Release Notes
Date: 29th March 2021
[Feature] #154 Perform groupby operation on grouped columns directly:
group = df.groupby(['A"])group.min()​​groupby.apply((x) => x.add(2))groupby.col(["C"]).apply((x) => x.min())
Contributors: @steveoni @PrawiraGenestonlia @woosuk288 @risenW
Date: 6th March 2021
We added/updated the following features:
Fix error thrown when danfojs CDN is placed in an HTML header
Smaller bundle size for browser: From ~1.7mb to about ~550kb
Stopped bundling Danfojs with Plotly. This means that as of v0.2.3, we no longer ship with Plotly distribution due to the huge size. Plotly plots are still supported, but in order to make them, you have to explicitly add the Plotly CDN or package.
A simple example:
<!DOCTYPE html><html lang="en"><head><meta charset="UTF-8"><meta name="viewport" content="width=device-width, initial-scale=1.0"><script src="https://cdn.plot.ly/plotly-1.2.0.min.js"></script><script src="https://cdn.jsdelivr.net/npm/[email protected]/lib/bundle.min.js"></script>​<title>Document</title></head>​<body>​<div id="div1"></div><div id="div2"></div><div id="div3"></div>​<script>​dfd.read_csv("https://raw.githubusercontent.com/plotly/datasets/master/finance-charts-apple.csv").then(df => {​df['AAPL.Open'].plot("div1").box() //makes a box plot​df.plot("div2").table() //display csv as table​new_df = df.set_index({ key: "Date" }) //resets the index to Date columnnew_df.plot("div3").line({ columns: ["AAPL.Open", "AAPL.High"] }) //makes a timeseries plot​}).catch(err => {console.log(err);})​</script></body>​</html>
[LATEST] Release (v0.2.2)​
Date: 14th February 2021
We added/updated the following features:
Fix babel runtime issue in node version
Smaller size in browser version in this version. From ~8mb to about ~5mb
Fix browser tag issue and returns back to specific versioning instead of @latest
Danfojs now ships with an exported version of tensorflowjs-node (2.8.5). This fixes the double dependency issue when building ML models, as you no longer need to install/import tensorflowjs separately.
To use tensoflowjs-node, you can reference it from danfo as shown below:
const dfd = require("danfojs-node")const tf = dfd.tf //contains a reference to the tensorflowjs-node library​const model = tf.sequential();model.add(tf.layers.dense({ inputShape: [7], units: 124, activation: 'relu', kernelInitializer: 'leCunNormal' }));model.add(tf.layers.dense({ units: 64, activation: 'relu' }));model.summary();
Date: 25th September 2020
We added/updated the following features:
​Read JSON files from local and remote URL into DataFrame (New feature)
​Read Excel files from local or remote URL into DataFrame (New feature)
Fix string comparison bug in query function (Fix)
Fix null value bug when creating DataFrame from JSON files (Fix)
Add relative import for reading files into DataFrame
Add sort_index function to DataFrame and Series (New feature)
Minor patch and overall optimizations (Fix)
Contributors: Rising Odegua, Stephen Oni, Jhenner Tigreros, Aditya Zope
Date: 15th August 2020
This is a minor release for browser-based environments. We added/updated the following features:
Ability to create DataFrame/Series from Tensors
Add iloc function for Series
Update addColumns function to accept Series
Add inplace option for fillna​
Fix bug in rename function of DataFrame
Add Inplace option for query​
Fixed upper bound bug in indexing
Contributors: Rising Odegua, Stephen Oni
​
