Mechanical Parts data cost data and shape cluster

Dataset Ⅰ：To obtain the prices of parts from the manufacturing characteristics and other manufacturing processes, feature quantity expression is innovatively applied. By identifying manufacturing features and calculating the feature quantities, the feature quantities are described in the form of assignments as data. To obtain the prices of parts intelligently, the most widely used and mature deep-learning method is adopted to realize the accurate quotation of parts.

Dataset Ⅱ：For 3D mechanical CAD model shape clustering, this study established an inertial feature descriptor, extracted the normalized PI and MI features of the model, and constructed a DEC network with the inertial features as input. DEC was initialized using a stacked autoencoder, nonlinear mapping of feature data to the feature space was achieved using an encoder, and complete clustering analysis in the feature space.