Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 20 of 23 for “"ONNX"”.

  1. An optimizing compiler for ONNX models on heterogeneous systems

    … models, which is based on an open format - ONNX of deep learning models, running on heterogeneous systems. It consists of a front-end and a back-end to address those above issues. The goal of this neural network compiler is also to map high-level neural network models to low-level executable …

    uiuc Repository record for An optimizing compiler for ONNX models on heterogeneous systems (opens in a new tab)

  2. GPU ACCELERATED BIAS MITIGATION IN FACE DETECTION USING VARIATIONAL AUTO ENCODERS AND ONNX RUNTIME

    … deployment, the trained model was exported via ONNX Runtime and optimized for GPU inference. On GPU, the ONNX-exported model attains nearly a 19× throughput speed-up over CPU execution, enabling real-time face detection. In summary, this research demonstrates a fast and fair face detection …

    mcmaster Repository record for GPU ACCELERATED BIAS MITIGATION IN FACE DETECTION USING VARIATIONAL AUTO ENCODERS AND ONNX RUNTIME (opens in a new tab)

  3. Převod modelů mezi nástroji strojového učení pro mobilní platformy

    … je formát Open Neural Network Exchange (ONNX). Vývoj softwarovej podpory pre hardwarové akcelerátory na vstavaných systémoch je drahý, a ONNX je len výnimočne podporovaný. Potrebné ovládače sú typicky implementované iba pre formát TensorFlow Lite (TFLite). Aktuálne možnosti pre konverziu …

    brno-tech Repository record for Převod modelů mezi nástroji strojového učení pro mobilní platformy (opens in a new tab)

  4. Rozpoznávač řeči běžící čistě ve webovém prohlížeči.

    … a modelu NeMo Parakeet pomocí technologie ONNX Runtime Web. Závěr práce je věnován testování a porovnání modelů z hlediska rychlosti, chybovosti a paměťové náročnosti. Vyhodnocen je i vliv kvantizace nebo volba výpočetního backendu pro jednotlivé části inference.

    brno-tech Repository record for Rozpoznávač řeči běžící čistě ve webovém prohlížeči. (opens in a new tab)

  5. Introducing the ModularML Framework - A transparent and modular machine learning framework made as a tool for research and education

    … performance compromises. The framework can parse ONNX models into dynamic C++ objects, modify the implementation of core computational functions (like GEMM), and use the models to perform inference. The framework reproduces results achieved in peer frameworks like PyTorch and TensorFlow. The …

    chalmers Repository record for Introducing the ModularML Framework - A transparent and modular machine learning framework made as a tool for research and education (opens in a new tab)

  6. Aplikace pro odstranění objektů z obrázků s využitím metod hlubokého učení

    … straně webové aplikace s využitím knihovny ONNX Runtime Web, pro efektivní využití zdrojů klientova zařízení jsou použity technologie WebGPU a WebAssembly. Výsledná aplikace umožňuje díky využitým neuronovým sítím vybrat a odstranit objekt pouze v několika málo kliknutích. Dle zpětné vazby …

    brno-tech Repository record for Aplikace pro odstranění objektů z obrázků s využitím metod hlubokého učení (opens in a new tab)

  7. MLPerf PowerPack: Defining MLPerf Inference Benchmark's Power Profile

    … functions that comprise MLPerf Inference are ONNX Runtime model prediction, system check, post-processing ONNX model output, image retrieval, backend initialization, and loading annotations using the COCO API. Defining the power consumption characteristics of each function with component-level …

    vt Repository record for MLPerf PowerPack: Defining MLPerf Inference Benchmark's Power Profile (opens in a new tab)

  8. Prieigos kontrolės plėtojimas DevOps įrankiuose: laiku pagrįstų taisyklių įgyvendinimas /

    … integration. The trained models were exported to ONNX format and integrated into the policy engine for automatic access revocation upon anomaly detection.

    vilnius Repository record for Prieigos kontrolės plėtojimas DevOps įrankiuose: laiku pagrįstų taisyklių įgyvendinimas / (opens in a new tab)

  9. Efficient Processing of Convolutional Neural Networks on the Edge: A Hybrid Approach Using Hardware Acceleration and Dual-Teacher Compression

    … including Vitis-AI, OpenVINO, TensorRT, and ONNX, without architectural adjustments. This provides a scalable solution for deploying high-accuracy CNNs on low-power edge devices across various applications, such as autonomous vehicles, surveillance systems, robotics, healthcare, and smart …

    ucf Repository record for Efficient Processing of Convolutional Neural Networks on the Edge: A Hybrid Approach Using Hardware Acceleration and Dual-Teacher Compression (opens in a new tab)

  10. Understanding Open-Source Contributor Profiles in Popular Machine Learning Libraries

    … scikit-learn, Keras, MXNet, Theano/Aesara, ONNX, and deeplearning4j), we identify four contributor profiles: Core-Afterhour, Core-Workhour, Peripheral-Afterhour, and Peripheral-Workhour. We find that: 1) project experience, authored files, collaborations, pull requests comments received and …

    queens Repository record for Understanding Open-Source Contributor Profiles in Popular Machine Learning Libraries (opens in a new tab)

