带你从零实现一个高性能的深度学习推理库,支持大模型 llama2 、Unet、Yolov5、Resnet等模型的推理。Implement a high-performance deep learning inference library step by step
-
Updated
Jun 6, 2024 - C++
带你从零实现一个高性能的深度学习推理库,支持大模型 llama2 、Unet、Yolov5、Resnet等模型的推理。Implement a high-performance deep learning inference library step by step
Data Science Project: Comparing 3 Deep Learning Methods (CNN, LSTM, and Transfer Learning).
Represent trained machine learning models as Pyomo optimization formulations
American Sign Language (ASL) Detection using CNN
QReLU and m-QReLU: Two novel quantum activation functions for Deep Learning in TensorFlow, Keras, and PyTorch
Neural Network from Scratch using Numpy for MNIST Dataset
Neural Network from Scratch using Numpy for MNIST Dataset
Implementations of neural networks in python for the classification of MNIST datasets.
"The 'Activation Functions' project repository contains implementations of various activation functions commonly used in neural networks. "
identify digits from a dataset of tens of thousands of handwritten images.
Feedforward Neural Network from scratch - backpropagation, gradient descent, activation functions
TorchAct, collection of activation function for PyTorch. https://pypi.org/project/torchact/
Softmax-as-intermediate-layer-CNN
Building Convolutional Neural Networks From Scratch using NumPy
A from-scratch implementation of a multi-layer, densely connected neural network that can classify handwritten images of numbers. Trained on the MNIST dataset with no external AI library dependencies!
A web app where user can draw Bengali digit and the AI model can detect handwritten digit and predict the digit.
Laboratories - for 'Image and Video Analysis and Technology' M.Sc. Course ECE @ntua
Building a HTTP-accessed convolutional neural network model using TensorFlow NN (tf.nn), CIFAR10 dataset, Python and Flask.
Add a description, image, and links to the relu topic page so that developers can more easily learn about it.
To associate your repository with the relu topic, visit your repo's landing page and select "manage topics."