关于PPgEEC深度学习基础的研究生课程知识库【国外】.zip资源-CSDN文库

共52个文件

ipynb：20个

tex：11个

pdf：8个

版权申诉

159 浏览量 2024-03-26 13:49:50 上传评论收藏 53.5MB ZIP 举报

: "PPgEEC深度学习基础的研究生课程知识库" 【正文】: 深度学习作为人工智能领域的重要分支，已经在计算机视觉、自然语言处理、语音识别等多个领域取得了显著成果。PPgEEC（可能代表某种课程代码或项目名称）的深度学习基础研究生课程旨在系统地教授这一前沿技术的核心概念和应用。本知识库针对国内外研究生，提供了深入学习深度学习的宝贵资源，帮助学生理解和掌握深度学习的基本原理和实践技巧。 1. **深度学习概述**：深度学习是机器学习的一个子领域，它模仿人脑的工作方式，通过多层非线性变换对数据进行抽象和建模。其核心是神经网络，通过大量训练数据调整权重参数，实现模式识别和预测。 2. **神经网络基础**： - 神经元模型：理解神经网络的基础单元，包括输入、权重、偏置和激活函数。 - 卷积神经网络 (CNN)：在图像处理中的应用，如特征提取和图像分类。 - 循环神经网络 (RNN)：适用于序列数据，如自然语言处理。 - 长短期记忆网络 (LSTM)：解决RNN中的梯度消失问题，优化序列建模。 3. **深度学习框架**：学习如何使用TensorFlow、PyTorch、Keras等主流深度学习框架，了解它们的优缺点以及应用场景。 4. **损失函数与优化器**： - 损失函数：如均方误差、交叉熵等，用于评估模型预测的准确程度。 - 优化器：如梯度下降、随机梯度下降、Adam等，用于更新模型参数以最小化损失。 5. **模型训练与超参数调优**： - 数据预处理：增强数据集，提高模型泛化能力。 - 训练循环：理解批量训练和epoch的概念。 - 超参数调优：网格搜索、随机搜索等方法，找到最佳模型配置。 6. **模型评估与验证**： - 混淆矩阵：理解真正例、假正例、真负例和假负例。 - AUC-ROC曲线：评估二分类模型性能。 - 交叉验证：提高模型稳健性，减少过拟合。 7. **深度学习应用**： - 自动驾驶：利用深度学习进行目标检测和路径规划。 - 医疗诊断：使用深度学习辅助医生识别疾病。 - 推荐系统：通过深度学习构建个性化推荐。 8. **最新研究进展**：深入了解最新的深度学习研究，如生成对抗网络 (GANs)、变分自编码器 (VAEs) 和Transformer模型，以及在元学习、强化学习和无监督学习等方面的应用。文件“eec2003-main”可能是课程的主要资料，可能包含了课件、讲义、练习题和项目指南等内容，为学生提供了一个全面学习和实践深度学习的平台。通过这个课程，学生不仅能够掌握深度学习的基本理论，还能通过实际操作提升解决实际问题的能力，为未来在AI领域的研究和开发打下坚实基础。