  11. Skinclusive AI: Towards Equitable Skin Cancer Detection for Deployment on Edge Devices

    … attributes. The final model is converted to ONNX format and quantized for deployment on both general-purpose and resource-constrained computer hardware, including an Intel Xeon (via Google Colab) and a Raspberry Pi 3 (RPi3). Our final quantized system achieved an average inference of 1.91 …

    calpoly Repository record for Skinclusive AI: Towards Equitable Skin Cancer Detection for Deployment on Edge Devices (opens in a new tab)

  12. Methods for Optimizing BERT Model on Edge Devices - Accelerating Biomedical NLP with Pruned and Quantized BERT Models

    … stage, dynamic 8-bit integers quantization using ONNX Runtime was applied to standard BERT, BioBERT, and DistilBERT. The third stage involved combining both pruning and quantization, further optimizing the pre-trained standard BERT and BioBERT transformers. Experiments were run on annotated …

    chalmers Repository record for Methods for Optimizing BERT Model on Edge Devices - Accelerating Biomedical NLP with Pruned and Quantized BERT Models (opens in a new tab)

  13. Real-Time Anomaly Detection in OT Networks Using GRU-Based Autoencoders

    … the Seq100 model was quantized, exported to ONNX, and deployed on an AMD Neural Processing Unit (NPU) and an NVIDIA V100 GPU. With a mean inference time of just 5.6 milliseconds on the NPU, the model demonstrated strong real-time feasibility in resource-constrained environments. This thesis …

    arkansas Repository record for Real-Time Anomaly Detection in OT Networks Using GRU-Based Autoencoders (opens in a new tab)

  14. Meta-Solving via Machine Learning for Automated Reasoning

    … engine over the Open Neural Network Exchange (ONNX) -- a standardized model format for deep learning and classical ML, and VNN-LIB -- a specification standard over the input-output behavior of ML systems. \pierce supports the entirety of the VNN-LIB and most of ONNX v18.

    waterloo Repository record for Meta-Solving via Machine Learning for Automated Reasoning (opens in a new tab)

  15. An Edge Computing Device Optimized and Transfer Learning Enhanced Deep Learning Model for Detecting Wildfire Flame and Smoke

    … frameworks, Open Neural Network Exchange (ONNX) and Open Visual Inference and Neural Network Optimization (OpenVINO), to accelerate CPU-based inference. Compared to the original YOLOv5n, the modified model converted to OpenVINO demonstrates a 423% increase in inference speed - up to 31.9 …

    unlv Repository record for An Edge Computing Device Optimized and Transfer Learning Enhanced Deep Learning Model for Detecting Wildfire Flame and Smoke (opens in a new tab) Download PDF for An Edge Computing Device Optimized and Transfer Learning Enhanced Deep Learning Model for Detecting Wildfire Flame and Smoke (opens in a new tab)

  16. Workflow recommendations using deep learning and machine learning tools for exploratory and predictive data analysis in Galaxy

    … in an open neural network exchange format (ONNX), and other supporting datasets become available in Galaxy.<br><br>The usage of the JupyterLab tool in Galaxy is demonstrated by two use cases - prediction of infected regions in COVID-19 CT scans and 3D structure of proteins (publication P3). …

    freiburg-diss Repository record for Workflow recommendations using deep learning and machine learning tools for exploratory and predictive data analysis in Galaxy (opens in a new tab)

  17. Learning-Based Control and Tactile Perception for Robotic Systems

    … PPO-based whole-body tracking in Isaac Lab, and ONNX-based inference inside a 50 Hz policy loop supported by a 500 Hz ROS 2 servo loop. A central design choice is to separate slow semantic skill grounding from fast balance-critical motor execution: a cloud-hosted multimodal large language model …

    washington Repository record for Learning-Based Control and Tactile Perception for Robotic Systems (opens in a new tab)

  18. Pain Recognition Performance on a Single Board Computer

    Emotion recognition is a quickly growing field of study due to the increased interest in building systems which can classify and respond to emotions. Recent medical crises, such as the opioid overdose epidemic in the United States and the global COVID-19 pandemic has emphasized the importance of …

    usf Repository record for Pain Recognition Performance on a Single Board Computer (opens in a new tab)

  19. Sémantická segmentace obrazu z kamery mobilního robotu

    Diplomová práca bola zameraná na riešenie problému sémantickej segmentácie pre kameru mobilného robota, ktorý je postavený na menej výkonnom hardvéry. Výberom a implementáciou vhodnej konvolučnej siete bola docielená predikcia v reálnom čase aj na staršej grafickej karte akou je Nvidia GTX 1050Ti

    brno-tech Repository record for Sémantická segmentace obrazu z kamery mobilního robotu (opens in a new tab)

  20. Detekce vozidla v obraze

    Tato bakalářská práce je zaměřena na detekci vozidel v obraze. V práci je rozebrána metoda detekce vozidel pomocí konvolučních neuronových sítí, jejich struktury a modely. Všechny skripty byly realizovány v programovacím jazyce Python s rozhraním Tensorflow Object Detection API. První část …

    brno-tech Repository record for Detekce vozidla v obraze (opens in a new tab)

Page 1 of 2