensor data and the wearable data, determine a position of the one or more wearable devices relative to the one or more vehicle sensors, and based on the position, control one or more features of the refuse vehicle. A charging system includes one or more wireless charging devices located at a position corresponding to the one or more wearable devices.

Abstract: A vocational vehicle includes a chassis, a cab supported by the chassis, a body supported by the chassis and defining a compartment, a lift assembly coupled to the body so that the lift assembly is configured to move between a lowered position and a raised position along a path, a sensor defining a field of view that at least partially includes the path of the lift assembly, and a vehicle control system in communication with the lift assembly and the sensor. The vehicle control system includes a controller having a processor and at least one memory. The controller is configured to detect that the lift assembly is activated to move and disable the sensor.

Abstract: Some embodiments provide a compiler for optimizing the implementation of a machine-trained network (e.g., a neural network) on an integrated circuit (IC). The compiler of some embodiments receives a specification of a machine-trained network including multiple layers of computation nodes and generates a graph representing options for implementing the machine-trained network in the IC. In some embodiments, the compiler also generates instructions for gating operations. Gating operations, in some embodiments, include gating at multiple levels (e.g., gating of clusters, cores, or memory units). Gating operations conserve power in some embodiments by gating signals so that they do not reach the gated element or so that they are not propagated within the gated element. In some embodiments, a clock signal is gated such that a register that transmits data on a rising (or falling) edge of a clock signal is not triggered.

Abstract: Described herein are methods, compositions, and systems derived from uncultivated microorganisms useful for gene editing.

Abstract: Some embodiments provide a compiler for optimizing the implementation of a machine-trained network (e.g., a neural network) on an integrated circuit (IC). The compiler of some embodiments receives a specification of a machine-trained network including multiple layers of computation nodes and generates a graph representing options for implementing the machine-trained network in the IC. In some embodiments, the graph includes nodes representing options for implementing each layer of the machine-trained network and edges between nodes for different layers representing different implementations that are compatible. In some embodiments, the graph is populated according to rules relating to memory use and the numbers of cores necessary to implement a particular layer of the machine trained network such that nodes for a particular layer, in some embodiments, represent fewer than all the possible groupings of sets of clusters.

Abstract: A refuse vehicle includes a chassis, a body supported on the chassis, a pump, a tank configured to supply hydraulic fluid to the pump, and a service indicator configured to provide an indication of a fluid level within the tank. The service indicator includes a light, a graphic on a display, or a symbol in an augmented reality overlay. The service indicator is configured to provide the indication of the fluid level within the tank in response to a request signal.

Abstract: Described herein are methods, compositions, and systems derived from uncultivated microorganisms useful for gene editing.

Abstract: Some embodiments provide a method for generating neural network program instructions for a neural network inference circuit to execute a neural network. The neural network inference circuit includes a particular amount of available memory. The method receives a specification of the neural network including multiple layers. The method determines (i) a required amount of weight memory for the neural network and (ii) required amounts of activation memory for each of a set of layers of the neural network. When the required amount of weight memory and the required amount of activation memory for at least one layer is greater than the particular amount of available memory, the method generates the program instructions for the neural network inference circuit to execute a first set of the layers of the neural network multiple times for different blocks of input data and execute a second set of the layers in a single pass.

Abstract: The present disclosure provides for endonuclease enzymes as well as methods of using such enzymes or variants thereof.