Thank you Ashutosh for creating the very useful video for beginners, I have a question, I am doing the model development under Databricks notebook and for registry I am using MLFlow, do i still need other tool for deployment to prod ?
It depends. Using mlflow you can expose the model serving api and then it can be consumed in other applications. FastAPi is another way to create an api for the same.
Hello, mlflow server in your case was local that you used in endpoint, what if i am usint a remote sever like dagshub, how will i take the Letest model automatically in my dagshub
Thank you so much for the tutorials. I have been following it closely, but now I am stuck with part 3. Facing some internal server error which I do not know how to fix.
Very useful and u r great trainer... Your clear explanation makes better understanding sir. Pls how to create docker file and cotainer using mlflow registry artifcats.... 🙏🙏🙏🙏
Could you please help me to clarify these doubts? 1. How to run the below command dynamically based on the model I want to use ? `mlflow models serve --model-uri models:/iris-classifier/Production -p 1234 --no-conda` 2. How to automate the above process of picking the desired model and then starting the mlflow server? Thanks.
superb sir the way of explain each and evey point is awesome, i need one help from side sir , please create one project where u will be useing MLOPS(mlflow, DVC) and docker please sir request = ["humble_requset form my side 🤗🙏" for req in range(10000000000000000) ] print(request)
Very informative. Can you create a video on how to use mlflow within databricks? Specifically, calling the model using curl and python api in databricks. Thanks!
Model Serving using MLFlow model registry: ruclips.net/video/1EYVQGdXVOo/видео.html
we actually need some good people like u....
Thank you for explicitly explaining, each and everything with enormous patience.
Thank you for the kind words. Always glad to know when videos do the job.
Awesome Tutorial...
Thanks Krish
Your channel is a gem sir . Thank you for putting in so much of efforts . Helps a lot really !! 🙂
Thank you very much. Glad to know it helped.
You have made great videos on this subject and i am subscribing to encourage you!
Thank you.
you are a gem
Thank you 😊
Awesome explanation.Kindly make a video in
CI/CD mlops in databricks.
Ok will try to make one on this topic. Thank you
Very insightful, keep making such videos 😊
Thank you.
Very insightful
The video is soo dam good u r a gem
Thank you.
Thank you Ashutosh for creating the very useful video for beginners, I have a question, I am doing the model development under Databricks notebook and for registry I am using MLFlow, do i still need other tool for deployment to prod ?
It depends. Using mlflow you can expose the model serving api and then it can be consumed in other applications. FastAPi is another way to create an api for the same.
Hello, mlflow server in your case was local that you used in endpoint, what if i am usint a remote sever like dagshub, how will i take the Letest model automatically in my dagshub
Grateful for your succinct and clear elucidation. Would you kindly provide the Git or Collab link pertaining to the remaining segment?
Thank you so much for the tutorials. I have been following it closely, but now I am stuck with part 3. Facing some internal server error which I do not know how to fix.
Btw why don't you do the inference directly in fastapi? Isn't it using mlflow endpoint adding overhead cost?
Very useful and u r great trainer... Your clear explanation makes better understanding sir. Pls how to create docker file and cotainer using mlflow registry artifcats.... 🙏🙏🙏🙏
Ok, will try to make one video on this topic।
Video is very explained and good but git is missing code file for this one.
Could you please help me to clarify these doubts?
1. How to run the below command dynamically based on the model I want to use ?
`mlflow models serve --model-uri models:/iris-classifier/Production -p 1234 --no-conda`
2. How to automate the above process of picking the desired model and then starting the mlflow server?
Thanks.
You can use shell scripting and send the model uri as an argument or pass via command line
When do you plan to add flask application to this wonderful series?
Very soon. Most probably the next video which is coming on Monday.
Just uploaded the video on MLFlow model deployment using flask. You can check it here: ruclips.net/video/XZSiHpOoOR0/видео.html
very informative videos.. Could you please share the code of fastAPI and Flask for model deployment? Thank you so much.
fast api ask me to give token.....which token should I provide?
superb sir the way of explain each and evey point is awesome, i need one help from side sir , please create one project where u will be useing MLOPS(mlflow, DVC) and docker please sir
request = ["humble_requset form my side 🤗🙏" for req in range(10000000000000000) ]
print(request)
Very informative. Can you create a video on how to use mlflow within databricks? Specifically, calling the model using curl and python api in databricks. Thanks!
Sure. Will try create one on this topic.
sir files are not present in the github, can you please check .
what is invocations ?
Cannot find any file on fastapi manual in the the github link...please reload the document in the github..thanks
@ashutosh : Too good explanation. But can you upload this notebook
Can you upload all the files on your github please?
Hi Sandeep, you can download the files from the link below:
ashutoshtripathi.com/2023/03/25/machine-learning-model-deployment-using-docker-container/
sir where can i find the main and main2 python files?
i will upload the files in GitHub and will share the link. You will find it in the description. Just give me 2 days, I came on vacation.
@@AshutoshTripathi_AI where to get notebooks for deployment?
@@sarthak7413 you can find the notebook link in the description.
@@AshutoshTripathi_AI Can you please help me with exact file name.I am unable to locate it in github repo
@@AshutoshTripathi_AI : They are not there in git hub link Ashutosh..