A repo about using Tensorflow.js.
- Node.js
- TensorFlow.js
- Parcel
- Http-server
To predict a person's weight based on his height.
根据身高预测体重。
To predict the result of a linear-reggression function.
判断一个线性回归的结果，只有一个神经元。
To predict a logistic regression problem, which introduce the loss.
判断一个逻辑回归问题。
A multi-layer neural network is used to solve an XOR problem.
通过多层神经网络解决XOR问题。
Use weight decay / dropout to solve the overfit problem.
通过 权重衰减/丢弃法 来解决过拟合问题。
Use Iris data to train the classification model.
通过经典的鸾尾花数据来训练带有softmax的多层神经网络分类模型。
Classical model to recognize handwritten numbers.
通过Mnist数据训练模型识别手写数字>
This chapter is to show you how to use the pre-training model to classify the items.
通过使用预训练模型进行物品分类。
Classify the logo through transfer learning model.
通过迁移学习模型识别三种商标。
Use pre-training model to recognize speech.
使用tfjs-models的speech-commands包加载模型，训练语音识别。
- First step
npm install
npm i parcel -g
npm i http-server -g
- Second step
http-server data --cors
- Final step
parcel ***(which part u want to use, for example: iris)/index.html
Then you can see the page at localhost:1234