资源推荐

资源详情

资源评论

收起资源包目录

关于PPgEEC深度学习基础的研究生课程知识库【国外】.zip （52个子文件）

eec2003-main

Lesson #03

Lesson #03 Task #01 Regularization.ipynb 24MB

Lesson #03 Better Generalizaton vs Better Learning.pdf 2.95MB

Lesson #03 Task #02 Data Scaling and Weight Initialization.ipynb 29KB

Lesson #03 Task #03 Better Deep Learning.ipynb 997KB

Lesson #09

Lesson #09.pdf 1.34MB

Lesson_09.ipynb 107KB

Lesson_09_Cat_and_Dogs.ipynb 48KB

Lesson #02

Figures

Figures.ipynb 176KB

Perceptron

fig_2_perceptron_bias.tex 1KB

fig_6_one_hidden_B.tex 2KB

fig_9_dimension.tex 2KB

fig_4_perceptron_z.tex 1KB

fig_8_onehidden.tex 1KB

fig_1_perceptron_N.tex 1KB

latexmkrc 62B

fig_3_perceptron_example.tex 996B

fig_5_perceptron_z2.tex 2KB

fig_6_one_hidden.tex 2KB

fig_7_two_hidden.tex 2KB

fig10_dimension_1output.tex 2KB

Notebooks

Lesson 02 Task 04 - Better Learning part I.ipynb 1.19MB

Lesson 02 Task 03 - Optimization Methods.ipynb 301KB

Lesson 02 Task 02 - Introduction to TF.ipynb 58KB

Lesson 02 Task 01 - TensorFlow 2.x + Keras Crash Course.ipynb 135KB

Lesson #02 Introduction to Deep Learning and TensorFlow.pdf 6.48MB

Datasets

train_catvnoncat.h5 2.45MB

test_catvnoncat.h5 602KB

data.mat 6KB

Lesson #08

Lesson_08.ipynb 65KB

Lesson #01

Lesson #01 Course Outline.pdf 8.64MB

Lesson #01 Task 01 - Functions, Context Managers and Decorators.ipynb 155KB

Datasets

example_log.txt 2.05MB

nba_players_2013.csv 76KB

Lesson #01 Task 02 - Object-Oriented, Fiunctional Programming.ipynb 128KB

.gitignore 10B

Lesson #06

Lesson #06 CNN Architectures.pdf 4.54MB

Lesson_06.ipynb 113KB

Lesson #05

[Step 1] Foundations of DL for CV and Image Data Preparation.ipynb 2.29MB

Lesson #05 Fundamentals of CNN.pdf 3.1MB

[Step 2] Convolutional Neural Networks and Case Studies.ipynb 426KB

Lesson #04

Lesson #04.pdf 4.36MB

Dataset

train_catvnoncat.h5 2.45MB

thumbs_up.jpg 989KB

test_catvnoncat.h5 602KB

train_signs.h5 12.67MB

test_signs.h5 1.41MB

Lesson #04 Task #02 Hyperparameter Tuning using Weights and Biases.ipynb 425KB

Lesson #04 Task #01 Hyperparameter Tuning using Keras Tuner.ipynb 50KB

Lesson #04 Task #03 Batch Normalization.ipynb 205KB

README.md 6KB

Lesson #07

Lesson #07.pdf 3.22MB

Lesson_7.ipynb 119KB

## Universidade Federal do Rio Grande do Norte ### Programa de Pós-Graduação em Engenharia Elétrica e de Computação PPgEEC #### EEC2003 - Special Topics in AI - Lesson #01 Introduction - Course Outline & Presentation - [Google Colab Introduction](https://www.loom.com/share/8a4f0d34b3cb4d9ea04b6dcf0b3d1aca) - [Google Colab Cont.](https://www.loom.com/share/d96cb0af7d9c4416bfe8145c93248a11) - Lesson #02 Fundamentals - [Introduction to Deep Learning and TensorFlow - Outline](https://www.loom.com/share/caeb19f6f7694bfdba3687a46b37298d) - [The perceptron](https://www.loom.com/share/bccf2bc2c7f24652b7b3b73825e0100f) - [Building Neural Networks](https://www.loom.com/share/f0ca49107b52458699210cbda8d3cb76) - [Matrix Dimension](https://www.loom.com/share/31862a3448f6427097e16adc773592a1) - [Applying Neural Networks](https://www.loom.com/share/f5ef63a357604bcebb577458cbfe85f6) - [Training a Neural Network](https://www.loom.com/share/38f251f7949d4d3c99097395ab9e3b74) - [Backpropagation with Pencil & Paper](https://www.loom.com/share/7093fed68d7342b189ef2f9b85e93b2d) - [Learning rate & Batch Size](https://www.loom.com/share/183248cfec9f46a5bc0ae7ec410aa291) - [Exponentially Weighted Average](https://www.loom.com/share/b84b1452ab5d4193b63481910d9323b1) - [Adam, Momentum, RMSProp, Learning Rate Decay](https://www.loom.com/share/101a5956c6f04d31843f37c4be089978) - Lesson #03 Better Generalization vs Better Learning - Better Generalization - [Better Deep Learning - Outline](https://www.loom.com/share/33ceae6510ca4321b95425efc7c7828e) - [Spliting Data](https://www.loom.com/share/436be4492b0549baba57c52d40941cc3) - [Bias vs Variance](https://www.loom.com/share/9cc90385906d458b9baafc19c686cc8e) - [Weight Regularization](https://www.loom.com/share/6f8e8101bee243318302cb3742fbdb8c) - [Weight Constraint](https://www.loom.com/share/b65c8294dcda4746a0a9a9c9ea3b5cb4) - [Dropout](https://www.loom.com/share/c32f0a35d56b426ca988e05926787936) - [Promote Robustness with