Default Learning Rates for Optimizers are higher than those in Tensorflow.js/Keras · ml5js/ml5-library#1249 | Good First Issue

(2 comments) (0 reactions) (0 assignees)JavaScript (905 forks)batch import

bughelp wantedneural-network

Repository metrics

Stars: (6,136 stars)
PR merge metrics: (30d に merged PR はありません)

説明

Dear ml5 community,

I'm submitting a new issue. Please see the details below.

→ Step 1: Describe the issue 📝

The default Optimizer learning rates are significantly higher than those in tensorflow.js. This can lead to sub-optimal accuracy rates when training.

→ Step 2: Screenshots or Relevant Documentation 🖼

Default ML5.js Regression Adam Optimizer learning rate: .20 Default Tensorflow.js Adam Optimizer learning rate: .001

コントリビューターガイド

調査方針: ml5.js の NeuralNetwork/index.js のデフォルト学習率と TensorFlow.js のデフォルトを比較し、一致するように更新します。
技術スタック: javascript
領域: machine learning
Issue 種別: バグ
難度: 2
推定時間: 1時間未満
活動状況: アクティブ
明確さ: 明確
前提条件: JavaScript
初心者向け度: 85