31results about How to "Dig" patented technology

The present invention discloses an extraction method for metagenome, and a method of probe hybridization-bead capture is used to reduce the host DNA content in a host metagenomic DNA sample. According to the method, the Alu repeat sequence of the host DNA sequences and conserved sequence at both ends thereof are used as templates to design one-way or two-way probes, wherein the 5 'end of each probe is modified with biotin, streptavidin coated beads are used to capture host DNA hybridized with probes, in order to achieve aims of weakening host DNA background in a metagenome, and enhancing sequencing data efficiency.

The invention discloses an image hash code training model algorithm and a classification learning method based on binary weights. The model algorithm steps include: selecting a loss function, determining an objective equation, and performing binary encoding on classifiers and training image features; Code for unified learning, update binary code, optimize loss function; derive hash code training model. Also disclosed is a classification learning method using a hash code image training model based on binary weights. The steps include: the image to be searched is obtained through the hash code training model based on binary weights to obtain the hash code, and the hash code and the hash code are obtained. The Hamming distance between the binary codes of the classifiers; find the minimum Hamming distance in the Hamming distance, and obtain the corresponding classifier, which is the category of the image to be searched. The invention can classify images in multiple image categories and high-latitude scenes, improve the performance of algorithms on large-scale data sets, and is accurate, efficient, fast and low in memory consumption.

The present invention relates to a human body posture estimation method based on joint relationship, comprising the following steps: S1: building a joint relationship module, including two sub-modules, a channel-based feature relationship module and an adjacent joint space relationship module; S2: building a human body based on joint relationship Pose estimation model; S3: use the labeled human pose data to train the human pose estimation model based on the joint relationship; S4: use the trained human pose estimation model with the joint relationship module to perform the human pose estimation task based on a single image, and obtain the prediction human body posture. Compared with the prior art, the present invention effectively overcomes the problem that it is difficult to detect the position of limb joints with high degrees of freedom such as wrist joints, ankle joints and hidden invisible joints in the image, and the accuracy of human body posture estimation is high.

First, analyze and preprocess the weather data and traffic crossing data, input them into the threshold recurrent neural network (GRU) to extract the high-order features of traffic flow, and then input the high-order features into the gradient boosting decision tree regression model (GBDT) for short-term traffic flow prediction. The invention can carry out more in-depth excavation and analysis on traffic flow data, and after changing the output layer to GBDT, it can effectively improve the accuracy of short-term traffic flow prediction.

A neural network-based online car-hailing supply and demand forecasting method collects various online car-hailing travel data, uses convolutional neural networks to extract features and reduce dimensions, and then inputs the extracted feature maps into the gated recurrent neural network To predict the supply and demand of online car-hailing, adjust the model during the training process, and finally predict the balance between supply and demand of online car-hailing through the adjusted model. The method of the present invention can carry out more in-depth excavation and analysis on the supply and demand balance data of online car-hailing, so it is more accurate and better in performance when predicting the supply and demand of online car-hailing.

The invention discloses a substation clustering method that takes into account voltage sag monitoring information and associated factors, including establishing voltage sag event monitoring information indicators and associated factor indicators; inputting voltage sag characteristic index data of power grid substations in areas to be analyzed , and carry out normalization processing; using the fuzzy C-means clustering method, clustering the substations of the power grid in the area to be analyzed according to the three voltage sag characteristic indicators respectively, and obtaining the clustering results in three dimensions; selecting key indicators, using fuzzy The C-means clustering method, according to the key indicators, clusters the substations of the power grid in the area to be analyzed to obtain a comprehensive clustering result; mainly based on the comprehensive clustering results of the substations, combined with the clustering results of three dimensions, the voltage sag of each category of substations is analyzed characteristics, and give preliminary suggestions for voltage sag control. The clustering method of the invention has the advantages of comprehensive index setting, good classification effect, rich extracted information and the like.

The invention discloses a transfer learning method based on machine reading to sequence model, comprising the following steps: (1) pre-training a machine reading model, the machine reading model includes a coding layer and a model layer based on a recurrent neural network; (2) ) set up a sequence model, the sequence model includes an encoder and a decoder based on a recurrent neural network; (3) extract the parameters of the encoding layer and the model layer in the trained machine reading model, and migrate to the sequence model to be trained, As part of the initialization parameters when training the sequence model; (4) train the sequence model until the model converges; (5) use the trained sequence model for text sequence prediction tasks. By using the present invention, the information contained in the text can be more deeply excavated, and the quality of the generated text sequence can be improved.

The invention discloses a method and a device for adjusting the granularity of a chart, belonging to the field of data mining. The method includes: acquiring the coordinates of the first point triggered by the user on the chart; acquiring the sliding position of the user operation in real time, and determining the granularity corresponding to each position; acquiring data corresponding to each granularity; keeping the first point at the granularity The coordinates in the adjustment direction remain unchanged, and the data on the chart are updated in real time according to the various granularities and corresponding data. The device includes: a first trigger module, a first acquisition module and a first adjustment module. The invention realizes the real-time adjustment and precise adjustment of the granularity of charts, improves the efficiency and accuracy of granularity adjustment, and simplifies user operations.