This seems like a novice question and yet I don't understand it. How do I import and use a third party js library in vue 3? In this case I'm trying to use Danfo.js https://danfo.jsdata.org/getting-started by doing npm install danfojs (even though it only shows to use cdn for browser use I think this is the same thing but correct me if wrong). Then idk if this is something that I import in each file I want to use or if I do it in main.js and it works globally automatically or what. I tried making main.js

import { createApp } from 'vue'
import App from './App.vue'

import danfo from 'danfojs';

const app = createApp(App)
app.use(danfo)
app.mount('#app')

and then idk if that's correct but if so then how do I call dfd from inside the setup() of a component

function danfoTest(){
      console.log('idk', dfd)
      const json_data = [
        { A: 0.4612, B: 4.28283, C: -1.509, D: -1.1352 },
        { A: 0.5112, B: -0.22863, C: -3.39059, D: 1.1632 },
        { A: 0.6911, B: -0.82863, C: -1.5059, D: 2.1352 },
        { A: 0.4692, B: -1.28863, C: 4.5059, D: 4.1632 }]
                    
        const df = new dfd.DataFrame(json_data)
        console.log('here')
        df['A'].print()
    }

Idk if this is a lack of vue 3 understanding or lack of Danfo.js understanding but either way would appreciate some help, thanks!

Also is is possible can is only option? When adding the <script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/index.min.js"></script> tag to index.js it did work but I was getting errors in terminal about dfd not being defined although calling dfd did work. I assume because it loads the script later idk either way I think I want the npm install way and the npm install danfojs-node I believe is for a node version not the browser version which is why I did npm install danfojs

Also side question is this a long-term supported project with tensor or more of a side project

I am trying to install DanfoJS on windows 10 using below command:

npm install danfojs-node

but I get following error when I run npm start

Error: The Node.js native addon module (tfjs_binding.node) can not be found at path:

I tried to ran npm rebuild @tensorflow/tfjs-node build-addon-from-source but still having problem.

sortValues seems inconvinient because it does reset the index and its hard to reindex it if I only have numbers...

The Index gets ignored too often in general for example:

Series.toJSON() or toCSV() only includes the values. This is inconvinient. There should be an option to not drop the index. Pandas doesn't drop the index either.

Also is it possible to combine the functions valueCounts() and sortValues, just like pandas does? The missing index when passing a Series makes further calculations unnecessarily complicated..

If it is intended to leave a passed Series 1 dimensional there should be atleast a method that converts a Series to a Dataframe by adding a new Index. This is done by reindex() in pandas, but in danfojs this only seems to drop the current index and give it new values.

Hi everyone, I have issue when i install danfojs-node with npm :

