Full Text Available
Access Full Text at Repository
Search Results - training errors
-
End-to-end automatic speech recognition of code-switched speech
Published 2023Get full text
ThesisFull Text AvailableAccess Full Text at Repository -
Utilisation of performance analysis amongst Western Province club rugby coaches
Published 2018Get full text
ThesisFull Text AvailableAccess Full Text at Repository -
Classifying yield spread movements in sparse data through triplots
Published 2020Get full text
ThesisFull Text AvailableAccess Full Text at Repository -
Insurance recommendation engine using a combined collaborative filtering and neural network approach
Published 2021“…The system thus achieved good training performance and given further modifications, could be used in a production environment.…”
Get full text
ThesisFull Text AvailableAccess Full Text at Repository -
Analysis to indicate the impact Hindsight Bias have on the outcome when forecasting of stock in the South African equity market
Published 2023Get full text
ThesisFull Text AvailableAccess Full Text at Repository -
A comparison framework for deep learning RFI detection algorithms in radio astronomy
Published 2024Get full text
ThesisFull Text AvailableAccess Full Text at Repository -
Intelligent tool condition monitoring in high-speed turning of Titanium Ti-6Al-4V alloy
Published 2009ArticleFull Text AvailableAccess Repository -
Cape Town road traffic accident analysis: Utilising supervised learning techniques and discussing their effectiveness
Published 2023“…The RTA data was split into training, validation and test sets keeping the proportions of the injury-severity category consistent. …”
Get full text
ThesisFull Text AvailableAccess Full Text at Repository -
Monitoring and diagnosis of process systems using kernel-based learning methods
Published 2010Get full text
ThesisFull Text AvailableAccess Full Text at Repository -
Driving factors for national competitiveness in Africa as measured by GDP per capita
Published 2015Get full text
ThesisFull Text AvailableAccess Full Text at Repository -
The prevalence of substance use in anaesthesia practitioners in South Africa
Published 2021“…Years of experience and level of training were explored as possible risk factors for hazardous or harmful use. …”
Get full text
ThesisFull Text AvailableAccess Full Text at Repository -
Short-term sea level forecasting using machine learning techniques: A case study for South Africa
Published 2023“…Model performance was measured using the correlation coefficient (R), the root mean square error (RMSE), the bias and the Willmott indices of correlation. …”
Get full text
ThesisFull Text AvailableAccess Full Text at Repository -
Classification of Fallow and Perennial Fields in High-Resolution Multispectral Aerial Images
Published 2021“…In the first approach, the texture classification problem is reformulated as a texture segmentation problem in which each pixel in the image is individually labelled by training a classifier on the texture feature space. …”
Get full text
ThesisFull Text AvailableAccess Full Text at Repository -
Assessing the Principal Agent Problem in Mobile Money Services: Lessons from M – PESA in Lesotho
Published 2019“…Drivers of moral hazard are demonstrated by the influences and demands of agents’ core businesses and challenges in agent monitoring and training. The existence of the principal – agent problem has limited or no implications on access and availability of services. …”
Get full text
ThesisFull Text AvailableAccess Full Text at Repository -
Adapting Large-Scale Speaker-Independent Automatic Speech Recognition to Dysarthric Speech
Published 2023“…Re-initialization was achieved by randomly initializing the weights of Deep Speech and training from scratch. A separate encoder-decoder recurrent neural network consisting of far fewer parameters was also trained from scratch. …”
Get full text
ThesisFull Text AvailableAccess Full Text at Repository -
Investigation of brain ageing in HIV-positive individuals using convolutional neural networks
Published 2025“…To this end, we achieve a mean absolute error (MAE) of 2.57±1.94 years. Next, we fine-tune the pre-trained model on a smaller dataset, with a wider age range, aligned with that of our testing dataset from ENIGMA-HIV. …”
Get full text
ThesisFull Text AvailableAccess Full Text at Repository