Abstract: A method for characterizing an effective dielectric constant of a substrate including glass fibers, including: constructing a real model, where the real model includes: the substrate, a glass fiber-woven layer on the substrate, and a transmission line trace on the substrate; performing a phase shift simulation on the real model according to phase shift value; and obtaining first information; constructing a simplified model, where the simplified model includes: the substrate, and the transmission line trace on the substrate; setting a dielectric constant of the substrate as a variable; performing parameter scanning simulation; and obtaining a second information; and according to the first and second information, determining a dielectric constant value when the phase shift value of the simplified model equals the phase shift value of the real model, and using the dielectric constant value as the effective dielectric constant of the substrate including the glass fibers.

Abstract: Provided is a method and system for intelligent segmentation of fundus blood vessels based on retinal angiography. The method includes: acquiring fundus images with retinal angiography technology; performing a preprocessing operation on the fundus images to obtain preprocessed images; performing a superpixel segmentation on the preprocessed images to obtain segmented areas; performing a fundus blood vessel recognition on the segmented areas to obtain a recognition result; and constructing a trained fundus blood vessel segmentation model, and performing a feature segmentation on the recognition result by using the fundus blood vessel segmentation model to obtain a segmentation result. This method performs a multi-level feature extraction and fusion through the intelligent segmentation model, realizes effective extraction of multi-dimensional features of the fundus images, such as color, texture, shape and intensity, and improves segmentation accuracy through the feature fusion.

Abstract: In example implementations described herein, there are systems and methods for training a machine-trained model for industrial failure detection based on an initial set of training data comprising a plurality of input-output data sets each comprising input data and output data regarding at least one classification of the input data. The method may include identifying at least one classification of the input data, in a plurality of classifications that is under-represented. The method may also include automatically generating, for inclusion in a modified set of training data, additional input-output data sets for the identified at least, one classification to balance a representation of the classifications in the modified set of training data. Finally, the method may include training the machine-trained model based on the modified set of training data comprising at least a subset of the initial set of training data and the additional input-output data sets.

Abstract: Provided is an automatic calculation method for a surface defect area of a wet blue skin based on a DeeplabV3+ model, aiming at solving the problem that the automatic calculation of the surface defect area is difficult in a tannery. The method includes image data acquisition and preprocessing of a wet blue skin, training and verification of a wet blue skin surface defect segmentation model based on a DeeplabV3+ model, and automatic calculation of the surface defect area in a segmented image. The method can accurately segment common defects such as knife holes, brands and the like on the surface of a large-area wet blue skin, with a high speed of detecting a wet blue skin within 1 s.

Abstract: The present disclosure provides a model compiling method. The method includes that: a target computation sub-graph in a computation graph corresponding to a model to be compiled is obtained and the target computation sub-graph includes multiple memory-access-intensive operators and data corresponding to the multiple memory-access-intensive operators; multiple groups of computation tasks to be serially executed by a target thread are determined based on multiple threads set in an executor of the model and a computation task parallelism degree of each thread sequential rearrangement is performed on multiple computation tasks contained in the any group of computation tasks, and an execution instruction sequence both corresponding to the target thread and the any group of computation tasks is generated; and an executable program file for executing the model by the executor is determined based on the execution instruction sequence corresponding to the any group of computation tasks.

Abstract: A method for generating persona specific insights. The method may include receiving sensor data associated with a device; extracting features from the received sensor data; processing the features using a machine learning model to generate machine learning metrics; ingesting the machine learning metrics and the features to generate insights data associated with the device; generating personas data using the insights data and the features, and mapping the insights to the personas data; generating custom insights using the insights data, the personas data, and the features, wherein the custom insights are text-based summaries; and disseminating each of the custom insights to respective persona of the personas data to place service orders associated with the device.

Abstract: Systems and methods described herein involve generating an asset hierarchy from a plurality of images of a plurality of asset, which can involve executing a machine learning process to identify the plurality of assets from the plurality of images; identifying relationships between the identified plurality of assets based on feature extraction and analysis of extracted features of the identified plurality of assets; and generating the asset hierarchy from one or more asset hierarchy templates derived from the identified relationships between the identified plurality of assets.

Abstract: A communication method and apparatus are provided, are applied to a wireless local area network system that supports the 802.11 series protocols, for example, a next-generation Wi-Fi protocol of IEEE 802.11ax, such as 802.11be, Wi-Fi 7, or EHT, for another example, a next-generation of 802.11be, such as 802.11bn or Wi-Fi 8, 802.11ay, or 802.11aj, and may be further applied to a UWB-based wireless personal area network system, a sensing (sensing) system, and the like. A first communication apparatus sends a PPDU, and a second communication apparatus receives the PPDU. The PPDU corresponds to at least one target user, the PPDU includes a data field corresponding to the target user, and a parameter NSD,short corresponding to the target user is determined based on an RU/MRU used to transmit the data field and an MCS of the data field. This effectively reduces encoding/decoding time and improves encoding/decoding efficiency.

