MLOps Chat: How Should We Test ML Models? with Data Scientist Jeremy Jordan

  Рет қаралды 7,646

DVCorg

DVCorg

Күн бұрын

Пікірлер: 21
@firstpersonviews1996
@firstpersonviews1996 4 жыл бұрын
Interesting interview, thank you for maintaining such great channel!
@rajputjay9856
@rajputjay9856 4 жыл бұрын
Thanks for wonderful talk and knowledge chats
@eleonoracortese6967
@eleonoracortese6967 Жыл бұрын
Thank you for your tutorials!
@dvcorg8370
@dvcorg8370 Жыл бұрын
Glad you like them!
@venkateshvenkataramani2386
@venkateshvenkataramani2386 3 жыл бұрын
Wonderful content! Keep up the good work!
@dvcorg8370
@dvcorg8370 3 жыл бұрын
Thank you! Will do!
@rajangupta5432
@rajangupta5432 2 жыл бұрын
Hi elle, from Tester perspective can you make some videos where you can cover what are the steps a tester do when ML developer deploy a model and hand it over to QA for verification. Please
@nilanshrajput791
@nilanshrajput791 4 жыл бұрын
I cannot find the Apple library "Checklist" for testing for NLP models that he mentioned, can anyone provide relevant links for it.
@dvcorg8370
@dvcorg8370 4 жыл бұрын
Is this what you're looking for? github.com/marcotcr/checklist
@dvcorg8370
@dvcorg8370 4 жыл бұрын
@@nilanshrajput791 that's kinda funny :)
@dvcorg8370
@dvcorg8370 4 жыл бұрын
@@nilanshrajput791 that's awesome timing. Keep us posted on how your test suite goes!
@ndamulelosbg8887
@ndamulelosbg8887 2 жыл бұрын
Interesting insights!
@mohitsinghpawar5682
@mohitsinghpawar5682 4 жыл бұрын
Thanks for this video 🔥 . One question does CML supports flask with rest API .. like can it run APIs too on GitHub actions? Thanks
@mohitsinghpawar5682
@mohitsinghpawar5682 4 жыл бұрын
Either in CML or in DVC
@dvcorg8370
@dvcorg8370 4 жыл бұрын
@@mohitsinghpawar5682 Hi Mohit, indirectly, yes! So your flask app needs to be hosted on a machine that will be persistently on (if you want to use the API). But your GitHub Actions runner will turn off after the workflow is done. So you could use your GitHub Actions runner to deploy the flask app to another machine (maybe Heroku could help here?)
@mohitsinghpawar5682
@mohitsinghpawar5682 4 жыл бұрын
Thanks a lot 😊 ..Is there any docs or example for such ? Thanks
@dvcorg8370
@dvcorg8370 4 жыл бұрын
@@mohitsinghpawar5682 We don't have any ourselves, but these docs from other authors look helpful: blog.siliconvalve.com/2019/03/22/use-github-actions-to-deploy-a-python-flask-web-app-to-azure-app-service-on-linux/ medium.com/swlh/get-started-with-ci-cd-using-github-actions-ca32d34b2943 www.clouddev.engineering/how-to-deploy-a-python-api-with-github-actions-cicd/
@mohitsinghpawar5682
@mohitsinghpawar5682 4 жыл бұрын
@@dvcorg8370 thanks a lot❤️
@patriciarodrigues9132
@patriciarodrigues9132 3 жыл бұрын
Can we have more videos of using CI CD pipelines along with dvc to do a deployment perhaps using something like django
@dvcorg8370
@dvcorg8370 3 жыл бұрын
Noted, Patricia! Deploying an app that you can interact with is a compelling use case, particularly because it can help you understand the behavior of your deployed model in a way that visuals & tables alone can't.
Гениальное изобретение из обычного стаканчика!
00:31
Лютая физика | Олимпиадная физика
Рет қаралды 4,8 МЛН
Леон киллер и Оля Полякова 😹
00:42
Канал Смеха
Рет қаралды 4,7 МЛН
Chain Game Strong ⛓️
00:21
Anwar Jibawi
Рет қаралды 41 МЛН
99.9% IMPOSSIBLE
00:24
STORROR
Рет қаралды 31 МЛН
Trey Causey: Testing for Data Scientists
39:34
PyData
Рет қаралды 9 М.
5 concepts of A/B testing you should know as a Data Scientist
11:14
Behavioral Testing of ML Models (Unit tests for machine learning)
12:22
Talk - Carlos Kidman: Testing Machine Learning Models
31:12
Transformers (how LLMs work) explained visually | DL5
27:14
3Blue1Brown
Рет қаралды 4,1 МЛН
Seamless MLOps with Seldon and MLflow
50:22
Databricks
Рет қаралды 10 М.
Unit Testing for Data Scientists - Hanna Torrence
39:11
PyData
Рет қаралды 26 М.
Гениальное изобретение из обычного стаканчика!
00:31
Лютая физика | Олимпиадная физика
Рет қаралды 4,8 МЛН