20484001 results
Contributors: Jorge Segarra-Tamarit, Emilio Perez, Hector Beltran, Javier Perez
Date: 2020-02-13
... This dataset contains the code, data and results obtained by forecasting the GHI using a deep neural network. There are four main folders in the project: Data This folder contains all the data used from the two studied locations: UJI (latitude=39.99º, longitude=-0.06º) and La Celle (latitude=40.4º, longitude=6.0º). Sorted by year, month and day, each location has three kinds of data: • The files named as just a number are 151x151 irradiance estimates matrices centered in the same location obtained from http://msgcpp.knmi.nl.The spatial resolution is 0.03º for both latitude and longitude. • The files named Real_ are the irradiance measurements at the location • The files named CopernicusClear_ are the clear sky estimates from the CAMS McClear model Each file contains the 96 15-minute samples for the same day in Matlab format and UTC time. Code All the python scripts used to train the neural networks and perform the forecasts. The main files are: • tf1.yml: List of the modules and versions used. A clean Anaconda environment created from this file can run all the code in the project. • learnRadiation.py: The script to train a new model. Changing the variables “paper_model_name” and “location”. The first variable selects the kind of model to fit and the second one the training location. • predictOnly.py: Loads a trained model and performs the forecast. Notice that the model and location must match the ones used to train the model stored in the “training_path” folder Models This folder contains all the trained models and their forecasting results. There is also a training folder to contain the last trained model. Logdir This folder stores Tensorboard files during training A new model can be trained using “learnRadiation.py”. This script has three parameters • location: Selects the location where the model will be trained (UJI or LaCelle) • paper_model_name: This sets the inputs to match the ones used in the models from the article. • training_path: The folder to save the trained model “predictOnly.py” script allows performing the forecasts. It is important to set the same parameters as in the “learnRadiation.py” script. This program will generate the predictions and save them in the model folder. It also plots some days, which can be modified at the bottom of the script. For instance for UJI and model TOA & all real we would run: "python learnRadiation.py TOAallreal UJI training" This will train the neural network and save the results in the folder models/training. After this, we would generate the results and plot some days using: “python predictOnly.py TOAallreal UJI training” This will save the forecasts and real values in the training folder and show figures with 1 to 6 hour forecasts The models used for the article can also be evaluated by using predictOnly.py and targeting their folders. For instance, to evaluate the TOA & all real model used in the article, this command must be used: “python predictOnly.py TOAallreal UJI RtoaAllReal”
Data types:
Contributors: Kenli Li, Ahmad Salah, Khalid Hosny, Qi Tian, Mohamed Darwish
Date: 2020-02-13
... undefined
Data types:
Contributors: Laura Ortiz-Herrero, IRENE cardaba, Sofía Setien, Luis Bartolomé, marialuz alonso, Miren Itxaso Maguregui
Date: 2020-02-13
... Raw data acquired by FTIR-ATR for two paint brands and applied to the chemometric analysis. Each column gives the absorbance measured at each wavenumber recorded for each paint sample. The accelerated ageing time of each paint sample is expressed as a logarithmic function.
Data types:
Contributors: Renato Bittar
Date: 2020-02-13
... To determine whether the cephalometric measurements studied could correlate with the polysomnographic parameters in order to provide a positive or negative predictive value for association with OSAS.
Data types:
Contributors: Olga Jakšić
Date: 2020-02-13
... For known kinetic parameters, trained on the numerical solutions to coupled Riccati equations, ANN gives the pair of steady state values for the components in a binary mixture.
Data types:
Contributors: Steven Luksic
Date: 2020-02-13
... Data used for the publication, being imageJ code for segmentation of bubbles from in-situ CT scans of glass-melting, spread sheets of bubble parameters output from Avizo particle measurement, and analysis of the same.
Data types:
Contributors: Mattia Barbero
Date: 2020-02-13
... This data set contains data from the Montgat library (Spain) in the period October 2017 to October 2018. In particular it contains the hourly total energy consumption, the hourly consumption from the HVAC system, the hourly external temperature (From 24 October 2017), the 5-minutes internal temperature and the set-point temperature.
Data types:
Contributors: Daniel Díaz, Angelo Castruccio, Felipe Zuñiga
Date: 2020-02-13
... Sensitivity tests for 3-D modeling presented in the manuscript
Data types:
Contributors: Agus Arip Munawar, Yuswar Yunus, Devianti Devianti
Date: 2020-02-13
... Dataset of near infrared spectroscopy (NIRS) were presented and can be used to predict agricultural soil fertility properties. Near infrared spectra data were acquired in wavelength range from 1000 to 2500 nm for a total of 40 bulk soil samples amounted of 10 g per each bulk. Soil fertility properties, presented as soil nitrogen (N), phosphorus (P). potassium (K), soil pH, magnesium (Mg) and calcium (Ca), were measured by means of wet chemical analysis.
Data types:
Contributors: Vinh Truong Hoang
Date: 2020-02-12
... The demands and application of face recognition are growing quickly in recent years due to the development of advances technology. One of the most challenging tasks of face recognition is to be recognized the facial image with limited training samples (a few or single image per person). This real-world problem can be seen in access control, physical security and international border control. In order to provide a new benchmark facial dataset for this issue, the FACEID-550 dataset is constructed (the full dataset is available from: http://dx.doi.org/10.17632/zs2vnf9692.2). Itcontains total 11,550 color images (256 × 256 pixels) of 550 subjects. Each person has one ID photo and twenty daily (or selfie) photos.
Data types: