Examples built with TensorFlow.js
See the codeThis repository contains a set of examples implemented in TensorFlow.js.
Each example directory is standalone so the directory can be copied to another project.
| Example name | Demo link | Input data type | Task type | Model type | Training | Inference | API type | Save-load operations |
|---|---|---|---|---|---|---|---|---|
| abalone-node | Numeric | Loading data from local file and training in Node.js | Multilayer perceptron | Node.js | Node.js | Layers | Saving to filesystem and loading in Node.js | |
| addition-rnn | π | Text | Sequence-to-sequence | RNN: SimpleRNN, GRU and LSTM | Browser | Browser | Layers | |
| addition-rnn-webworker | Text | Sequence-to-sequence | RNN: SimpleRNN, GRU and LSTM | Browser: Web Worker | Browser: Web Worker | Layers | ||
| angular-predictive-prefetching | Numeric | Multiclass predictor | DNN | Browser: Service Worker | Layers | |||
| baseball-node | Numeric | Multiclass classification | Multilayer perceptron | Node.js | Node.js | Layers | ||
| boston-housing | π | Numeric | Regression | Multilayer perceptron | Browser | Browser | Layers | |
| cart-pole | π | Reinforcement learning | Policy gradient | Browser | Browser | Layers | IndexedDB | |
| chrome-extension | Image | (Deploying TF.js in Chrome extension) | Convnet | Browser | ||||
| custom-layer | π | (Defining a custom Layer subtype) | Browser | Layers | ||||
| data-csv | π | Building a tf.data.Dataset from a remote CSV | ||||||
| data-generator | π | Building a tf.data.Dataset using a generator | Regression | Browser | Browser | Layers | ||
| date-conversion-attention | π | Text | Text-to-text conversion | Attention mechanism, RNN | Node.js | Browser and Node.js | Layers | Saving to filesystem and loading in browser |
| electron | Image | (Deploying TF.js in Electron-based desktop apps) | Convnet | Node.js | ||||
| fashion-mnist-vae | Image | Generative | Variational autoencoder (VAE) | Node.js | Browser | Layers | Export trained model from tfjs-node and load it in browser | |
| interactive-visualizers | Image | Multiclass classification, object detection, segmentation | Browser | |||||
| iris | π | Numeric | Multiclass classification | Multilayer perceptron | Browser | Browser | Layers | |
| iris-fitDataset | π | Numeric | Multiclass classification | Multilayer perceptron | Browser | Browser | Layers | |
| jena-weather | π | Sequence | Sequence-to-prediction | MLP and RNNs | Browser and Node | Browser | Layers | |
| lstm-text-generation | π | Text | Sequence prediction | RNN: LSTM | Browser | Browser | Layers | IndexedDB |
| mnist | π | Image | Multiclass classification | Convolutional neural network | Browser | Browser | Layers | |
| mnist-acgan | π | Image | Generative Adversarial Network (GAN) | Convolutional neural network; GAN | Node.js | Browser | Layers | Saving to filesystem from Node.js and loading it in the browser |
| mnist-core | π | Image | Multiclass classification | Convolutional neural network | Browser | Browser | Core (Ops) | |
| mnist-node | Image | Multiclass classification | Convolutional neural network | Node.js | Node.js | Layers | Saving to filesystem | |
| mnist-transfer-cnn | π | Image | Multiclass classification (transfer learning) | Convolutional neural network | Browser | Browser | Layers | Loading pretrained model |
| mobilenet | π | Image | Multiclass classification | Convolutional neural network | Browser | Layers | Loading pretrained model | |
| polynomial-regression | π | Numeric | Regression | Shallow neural network | Browser | Browser | Layers | |
| polynomial-regression-core | π | Numeric | Regression | Shallow neural network | Browser | Browser | Core (Ops) | |
| quantization | Various | Demonstrates the effect of post-training weight quantization | Various | Node.js | Node.js | Layers | ||
| sentiment | π | Text | Sequence-to-binary-prediction | LSTM, 1D convnet | Node.js or Python | Browser | Layers | Load model from Keras and tfjs-node |
| simple-object-detection | π | Image | Object detection | Convolutional neural network (transfer learning) | Node.js | Browser | Layers | Export trained model from tfjs-node and load it in browser |
| snake-dqn | π | Reinforcement learning | Deep Q-Network (DQN) | Node.js | Browser | Layers | Export trained model from tfjs-node and load it in browser | |
| translation | π | Text | Sequence-to-sequence | LSTM encoder and decoder | Node.js or Python | Browser | Layers | Load model converted from Keras |
| tsne-mnist-canvas | Dimension reduction and data visualization | tSNE | Browser | Browser | Core (Ops) | |||
| webcam-transfer-learning | π | Image | Multiclass classification (transfer learning) | Convolutional neural network | Browser | Browser | Layers | Loading pretrained model |
| website-phishing | π | Numeric | Binary classification | Multilayer perceptron | Browser | Browser | Layers |
Except for getting_started, all the examples require the following dependencies to be installed.
