Abstract: A computer device extracts local features of sample images based on a first part of a convolutional neural network (CNN) model. The sample images comprise a plurality of images taken at the same place. The device; aggregates the local features into feature vectors having a first dimensionality based on a second part of the CNN model. The device obtains compressed representation vectors of the feature vectors based on a third part of the CNN model. The compressed representation vectors have a second dimensionality less than the first dimensionality. The device trains the CNN model, and obtains a trained CNN mode satisfying a preset condition in accordance with the training.

Abstract: A pyridine-pyrimidine derivative, a preparation method therefor and a pharmaceutical use thereof. In particular, the present disclosure relates to pyridine-pyrimidine derivative as shown in general formula (I), a preparation method therefor, a pharmaceutical composition containing the derivative, and a use thereof as a therapeutic agent, especially a use thereof as an SOS1 inhibitor and a use thereof in preparation of drugs for treating diseases, conditions or disorders improved by inhibiting SOS1.

Abstract: Disclosed is a pyrazolo-heteroaryl derivative, a preparation method therefor, and medical use thereof. In particular, the present invention relates to a pyrazolo-heteroaryl derivative as shown in the general formula (I), a preparation method therefor, a pharmaceutical composition containing the derivative, and a use thereof as a therapeutic agent, particularly as ATR kinase inhibitor and in the preparation of drugs for the treatment and/or prevention of hyperproliferative diseases. The definition of each group in the general formula (I) is identical as in the specification.

Abstract: The present application relates to a substituted fused bicyclic derivative, a preparation method therefor, and an application thereof in medicines. Specifically, the present application relates to a novel substituted fused bicyclic derivative as represented by formula (I), a preparation method therefor, a pharmaceutical composition containing the derivative, an application thereof as a therapeutic agent, in particular, as an ERK inhibitor, and an application thereof in preparation of medicines for treatment and/or prevention of cancer, inflammation, or other proliferative diseases, wherein the definitions of the substituents in formula (I) are the same as those in the description.

Abstract: A computer device extracts local features of sample images based on a first part of a convolutional neural network (CNN) model. The sample images comprise a plurality of images taken at the same place. The device; aggregates the local features into feature vectors having a first dimensionality based on a second part of the CNN model. The device obtains compressed representation vectors of the feature vectors based on a third part of the CNN model. The compressed representation vectors have a second dimensionality less than the first dimensionality. The device trains the CNN model, and obtains a trained CNN mode satisfying a preset condition in accordance with the training.