npm WARN deprecated @types/[email protected]: This is a stub types definition. csv-parse provides its own type definitions, so you do not need this installed.
npm WARN deprecated [email protected]: Please upgrade to @mapbox/node-pre-gyp: the non-scoped node-pre-gyp package is deprecated and only the @mapbox scoped package will recieve updates in the future
npm WARN deprecated [email protected]: Critical security vulnerability fixed in v0.21.1. For more information, see https://github.com/axios/axios/pull/3410
npm ERR! code 1
npm ERR! path C:\Users\**\Desktop\NodeJS Developpement\Mysql JS\node_modules\@tensorflow\tfjs-node
npm ERR! command failed
npm ERR! command C:\WINDOWS\system32\cmd.exe /d /s /c node scripts/install.js
npm ERR! CPU-windows-3.2.0.zip
npm ERR! * Downloading libtensorflow
npm ERR!
npm ERR! * Building TensorFlow Node.js bindings
npm ERR! symlink ./lib/napi-v8 failed:  Error: Command failed: node scripts/deps-stage.js symlink ./lib/napi-v8
npm ERR!   * Symlink of lib\napi-v8\tensorflow.dll failed, creating a copy on disk.
npm ERR! node:internal/process/promises:245
npm ERR!           triggerUncaughtException(err, true /* fromPromise */);
npm ERR!           ^
npm ERR!
npm ERR! [Error: ENOENT: no such file or directory, copyfile 'C:\Users\**\Desktop\NodeJS Developpement\Mysql JS\node_modules\@tensorflow\tfjs-node\deps\lib\tensorflow.dll' -> 'C:\Users\**\Desktop\NodeJS Developpement\Mysql JS\node_modules\@tensorflow\tfjs-node\lib\napi-v8\tensorflow.dll'] {
npm ERR!   errno: -4058,
npm ERR!   code: 'ENOENT',
npm ERR!   syscall: 'copyfile',
npm ERR!   path: 'C:\\Users\\**\\Desktop\\NodeJS Developpement\\Mysql JS\\node_modules\\@tensorflow\\tfjs-node\\deps\\lib\\tensorflow.dll',
npm ERR!   dest: 'C:\\Users\\**\\Desktop\\NodeJS Developpement\\Mysql JS\\node_modules\\@tensorflow\\tfjs-node\\lib\\napi-v8\\tensorflow.dll'
npm ERR! }
npm ERR!
npm ERR!     at ChildProcess.exithandler (node:child_process:326:12)
npm ERR!     at ChildProcess.emit (node:events:369:20)
npm ERR!     at maybeClose (node:internal/child_process:1067:16)
npm ERR!     at Process.ChildProcess._handle.onexit (node:internal/child_process:301:5) {
npm ERR!   killed: false,
npm ERR!   code: 1,
npm ERR!   signal: null,
npm ERR!   cmd: 'node scripts/deps-stage.js symlink ./lib/napi-v8'
npm ERR! }

npm ERR! A complete log of this run can be found in:
npm ERR!     C:\Users\**\AppData\Local\npm-cache\_logs\2021-03-27T00_12_51_686Z-debug.log

The problem come from tensorflow and i have no idea to solve it. Thank you

Using the get started query example https://danfo.jsdata.org/api-reference/dataframe/danfo.dataframe.query

let data = {"A": [30, 1, 2, 3],
           "B": [34, 4, 5, 6],
           "C": [20, 20, 30, 40]}
           
let cols = ["A", "B", "C"]
let df = new dfd.DataFrame(data, { columns: cols })
df.print() //before query

let query_df = df.query({ "column": "B", "is": ">", "to": 5 })
table(query_df)

I modified the condition to

let data = {"A": [30, 1, 2, 3],
           "B": [34, 4, 5, 6],
           "C": [20, 20, 30, 40]}
           
let cols = ["A", "B", "C"]
let df = new dfd.DataFrame(data, { columns: cols })
df.print() //before query

let query_df = df.query({ "column": "B", "is": ">", "to": 40 })
table(query_df)

Error message

File format not supported for now

Expected Output: Return empty result/array

Describe the bug I get a Column names length mismatch error when trying to group by one column. If I try the same thing in Pandas everything works as it should.

To Reproduce Steps to reproduce the behavior:

dfd.read_csv('https://raw.githubusercontent.com/homomorfismo/spain-university-cutoffs/main/databases/spain_utf8.csv').then(a => df)
df.groupby(['University'])

Expected behavior Obtaining a list of dataframes each of which only contains observations with the same University value.

Screenshots

Desktop:

• OS: Windows 10
• Browser: Chrome
• Version: 95.0.4638.69 (64 bits)

This refactor converts the danfojs to use typescript. This is scheduled to be released as Danfo.Js v1.0 (stable), and will include a couple of breaking changes, bug fixes and new features.