Noise](https://www.loom.com/share/ed48470b14a3460eac572dcf9d8838c9) - [Early Stopping](https://www.loom.com/share/c738eed439a34794a6b99c555b99afad) - Better Learning - [Data scaling](https://www.loom.com/share/7008b640440d412498578e27b8557471) - [Vanishing/Exploding Gradient](https://www.loom.com/share/fb427d71b7a74e2dab226445941d2d41) - [Fix Vanishing Gradient with Relu](https://www.loom.com/share/0cdfb9ba531540fca075444f7d732fc6) - [Fix Exploding Gradient with Gradient Clipping](https://www.loom.com/share/2b560922401442b7b078faf06801a3ad) - Lesson #04 Hyperparameter Tuning & Batch Normalization - [Outline](https://www.loom.com/share/0c5ccb3514ae4dc8b674ea5789f26645) - [Hyperparameter Tuning Fundamentals](https://www.loom.com/share/19920a5abadd4ce0bea8eadc26c778ee) - [Keras Tuner and Weights and Biases](https://www.loom.com/share/f27dcb1d5779432a906e19db7a834c65) - [Wandb - Part #01](https://www.loom.com/share/fed7cfc2a5414ad58637244f84add9b8) - [Wandb - Part #02](https://www.loom.com/share/16255ea534b34691a90801fe1d34ce6d) - [Batch Normalization Fundamentals](https://www.loom.com/share/adf8e445186d44caa79a83d0f3af97d8) - [Batch Normalization Math Details](https://www.loom.com/share/b2dad925916e4ae58d6a3cf3223be945) - [Batch Normalization Case Study](https://www.loom.com/share/d8113419cd56463eab0094df1a687cf0) - Lesson #05 Fundamentals of Convolutional Neural Networks (CNN) - [Outline](https://www.loom.com/share/3b88a9f39080434e8ee9ec87135ea9b8) - [CNN Introduction & Motivation](https://www.loom.com/share/bb17b639264c4b079b375b63c4c16085) - [Convolutional Layer](https://www.loom.com/share/dd6b3d3ae5b3430ead7ed09174dd1b4c) - [Case Study of Convolutional Layer](https://www.loom.com/share/48b2566fae7b4c9794526e41ffbd624c) - [Pooling Layer](https://www.loom.com/share/055e32dcd2564b13890ce19d72daba32) - [Fully Connected Layer](https://www.loom.com/share/7912c7d6e1a54a97b39c15efb5a9fc88) - [Case Study - Signs Dataset](https://www.loom.com/share/2ea8f406635b489faa9f7f5ec675de73) - Lesson #06 Convolutional Neural Networks (CNN) Architecture I - [Outline](https://www.loom.com/share/c218c4ac06684b13aa07049b39f4b4df) - [Typical CNN Architecture](https://www.loom.com/share/9aaf99dc6e75403b96e17b08da935313) - [Best practices when building your own CNN](https://www.loom.com/share/f1016707d00f4fd99c9dfb6295072554) - [LeNet-5](https://www.loom.com/share/ddaf75c76e6b4ab39b3741f43a655764) - [Training LeNet-5 using MNIST dataset](https://www.loom.com/share/59ca5879a00242b4814f822fa3c6a430) - [ImageNet & ILSVRC](https://www.loom.com/share/c4f94499da3a4769bd751c2970c3d6ab) - [AlexNet](https://www.loom.com/share/bc8475491fcf49d1b17305a0441843f7) - [Dogs vs Cats Challenge + HDF5](https://www.loom.com/share/ba0bfb5f5f6044fead11ebebee3eedd3) - [CNN Architectures - Hands on Part #01](https://www.loom.com/share/068d3145cd9c4b0291e5d2a83a2c494b) - [CNN Architectures - Hands on Part #02](https://www.loom.com/share/b847f834f62c47faaaa936137d163e08) - Lesson #07 Convolutional Neural Networks (CNN) Architecture II - [Outline](https://www.loom.com/share/12dabb12f4df48f29802976dedf28a93) - [VGG](https://www.loom.com/share/54f9a9a0c7164752a0daa2e11ab94304) - [VGG case study using CIFAR-10](https://www.loom.com/share/f8543faa8a9544f690e8d83493d053ff) - [How to use 1x1 convolutions](https://www.loom.com/share/4fa05c8a49814494a658e9ad337a8ef6) - [GoogLeNet](https://www.loom.com/share/3f6e276e64b845f390174e86b45a9e5f) - [MiniGoogleLeNet using CIFAR-10](https://www.loom.com/share/62f243ae46ec4796bb767a7e6cbd78c9) - [DeeperGoogLeNet and Tiny ImageNet Challenge](https://www.loom.com/share/af4eee545ac24cc59dae0482443efa67) - Lesson #08 ResNet - Lesson #09 Transfer Learning - [Outline](https://www.loom.com/share/a748317324724d029678eb33c877b82f) - [Feature Extractors](https://www.loom.com/share/e19e7431012c42aaa3957ebf0e756e8b) - [Feature Extractors - Case Study](https://www.loom.com/share/7492b62692b24617a3c87f3635af8942) - [Fine-Tuning](https://www.loom.com/share/eaf9c878e7474a1b864bd5e967ae2b05) - [Fine-Tuning Case Study I - Flowers 17](https://www.loom.com/share/57830f6349e84360a71ceac147e0ac41) - [Fine-Tuning Case Study II - Cats and Dogs](https://www.loom.com/share/9253be684b06472ea9705eb1cf7cf321)

评论收藏

内容反馈

版权申诉