cd into the directory
If you are using yarn:
cd mnist-core
yarn
yarn watch
If you are using npm:
cd mnist-core
npm install
npm run watch
The convention is that each example contains two scripts:
yarn watch or npm run watch: starts a local development HTTP server which watches the
filesystem for changes so you can edit the code (JS or HTML) and see changes when you refresh the page immediately.
yarn build or npm run build: generates a dist/ folder which contains the build artifacts and
can be used for deployment.
If you want to contribute an example, please reach out to us on Github issues before sending us a pull request as we are trying to keep this set of examples small and highly curated.
Before you send a pull request, it is a good idea to run the presubmit tests and make sure they all pass. To do that, execute the following commands in the root directory of tfjs-examples:
yarn
yarn presubmit
The yarn presubmit command executes the unit tests and lint checks of all
the exapmles that contain the yarn test and/or yarn lint scripts. You
may also run the tests for individual exampls by cd'ing into their respective
subdirectory and executing yarn, followed by yarn test and/or yarn lint.
(top 24 of 25)
497 followers Β· starred Aug 2020
691 followers Β· starred May 2024
1,514 followers Β· starred Apr 2022
109 followers Β· starred Sep 2018
Examples built with TensorFlow.js
See the codeThis repository contains a set of examples implemented in TensorFlow.js.
Each example directory is standalone so the directory can be copied to another project.
| Example name | Demo link | Input data type | Task type | Model type | Training | Inference | API type | Save-load operations |
|---|---|---|---|---|---|---|---|---|
| abalone-node | Numeric | Loading data from local file and training in Node.js | Multilayer perceptron | Node.js | Node.js | Layers | Saving to filesystem and loading in Node.js | |
| addition-rnn | π | Text | Sequence-to-sequence | RNN: SimpleRNN, GRU and LSTM | Browser | Browser | Layers | |
| addition-rnn-webworker | Text | Sequence-to-sequence | RNN: SimpleRNN, GRU and LSTM | Browser: Web Worker | Browser: Web Worker | Layers | ||
| angular-predictive-prefetching | Numeric | Multiclass predictor | DNN | Browser: Service Worker | Layers | |||
| baseball-node | Numeric | Multiclass classification | Multilayer perceptron | Node.js | Node.js | Layers | ||
| boston-housing | π | Numeric | Regression | Multilayer perceptron | Browser | Browser | Layers | |
| cart-pole | π | Reinforcement learning | Policy gradient | Browser | Browser | Layers | IndexedDB | |
| chrome-extension | Image | (Deploying TF.js in Chrome extension) | Convnet | Browser | ||||
| custom-layer | π | (Defining a custom Layer subtype) | Browser | Layers | ||||
| data-csv | π | Building a tf.data.Dataset from a remote CSV | ||||||
| data-generator | π | Building a tf.data.Dataset using a generator | Regression | Browser | Browser | Layers | ||
| date-conversion-attention | π | Text | Text-to-text conversion | Attention mechanism, RNN | Node.js | Browser and Node.js | Layers | Saving to filesystem and loading in browser |
| electron | Image | (Deploying TF.js in Electron-based desktop apps) | Convnet | Node.js | ||||
| fashion-mnist-vae | Image | Generative | Variational autoencoder (VAE) | Node.js | Browser | Layers | Export trained model from tfjs-node and load it in browser | |
| interactive-visualizers | Image | Multiclass classification, object detection, segmentation | Browser | |||||
| iris | π | Numeric | Multiclass classification | Multilayer perceptron | Browser | Browser | Layers | |
| iris-fitDataset | π | Numeric | Multiclass classification | Multilayer perceptron | Browser | Browser | Layers | |
| jena-weather | π | Sequence | Sequence-to-prediction | MLP and RNNs | Browser and Node | Browser | Layers | |