Task and progress

• [x] Generic core
• [x] Frame class
• [x] Series class
• [x] File Readers /input and output
• [x] Utility class
• [x] Configuration class
• [x] Groupby class
• [x] Concat
• [x] Merge class
• [x] Preprocessing class
• [x] Str class (String)
• [x] Dt class (Datetime)
• [x] Indexing
• [x] Plotting class
• [x] Date range

Bug Fixes

• [x] Column data not being updated when mutating internal data array
• [x] Str class issue for non-strings
• [x] Better error message
• [x] Fix support for all JS date string
• [x] Fix loc slicing bug for row index with string labels
• [x] Fix wrong order of axis used during computation. axis=1 ==> row-wise operations and axis=0 ==> column-wise operation
• [x] apply function now works only across a specified axis

New Features

• [x] Ability to create an empty NDframe
• [x] Flag for toggling between low/high memory mode
• [x] Inplace support for all mutating operations
• [x] Ability to set Configuration values on NDframe creation
• [x] Support boolean mask for subsetting with iloc. E.g df.iloc({rows: df["count"].gt(5), columns: [0, 1]})
• [x] Support boolean mask for subsetting with loc. E.g df.loc({rows: df["count"].gt(5), columns: ["Count", "Size"]})
• [x] Update an existing column value via subsetting. E.g df["count"] = [1,3,4,5]
• [x] Add loc indexing support for Series
• [x] Add configuration support for formating table display in console
• [x] applyMap ==> Element wise apply function for DataFrame
• [X] and and or logical comparison support. E.g df.loc({rows: df['Salary_in_1000'].gte(100)).and(df['Age'].gt(60)) })
• [x] Streaming support for CSV and JSON files into NDframes
• [x] New IO function
  • [x] streamJSON ==> Supports streaming of local or remote JSON files into DataFrame.
  • [x] streamCSV ==> Supports streaming of local or remote CSV files into DataFrame.
  • [x] openCsvInputStream ==> Open a local/remote CSV file as a readable stream
  • [x] writeCsvOutputStream Open a local/remote CSV file as a writable stream
• [x] readCSV supports config values for headers, separator, etc.
• [x] toCSV supports config values for output
• [x] readJSON supports config values for headers, separator, etc.
• [x] toJSON supports config values for formating output
• [x] Query/Filter with multiple condition support. E.g df.loc({rows: df['Salary_in_1000'].gte(100)).and(df['Age'].gt(60)) })
• [x] streamCsvTransforme ==> Pipable stream transformer for incrementally transforming DataFrames

Is your feature request related to a problem? Please describe.

I normalized the data using StandardScaler. I then trained a tensorflowjs model. I now wish to predict. In order to carry this out, I need to first normalize and then use tensorflowjs's model predict() function - and then get the un-normalized prediction.

I assume to normalize that data that I will use for prediction, I re-use the same scaler to transform() (without running fit() first since we have already used fit() before during training - please let me know if I am wrong). But to get the un-normalized prediction, I need a reverse transform function

Describe the solution you'd like

sklearn appears to have an inverse_transform() function that needs to exist in Danfojs too

Describe alternatives you've considered

I am still trying to figure out how to get the inverse transform implemented, since I do not have the mean or standard deviation information with me. Going through Danfojs's source code to figure if it can be implemented or if I need to set up my own scaler and reverse transform functions

Describe the bug When creating a DataFrame, depending on the order of the columns the groupby() function works properly or returns an error.

To Reproduce This column order works perfectly:

let data = {
    worker: ["david", "david", "john", "alice", "john", "david"],
    hours: [5, 6, 2, 8, 4, 3],
    day: ["monday", "tuesday", "wednesday", "thursday", "friday", "friday"],
};
let df = new dfd.DataFrame(data);

df.groupby(["day"]).col(["hours"]).sum().print()

// ╔════════════╤═══════════════════╤═══════════════════╗
// ║            │ day               │ hours_sum         ║
// ╟────────────┼───────────────────┼───────────────────╢
// ║ 0          │ monday            │ 5                 ║
// ╟────────────┼───────────────────┼───────────────────╢
// ║ 1          │ tuesday           │ 6                 ║
// ╟────────────┼───────────────────┼───────────────────╢
// ║ 2          │ wednesday         │ 2                 ║
// ╟────────────┼───────────────────┼───────────────────╢
// ║ 3          │ thursday          │ 8                 ║
// ╟────────────┼───────────────────┼───────────────────╢
// ║ 4          │ friday            │ 7                 ║
// ╚════════════╧═══════════════════╧═══════════════════╝

