Fraud detection workflow example#676
Conversation
Signed-off-by: Jaya Venkatesh <jjayabaskar@nvidia.com>
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
Signed-off-by: Jaya Venkatesh <jjayabaskar@nvidia.com>
Signed-off-by: Jaya Venkatesh <jjayabaskar@nvidia.com>
Signed-off-by: Jaya Venkatesh <jjayabaskar@nvidia.com>
Signed-off-by: Jaya Venkatesh <jjayabaskar@nvidia.com>
|
I've only taken a quick look, but I noticed that the instructions are quite generic, and I think it would be helpful if we choose a provider (AWS, Brev or whatever is easier) and have the end to end example shown. Currently, when I read it I'm no sure the user will be quite sure where to start or how to set things up. How did you set up things and do the running? |
Signed-off-by: Jaya Venkatesh <jjayabaskar@nvidia.com>
Signed-off-by: Jaya Venkatesh <jjayabaskar@nvidia.com>
…esh19/deployment into fraud-detection-mlops-example
Signed-off-by: Jaya Venkatesh <jjayabaskar@nvidia.com>
| @@ -0,0 +1,635 @@ | |||
| { | |||
There was a problem hiding this comment.
| @@ -0,0 +1,635 @@ | |||
| { | |||
There was a problem hiding this comment.
Fix link for url to get API key, this has changed. The right link is https://org.ngc.nvidia.com/account/api-key
Reply via ReviewNB
| @@ -0,0 +1,635 @@ | |||
| { | |||
There was a problem hiding this comment.
We should source bashrc after install, follow thsi instructions we have in gpu deployment https://github.com/NVIDIA/accelerated-computing-hub/blob/main/tutorials/gpu-deployment/gpu-deployment-from-scratch.md#conda
Reply via ReviewNB
| @@ -0,0 +1,635 @@ | |||
| { | |||
There was a problem hiding this comment.
Add instructions to curl all the files needed into the brev instance, and tell the user to cd into the fraud-detection-mlops-pipeline
curl -sSL "URL" | bsdtar -xvf-
In the section
cp .env.example .env
add a source .env command so it loads the keys.
In the section
This prompts you to download transactions.tgz from IBM Box and extract it.
we should rewrite this to download locally they cred_card.zip, unpack it and then copy to brev the transactions.gz. Mention that they can either download the file locally and then copied it into brev using brev copy, or open jupyter and drag and drop it. Just be explicit on how to do it.
In the section
Start the infrastructure
We need to explain when to use which docker. compose, and what is each of the options for.
In the section
Verify the services are running:
Add explanation on how to port forward on brev, so they can open the links in their local machine browser.
Reply via ReviewNB
Signed-off-by: Jaya Venkatesh <jjayabaskar@nvidia.com>
Signed-off-by: Jaya Venkatesh <jjayabaskar@nvidia.com>
Signed-off-by: Jaya Venkatesh <jjayabaskar@nvidia.com>
2 participants
Towards #667
Adds a new workflow example demonstrating how to wrap the NVIDIA Financial Fraud Detection AI
Blueprint in production infrastructure
using Prefect (orchestration), MLflow (experiment tracking + model registry), and Triton (champion/challenger
serving with native versioning).
The notebook walks through the architecture, each pipeline stage, running and monitoring the pipeline, and
scaling to multiple machines.