| lstm-text-generation | π | Text | Sequence prediction | RNN: LSTM | Browser | Browser | Layers | IndexedDB |
| mnist | π | Image | Multiclass classification | Convolutional neural network | Browser | Browser | Layers | |
| mnist-acgan | π | Image | Generative Adversarial Network (GAN) | Convolutional neural network; GAN | Node.js | Browser | Layers | Saving to filesystem from Node.js and loading it in the browser |
| mnist-core | π | Image | Multiclass classification | Convolutional neural network | Browser | Browser | Core (Ops) | |
| mnist-node | Image | Multiclass classification | Convolutional neural network | Node.js | Node.js | Layers | Saving to filesystem | |
| mnist-transfer-cnn | π | Image | Multiclass classification (transfer learning) | Convolutional neural network | Browser | Browser | Layers | Loading pretrained model |
| mobilenet | π | Image | Multiclass classification | Convolutional neural network | Browser | Layers | Loading pretrained model | |
| polynomial-regression | π | Numeric | Regression | Shallow neural network | Browser | Browser | Layers | |
| polynomial-regression-core | π | Numeric | Regression | Shallow neural network | Browser | Browser | Core (Ops) | |
| quantization | Various | Demonstrates the effect of post-training weight quantization | Various | Node.js | Node.js | Layers | ||
| sentiment | π | Text | Sequence-to-binary-prediction | LSTM, 1D convnet | Node.js or Python | Browser | Layers | Load model from Keras and tfjs-node |
| simple-object-detection | π | Image | Object detection | Convolutional neural network (transfer learning) | Node.js | Browser | Layers | Export trained model from tfjs-node and load it in browser |
| snake-dqn | π | Reinforcement learning | Deep Q-Network (DQN) | Node.js | Browser | Layers | Export trained model from tfjs-node and load it in browser | |
| translation | π | Text | Sequence-to-sequence | LSTM encoder and decoder | Node.js or Python | Browser | Layers | Load model converted from Keras |
| tsne-mnist-canvas | Dimension reduction and data visualization | tSNE | Browser | Browser | Core (Ops) | |||
| webcam-transfer-learning | π | Image | Multiclass classification (transfer learning) | Convolutional neural network | Browser | Browser | Layers | Loading pretrained model |
| website-phishing | π | Numeric | Binary classification | Multilayer perceptron | Browser | Browser | Layers |
Except for getting_started, all the examples require the following dependencies to be installed.
cd into the directory
If you are using yarn:
cd mnist-core
yarn
yarn watch
If you are using npm:
cd mnist-core
npm install
npm run watch
The convention is that each example contains two scripts:
yarn watch or npm run watch: starts a local development HTTP server which watches the
filesystem for changes so you can edit the code (JS or HTML) and see changes when you refresh the page immediately.
yarn build or npm run build: generates a dist/ folder which contains the build artifacts and
can be used for deployment.
If you want to contribute an example, please reach out to us on Github issues before sending us a pull request as we are trying to keep this set of examples small and highly curated.
Before you send a pull request, it is a good idea to run the presubmit tests and make sure they all pass. To do that, execute the following commands in the root directory of tfjs-examples:
yarn
yarn presubmit
The yarn presubmit command executes the unit tests and lint checks of all
the exapmles that contain the yarn test and/or yarn lint scripts. You
may also run the tests for individual exampls by cd'ing into their respective
subdirectory and executing yarn, followed by yarn test and/or yarn lint.
(top 24 of 25)
497 followers Β· starred Aug 2020
691 followers Β· starred May 2024
1,514 followers Β· starred Apr 2022
109 followers Β· starred Sep 2018