ranfang66 / github-page-comments Goto Github PK

https://kissfire.icu/2023/02/10/keras%E6%A8%A1%E5%9E%8B%E8%BD%AC%E6%8D%A2%E5%88%B0tflite%E6%A0%BC%E5%BC%8F%E5%B9%B6%E9%83%A8%E7%BD%B2%E5%88%B0rk3588/#more

keras模型的保存和加载keras模型主要有以下几种保存的形式： save/load weights123456## 模型保存： 只保存权重参数，不保存模型网络结构model.save_weights(‘./checkpoins/my_checkpoint’)## 加载模型model = create_model()model.load_weights(‘./checkpoints/m

https://kissfire.icu/2023/09/22/keras%E5%85%A5%E9%97%A8%E5%AD%A6%E4%B9%A0/#more

Keras快速入门深度学习对于数据预处理的要求神经网络不能直接处理原始数据，比如原始的文本，图像，表格等原始数据，需要将原始数据导入并转换为向量化(vectorized)和标准化(standardized)的表示。 文本文件需要被读取成字符串张量，然后再分割成单词，最终这些单词还需要被索引化并转换为整数张量。 图像需要被读取并解码成为整形张量，然后转换为浮点型并正则化到较小范围的数值（通常使用0

https://kissfire.icu/about/

👋 Hi, I’m @RanFang66 🏫 I graduated from Harbin Institude of Technology with a master degree of electrical engineering 👀 I’m interested in embeded software development, Linux, raspberry PI, OpenCV

https://kissfire.icu/2022/10/01/pid-controller-in-embedded-system/#more

PID IntroductionThe model of PID controller$$u(t)=K_pe(t)+K_i\int e(t)dt+K_d \frac{\mathrm{d}e(t)}{\mathrm{d}t}$$ Where u is the output of PID controller and e is the error of reference and feedb

https://kissfire.icu/cv/

The homepage of Ran Fang, a learner and developer, where to record his life and work

https://kissfire.icu/guestbook/

The homepage of Ran Fang, a learner and developer, where to record his life and work