Abstract: A pharmaceutical combination including a Raf inhibitor which is Compound of formula (I) or a pharmaceutically acceptable salt thereof, (b) a KRAS G12C inhibitor, and/or (c) trametinib.

Abstract: A semiconductor device includes a substrate, a buffer layer, a gradient layer, an active layer, a window layer, and an optical filtering layer. The substrate includes a first element and a second element. The buffer layer is disposed on the substrate. The gradient layer is formed on the buffer layer, and includes sublayers. Each sublayer includes the first, second, and third elements. For each sublayer, a lattice constant thereof is adjusted by changing a ratio of the second element to the third element. The active layer is formed on the gradient layer, and includes the first, second, and third elements. The window layer is formed on the active layer. The optical filtering layer includes the first, second, and third elements, and is formed on the window layer to block a portion of light having a wavelength in a predetermined wavelength range.

Abstract: An intraocular lens (IOL) suspension structure includes a first fixing plate and a second fixing plate, fixedly interconnected by guide columns. A first moving plate and a second moving plate, slidably matched with the guide columns, are arranged between the two fixing plates. Curved needles are arranged on a left end face of the first moving plate through clamping mechanisms, and the clamping mechanisms are arranged on a right end face of the second moving plate. Auxiliary holders are arranged between the first fixing plate and the second fixing plate. The IOL suspension structure is ingeniously designed, functionally practical, and structurally simple. It effectively solves the problem in existing devices where curved needles on a first movable stage often fail to realign with connecting tubes on a second movable stage after passing through the eyeball and an IOL, significantly compromising surgical efficiency.

Abstract: Disclosed are an optimization method for distributed execution of a deep learning task and a distributed system. The method includes that: a computation graph is generated based on a deep learning task and hardware resources are allocated for the distributed execution of the deep learning task; the allocated hardware resources are grouped to obtain at least one grouping scheme; for each grouping scheme, tensor information related to multiple operators contained in the computation graph is split based on the value of at least one factor under this grouping scheme to obtain multiple candidate splitting solutions; and an optimal efficiency solution for executing the deep learning task of the hardware resources is selected by using a cost model. Through operator splitting based on device grouping combined with optimization solving based on the cost model, automatic optimization of distributed execution for various deep learning tasks is realized.

Abstract: A method for calibrating the reception of a phased array antenna with a built-in local oscillator and an electronic device, are provided. The method includes: constructing and configuring a test system, where a transmitting antenna is located on the central axis of a phased array antenna, and the transmitting antenna is connected to a signal source; dividing the phased array antenna into a plurality of centrally symmetric ring areas according to the central axis of the phased array antenna, and each ring area has one or more centrally symmetric sub-units, and each sub-unit includes a plurality of antenna channels of the same number; and receiving signals using the phased array antenna and transmitting signals with the transmitting antenna, and performing amplitude and phase calibrations between antenna channels within sub-units of the ring areas, between sub-units of the ring areas, and between different ring areas.

Abstract: A method for calibrating the transmission of a phased array antenna with a built-in local oscillator and an electronic device are provided. The method includes: constructing and configuring a test system, where a receiving antenna is located on a central axis of the phased array antenna; dividing the phased array antenna according to a layout of the phased array antenna, where the layout of the phased array antenna is divided into centrally symmetric ring areas, and each of the ring areas includes one or more centrally symmetric sub-units, and each of the sub-units includes a same number of antenna channels; and transmitting signals using the phased array antenna and receiving signals by the receiving antenna, and performing amplitude and phase calibration between the antenna channels within the sub-units of the ring areas, between the sub-units of the ring areas, and between the ring areas.

Abstract: A catalytic cracking catalyst contains 10-70 wt % of a cracking active component, 10-60 wt % of a binder and 10-70 wt % of a clay. The cracking active component has 5-100 wt % of a first Y-type molecular sieve and 0-95 wt % of a second molecular sieve. The first Y-type molecular sieve is a modified molecular sieve based on the crystal modification of kaolin and has the sodium oxide content of less than 2 wt %. The process for preparing the catalyst includes the steps of vigorously mixing and stirring a cracking active component comprising a modified molecular sieve based on the crystal modification of kaolin, a binder and a clay with water; spray drying; washing; filtering; and drying. The catalyst is used in the catalytic cracking reaction of heavy oils, and has a good coke selectivity, as well as a higher heavy oil conversion rate.

Abstract: A method for detecting visual anomalies.

Abstract: A ZSM-5/? core-shell molecular sieve has a core composed of at least two crystal grains of ZSM-5 molecular sieve and a shell composed of a plurality of crystal grains of ? molecular sieve. The ZSM-5 molecular sieve grains has an average grain size of 0.05-15 ?m. The core-shell molecular sieve has a shell coverage of 50-100%, a shell thickness of 10-2000 nm, and an average grain size of the ? molecular sieve grains in the shell of 10-500 nm. A ratio of a height of a diffraction peak at 2?=22.4° to a height of a diffraction peak at 2?=23.1° in an X-ray diffraction pattern of the ZSM-5/? core-shell molecular sieve is 0.1-10:1.