df.groupby(["worker"]).count().print()
// ╔════════════╤═══════════════════╤═══════════════════╤═══════════════════╗
// ║            │ worker            │ hours_count       │ day_count         ║
// ╟────────────┼───────────────────┼───────────────────┼───────────────────╢
// ║ 0          │ david             │ 3                 │ 3                 ║
// ╟────────────┼───────────────────┼───────────────────┼───────────────────╢
// ║ 1          │ john              │ 2                 │ 2                 ║
// ╟────────────┼───────────────────┼───────────────────┼───────────────────╢
// ║ 2          │ alice             │ 1                 │ 1                 ║
// ╚════════════╧═══════════════════╧═══════════════════╧═══════════════════╝

But when I change the column order to the following it doesn't work:

let data = {
    hours: [5, 6, 2, 8, 4, 3],
    worker: ["david", "david", "john", "alice", "john", "david"],
    day: ["monday", "tuesday", "wednesday", "thursday", "friday", "friday"],
};
let df = new dfd.DataFrame(data);

df.groupby(["day"]).col(["hours"]).sum().print()
// Uncaught Error: Can't perform math operation on column hours
//    arithemetic groupby.ts:266
//    operations groupby.ts:417
//    count groupby.ts:431

df.groupby(["worker"]).count().print()
// Uncaught Error: Can't perform math operation on column hours
//    arithemetic groupby.ts:266
//    operations groupby.ts:417
//    count groupby.ts:431

Expected behavior I would expect that changing the order of the columns wouldn't make any change on the result.

Desktop (please complete the following information):

• OS: Windows 11
• Browser: Firefox v97.0.1, Chrome v98.0.4758.102, Edge v98.0.1108.56
• Version: -

Additional context I'm using the browser version, not the node.js one.

We should be able to pass a dataset to a function called train_test_split and get back splits of training and testing data similar to the sklearn train_test_split function here. .

Describe the bug I'm a little unsure if this is a bug? Or some sort of un-documented restriction?

If you try to groupby with 3 columns, it only groups by the first. If you try to groupby with 2 columns, it only groups by both.

Version

• Browser version of danfo.
• danfo - 0.2.5

To Reproduce

• Generate a dataframe with 3 text columns and one numeric column.
• Run a groupby - const dfGroup = df.groupby(["col1", "col2","col3"])
• Perform an aggregation dfGroup.col(["numeric_col"]).sum()

Expected behavior I get a dataframe with aggregation by all 3 columns. Instead I'm getting a dataframe with aggregation by a single column.

Is this currently a restriction or is this a bug?

Bumps json5 from 1.0.1 to 1.0.2.

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

Bumps json5 from 1.0.1 to 1.0.2.

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

Describe the bug I've installed danfojs@^1.1.2 in my React project using Vite and I'm getting the following error:

[plugin:vite:import-analysis] Failed to resolve entry for package "danfojs". The package may have incorrect main/module/exports specified in its package.json.

I really don't know what to do, I tried googling the error and found nothing. At first I thought it was a bug involving Vite, but then I tried using CRA and the bug still happens.

Screenshots

I'm using [email protected] in the browser, loaded from the official CDN link taken from the documentation. Tested in Chrome and Firefox

df = new dfd.DataFrame([ { "A" : "10"}, { "A" : "100"}]) df.dtypes >>> [ "int32" ] df["A"].sum() >>> "010100"

if i do:

df2 = df.asType("A", "int32") df2.dtypes >>> [ "int32" ] df2["A"].sum() >>> 110

file.csv

  import { readCSV } from 'danfojs-node';

  const ndf = await readCSV('/Users/ellison/Desktop/file.csv');
  ndf.print();
  ndf.loc({ rows: [1], columns: ['Variable 1'] }).print(); // error

Bumps flat to 5.0.2 and updates ancestor dependency mocha. These dependencies need to be updated together.

Updates flat from 4.1.1 to 5.0.2

Updates mocha from 7.2.0 to 10.2.0

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.