Deliver uncompromised performance for diverse workloads across multiple architectures.
URL: https://www.youtube.com/watch?v=3LboGde_1WM
Derma diseases are common conditions with a variety of causes. Some of them can cause death like Melanoma, but with early diagnosis, we can save many lives. Dermato.AI is an AI tool that can identify, detect, and segment of 3 types of similar-looking skin diseases.
Posted:
URL: https://github.com/Nitin-Mane/Freshwater-Quality-Prediction-using-oneAPI
This project aims to predict freshwater quality using machine learning techniques, specifically Adaptive Particle Swarm Optimization (APSO) and Convolutional Neural Network (CNN). The APSO algorithm is used for feature selection, while the CNN is used for classification. The model achieved an accura
Posted:
URL: https://github.com/Nitin-Mane/Automated-Weed-Detection
The proposed method uses a Residual Network (ResNet) architecture with Auto Mixed Precision (AMP) to classify images of crops and weeds. The use of AMP allows for dynamic adjustment of the precision of computations during training and inference, which can improve the model's performance.
Posted:
URL: https://github.com/Santhoshkumard11/fresh-water-classifier
Deploying a binary classification model on Azure Function to do on-demand predictions at scale powered by Intel oneAPi AI Analytics Toolkit. We have multiple versions of the model with varying f1 scores - the latest model has 94 as it's f1 score. Integrated GitHub actions with GitHub repo to make se
Posted:
URL: https://gitee.com/doraemon22333/parallel-operation
根据数学知识用三重循环实现矩阵乘法的基础上，改变循环的次序增大cache命中率，以及实现分块矩阵乘法，在这些基础上采用并行算法实现矩阵乘法，矩阵分块乘法
Posted:
URL: https://github.com/Csthyx/Conjugate_Gradient
共轭梯度算法也叫共轭斜量法。从理论上讲它是属于直接法，但在实际计算过程中，由于不可避免地会出现误差，所以常作为迭代法使用。该方法的最大特点是，当方程组的阶数很高时，往往只要经过比阶数小得多的迭代次数，就能得到满足精度要求的近似解。 使用openMP共轭梯度算法的串行以及并行版本，并统计在不同数据规模以及不同的并行线程数下，共轭梯度算法完成运算所耗费的时间并比较。
Posted:
URL: https://gitee.com/doraemon22333/parallel-operation-experiment.git
采用并行算法实现矩阵乘法，并将运行时间与直接矩阵相乘结果比较
Posted:
URL: https://github.com/IowaBattleship/TSP-Problem
我们在此次项目名为：并行化解决旅行商问题的各种算法，包括：遗传算法、粒子群算法、蚁群算法。旅行商问题是给定一系列城市和每对城市之间的距离，求解访问每一座城市一次并回到起始城市的最短回路 。此类问题属于NP难问题，无法在线性时间内解决，所以我们决定采用智能优化搜索算法，即各种仿生群体算法（遗传算法、粒子群算法、蚁群算法）来解决该问题，此类算法具有隐含并行性，对求解目标进行合理表示之后便可获得较优解，并且比较适合采用并行算法进行优化。
Posted:
URL: https://github.com/2022xbing/oneAPI-CrossEntropy.git
课堂作业
Posted:
URL: https://gitee.com/mars0417/matrix_block_multiplication
矩阵乘法的一般形式是C=A*B，使用串行计算的矩阵乘法需要三层循环，操作次数为2*M*N*K。 为了减少矩阵乘法的计算成本，本实验决定对计算结果的矩阵C进行分块，分成大小相同的若干块，这些工作块的计算分别要用到矩阵A和矩阵B的部分行和列。由gpu并行进行工作块的计算，从而达到矩阵分块乘法的并行实现，提高矩阵乘法的效率。 在将矩阵分块计算的时候，读取内存的次数并没有改变，只是利用gpu并行计算将计算时间缩短。因此设置缓存块，在每一块的运算中，将工作块按照缓存块大小分割，单次循环缓存足以计算出缓存块大小的矩阵A和矩阵B的数据，减少迭代次数，从而减少访问内存的次数，实现矩阵分块乘法的缓存优化。
Posted:
URL: https://github.com/lupinsan/matrix-mul.git
# matrix-mul one m*n matrix mul one n*r matrix
Posted:
URL: https://github.com/stan-maze/Parallel_Prime_Sieve.git
Parallel sieve method based on Esl's sieve
Posted:
URL: https://github.com/DungeonMasterVANsama/oneAPI_based_Parallel_AStar/tree/main
基于oneAPI的实现并行查找和结构优化的A*算法
Posted:
URL: https://build.opensuse.org/project/show/home:cabelo:intel
Individual initiative in Brazil of the world's first Linux image with native kernel 6.2-rc5 or higher with Intel ARC stable open source driver. All GPU detection procedure takes place automatically during installation.
Posted:
URL: https://github.com/QuietHuihui/oneAPI_TSP
高性能计算与云计算实验4以及大作业 使用oneAPI将遗传算法进行部分并行以及加速改良，提高其解决旅行商问题的效率。
Posted:
URL: https://github.com/Goleys/oneAPI_NeuralNetwork_Research.git
We will create a simple Neural Network using CUDA and DPC++. Then, we will use oneAPI to test our code using an Intel base workstation and compare it with the performance of an NVIDIA GPU.
Posted:
URL: https://github.com/Prajwal111299/Electric-Vehicle-Charging-Patterns
Using EVs' charging data, I explored when drivers are likely to plug in their cars, and how much additional electricity demand will be created when the number of EVs increases.
Posted:
URL: https://github.com/derolol/oneapi_dct
This project uses oneAPI to realize the parallelization of discrete cosine transform, and completes the performance test of the algorithm on the DevCloud platform
Posted:
URL: https://gitee.com/hulakk/parallel-computing-public
从线性代数角度求解线性方程组有一个经典方法叫做高斯消元法，该方法的基本思路是将方程组的增广矩阵通过初等行变换转换为阶梯矩阵，进而通过迭代求解出方程的解。但高斯消元法有两个严重的缺陷：主元为0时无法求解和主元太小时求解误差过大。列主元消元法是对高斯消元法的一种优化方法，先通过列交换将主元替换为绝对值较大的元素，规避了高斯消元法的以上两个缺陷。本项目着眼于将高斯消元法和优化后的列主元消元法分别用串行、并行两种方式实现，并在Intel提供的DevMesh平台上部署代码，比较运行效率，体会程序并行化的方法及意义。
Posted:
URL: https://github.com/Coolog/Logistic-Regression.git
高性能报告
Posted:
