Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 2
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Detection of glaucoma using three-stage training with EfficientNet

  • de Zarzà, I.
  • de Curtò, J.
  • Tavares De Araujo Cesariny Calafate, Carlos Miguel|||0000-0001-5729-3041
[EN] This paper sets forth a methodology that is based on three-stage-training of a state-of-the-art network architecture previously trained on Imagenet, and iteratively finetuned in three steps; freezing first all layers, then re-training a specific number of them and finally training all the architecture from scratch, to achieve a system with high accuracy and reliability. To determine the performance of our technique a dataset consisting of 17.070 color cropped samples of fundus images, and that includes two classes, normal and abnormal, is used. Extensive evaluations using baselines models (VGG16, InceptionV3 and Resnet50) are carried out, in addition to thorough experimentation with the proposed pipeline using variants of EfficientNet and EfficientNetV2. The training procedure is described accurately, putting emphasis on the number of parameters trained, the confusion matrices (with analysis of false positives and false negatives), accuracy, and F1-score obtained at each stage of the proposed methodology. The results achieved show that the intelligent system presented for the task at hand is reliable, presents high precision, its predictions are consistent and the number of parameters needed to train are low compared to other alternatives., This work is supported by the HK Innovation and Technology Commission (InnoHK Project CIMDA), the HK Research Grants Council (Project CityU 11204821) and City University of Hong Kong (Project 9610034). We acknowledge the support of Universitat Politècnica de València; R&D project PID2021-122580NB-I00, funded by MCIN/AEI/ 10.13039/501100011033 and ERDF.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

On the applicability of the Hadamard as an input modulator for problems of classification

  • de Curtò, J.
  • de Zarzà, I.
  • Yan, Hong
  • Tavares De Araujo Cesariny Calafate, Carlos Miguel|||0000-0001-5729-3041
[EN] The library explores the applicability of the Hadamard as an input modulator for problems of classification. It introduces a framework in C++ to use kernel approximates in the mini-batch setting with Stochastic Gradient Descent. The algorithm requires to compute the product of matrices Walsh Hadamard. A free-standing cache friendly Fast Walsh Hadamard that achieves compelling speed is provided, as well as a lightweight efficient CPU implementation of the method for research and practical purposes alike., This work is supported by HK Innovation and Technology Commission, Hong Kong (InnoHK Project CIMDA) and HK Research Grants Council, Hong Kong (Project CityU 11204821).