University of Rwanda Digital Repository
Not a member yet
17358 research outputs found
Sort by
Type inference for variable definitions and function returns: proposal for C23
No full textWe propose the inclusion of the so-called auto feature for variable definitions and function types into C. This feature allows to infer types from expressions that are used in initializers or return statements. This is part of a series of papers for the improvement of type-generic programming in C that has been introduced in N2734
Introduce the nullptr constant
No full textSince more than a decade C++ has already replaced the problematic definition of NULL which might be either of integer type or void*. By using a new constant nullptr, they achieve a more constrained specification, that allows much better diagnosis of user code. We propose to integrate this concept into C as far as possible without putting much strain on implementations. This is a split-off of N2368 that builds on the approval of the "keywords" patches
Empty Mind: An exploration towards an autonomous digital experience and aesthetics within a virtual live performance End report for DARIAH-EU THEME 2020: Arts Exchanges
No full textRapport d'évaluation de l'unité ASTRO - Agrosystèmes Tropicaux: Campagne d'évaluation 2020-2022 - Vague B: SOUS TUTELLE DES ÉTABLISSEMENTS ET ORGANISMES : Institut national de recherche pour l'agriculture, l'alimentation et l'environnement - INRAE
No full textL’unité « Agrosystèmes Tropicaux » (ASTRO) est une unité de recherche pluridisciplinaire créée le 1er janvier 2010 et sous tutelle de quatre départements de recherche INRAE : Agroécosystèmes (AGROECOSYSTEM), département pilote ; Biologie et Amélioration des Plantes (BAP) ; Économie et Sociologie (ECOSOCIO) ; Santé des Plantes et Environnement (SPE), et d’un département associé : Aliments, produits biosourcés et déchets (TRANSFORM). L’unité est positionnée essentiellement sur l'implantation INRAE du Domaine Duclos en Guadeloupe, sur trois sites principaux répartis entre bureaux et laboratoires, et gère plusieurs implantations techniques. À la date du 1 er juin 2020, l'unité était constituée de 42 agents permanents (deux DR, cinq CR, 36 ITA et non cadre EPIC
Private expenditure on culture in the emirate of Abu Dhabi - Introductory report
No full textThis report summarises: (1) analysis of the results of the culture data derived from the Household and Income Expenditure Survey (HIES) implemented by the Statistics Centre Abu Dhabi (SCAD) in 2014 and 2019; and (2) assessment of the cultural indicators from the Abu Dhabi Consumer Price Index (CPI) data for the period 2014 to 2021. The HIES and CPI data shared by SCAD has been analysed according to the internationally acknowledged approach adopted by the European Statistical System Network on Culture (ESSnet-CULTURE) to document direct expenditure on cultural goods and services and the recent trends in prices for cultural goods and services
Representing motion as a sequence of latent primitives, a flexible approach for human motion modelling
No full textWe propose a new representation of human body motion which encodes a full motion in a sequence of latent motion primitives. Recently, task generic motion priors have been introduced and propose a coherent representation of human motion based on a single latent code, with encouraging results for many tasks. Extending these methods to longer motion with various duration and framerate is all but straightforward as one latent code proves inefficient to encode longer term variability. Our hypothesis is that long motions are better represented as a succession of actions than in a single block. By leveraging a sequence-to-sequence architecture, we propose a model that simultaneously learns a temporal segmentation of motion and a prior on the motion segments. To provide flexibility with temporal resolution and motion duration, our representation is continuous in time and can be queried for any timestamp. We show experimentally that our method leads to a significant improvement over state-of-the-art motion priors on a spatio-temporal completion task on sparse pointclouds
Finding the best Machine Learning algorithm to forecast sedimentation in a wetland ecosystem in India: A comparative analysis
No full textMachine Learning (ML) methods are proposed within the academic literature as alternatives to traditional time series forecasting. Yet, scant evidence is available about their relative performance in terms of accuracy and computational requirements. The purpose of this study is to evaluate the performance of a range of ML and traditional statistical methods across multiple forecasting horizons. A subset of daily modeled sedimentation data in a coastal wetland in India for a 20 year period was used to carry out the study. After analyzing the data we created forecasting models that can forecast for daily, monthly, yearly and seasonal values of the parameter we are interested in. The classical time series models [1] considered for this analysis are Holt-Winters exponential smoothing, Moving Average(MA), AutoRegressive(AR), Autoregressive Moving Average(ARMA), AutoRegressive Integrated Moving Average(ARIMA), Seasonal AutoRegressive Integrated Moving Average(SARIMA), and Seasonal AutoRegressive Integrated Moving Average with Exogenous Variable(SARIMAX) models. RNN and Facebook Prophet models are used for deep learning model [2] analysis. After comparing the post-sample accuracy of popular ML methods thereupon of classical statistical ones, we found that the previous are dominated across both accuracy measures used and for all forecasting horizons examined and vice versa depending on the interval(i.e. daily, monthly or seasonal time interval) for which we're analyzing. Moreover, we observed that their computational requirements are considerably greater than those of classical statistical methods. The report explains how both the statistical and ML approaches work, discusses the results, and proposes some possible ways forward to decide which one to use. The empirical results found in our research stress the necessity for objective and unbiased ways to check the performance of forecasting methods which will be achieved through sizable and open competitions allowing meaningful comparisons and definite conclusions
Innovative ideas for teaching supports: Application to Graph theory
No full textTeaching graph theory with the most adequate tools requires time and ideas. We present how an open community of teachers shares contents and ideas on an innovative platform. The objective is to get the students autonomous in their training with activities that give them immediate feedback on their understanding. Beyond learning, the very large collection of exercises of various levels can also be used to evaluate the student's level. The proposed activities can be algorithm's code in classical programming languages (e.g. Java, Python) that the student can test with predefined tests proposed by the teacher or collections of generated questions