Tweet
Tweet

Slide 1

Slide 1 text

The right tool for the job SciPy 2024 | Julia Silge https://juliasilge.github.io/scipy2024

Slide 2

Slide 2 text

Hello! ο‚› @juliasilge ο“Ά @[email protected] ο…§ youtube.com/juliasilge  juliasilge.com https://juliasilge.github.io/scipy2024

Slide 3

Slide 3 text

https://juliasilge.github.io/scipy2024

Slide 4

Slide 4 text

Tools for data science Quarto Shiny Great Tables Vetiver Pins Exciting new work! πŸŽ‰ https://juliasilge.github.io/scipy2024

Slide 5

Slide 5 text

Using multiple programming languages What does it cost? What do you gain? What can we give? https://juliasilge.github.io/scipy2024

Slide 6

Slide 6 text

https://juliasilge.github.io/scipy2024

Slide 7

Slide 7 text

For the individual It is expensive to learn new things There are benefits to specialization https://juliasilge.github.io/scipy2024

Slide 8

Slide 8 text

In an organization Consistency Complexity https://juliasilge.github.io/scipy2024

Slide 9

Slide 9 text

There should be one, and preferably only one, obvious way to do it https://juliasilge.github.io/scipy2024

Slide 10

Slide 10 text

Pins πŸ“Œ Python R import pins board = pins.board_temp() board.pin_write( very_nice_data, "important-stuff", type = "parquet") library(pins) board <- board_temp() board |> pin_write( very_nice_data, "important-stuff", type = "parquet") https://juliasilge.github.io/scipy2024

Slide 11

Slide 11 text

Pins πŸ“Œ Python R import pins board = pins.board_temp() board.pin_read("important-stuff") library(pins) board <- board_temp() board |> pin_read("important-stuff") Cost for individuals Cost for our organization https://juliasilge.github.io/scipy2024

Slide 12

Slide 12 text

https://juliasilge.github.io/scipy2024

Slide 13

Slide 13 text

In an organization Everyone can be more productive https://juliasilge.github.io/scipy2024

Slide 14

Slide 14 text

Practicality beats purity https://juliasilge.github.io/scipy2024

Slide 15

Slide 15 text

https://juliasilge.github.io/scipy2024

Slide 16

Slide 16 text

Vetiver 🏺 Python R from vetiver import VetiverModel, VetiverAPI v = VetiverModel( model_fit, "my-important-model", prototype_data = X_train) api = VetiverAPI(v) api.run() library(vetiver) library(plumber) v <- vetiver_model( model_fit, "my-important-model") pr() |> vetiver_api(v) |> pr_run() https://juliasilge.github.io/scipy2024

Slide 17

Slide 17 text

MLOps is… Versioning Managing change in models βœ… Deploying Putting models in REST APIs 🎯 Monitoring Tracking model performance πŸ‘€ https://juliasilge.github.io/scipy2024

Slide 18

Slide 18 text

For the individual You can scale your impact Consider the long term Increase your vocabulary https://juliasilge.github.io/scipy2024

Slide 19

Slide 19 text

https://juliasilge.github.io/scipy2024

Slide 20

Slide 20 text

Building tools Learn from one community Bring to a different one https://juliasilge.github.io/scipy2024

Slide 21

Slide 21 text

https://juliasilge.github.io/scipy2024

Slide 22

Slide 22 text

https://juliasilge.github.io/scipy2024

Slide 23

Slide 23 text

https://juliasilge.github.io/scipy2024

Slide 24

Slide 24 text

Positron Positron is a next-generation data science IDE Positron is a very early stage project https://github.com/posit-dev/positron/ https://juliasilge.github.io/scipy2024

Slide 25

Slide 25 text

Thank you! ο‚› @juliasilge ο“Ά @[email protected] ο…§ youtube.com/juliasilge  juliasilge.com https://juliasilge.github.io/scipy2024