DataFrame.sample
Return a random sample of rows from DataFrame.
danfo.DataFrame.sample(num, options)
Parameters | Type | Description | Default |
|---|---|---|---|
num | Integer | The number of rows to return. Defaults to 5, which shuffles and return all rows. | 5 |
options | Object | seed: An integer specifying the random seed that will be used to create the distribution. Ensures reproducibility of generated samples. | { seed: 1 } |
Node
Browser
const dfd = require("danfojs-node")
async function sample_data() {
let data = {
Name: ["Apples", "Mango", "Banana", "Pear"],
Count: [21, 5, 30, 10],
Price: [200, 300, 40, 250],
};
let df = new dfd.DataFrame(data);
let s_df = await df.sample(2);
s_df.print();
}
sample_data()
Output
╔═══╤═══════════════════╤═══════════════════╤═══════════════════╗
║ │ Name │ Count │ Price ║
╟───┼───────────────────┼───────────────────┼───────────────────╢
║ 0 │ Apples │ 21 │ 200 ║
╟───┼─────────────────── ┼───────────────────┼───────────────────╢
║ 2 │ Banana │ 30 │ 40 ║
╚═══╧═══════════════════╧═══════════════════╧═══════════════════╝
By setting
seed when using sample, you can ensure that the random sampling is reproducible.Node
Browser
const dfd = require("danfojs-node")
async function load_data() {
let data = {
Name: ["Apples", "Mango", "Banana", "Pear"],
Count: [21, 5, 30, 10],
Price: [200, 300, 40, 250],
};
let df = new dfd.DataFrame(data);
let s_df = await df.sample(3, { seed: 2 });
s_df.print();
}
load_data()
Output
╔════════════╤═══════════════════╤═══════════════════╤═══════════════════╗
║ │ Name │ Count │ Price ║
╟────────────┼───────────────────┼───────────────────┼───────────────────╢
║ 1 │ Mango │ 5 │ 300 ║
╟────────────┼───────────────────┼───────────────────┼───────────────────╢
║ 2 │ Banana │ 30 │ 40 ║
╟────────────┼───────────────────┼───────────────────┼───────────────────╢
║ 0 │ Apples │ 21 │ 200 ║
╚════════════╧═══════════════════╧═══════════════════╧═══════════════════╝
Last modified 1yr ago