serve
ML-Workspace
Our great sponsors
serve | ML-Workspace | |
---|---|---|
11 | 7 | |
3,941 | 3,317 | |
1.5% | 0.8% | |
9.6 | 2.7 | |
8 days ago | 5 months ago | |
Java | Jupyter Notebook | |
Apache License 2.0 | Apache License 2.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
serve
-
Show Show HN: Llama2 Embeddings FastAPI Server
What's wrong with just using Torchserve[1]? We've been using it to serve embedding models in production.
-
How to leverage a local LLM for a client?
Looks like you are already up to speed loading LLaMa models which is great. Assuming this is Hugging Face PyTorch checkpoint, I think it should be possible to spin up a TorchServe instance which has in-built support for API access and HF Transformers. Since scale and latency aren’t a big concern for you, this should be good enough start.
- Is there a course that teaches you how to make an API with a trained model?
-
Pytorch eating memory on every api call
You could split the service in two, flask for the web part and a service to serve the model, I haven't used it but there is https://pytorch.org/serve/
-
Google Kubernetes Engine : Unable to access ports exposed on external IP
I'm attempting to set up inference for a torchserve container, and it's really tough to figure out what's not allowing me to connect to my network with the ports that I'm trying to expose. I'm using Google Kubernetes Engine and Helm via tweaking one of the tutorials at [torchserve](github.com/pytorch/serve). Specifically, it's the GKE tutorial [here](https://github.com/pytorch/serve/tree/master/kubernetes).
-
BetterTransformer: PyTorch-native free-lunch speedups for Transformer-based models
I did a Space to showcase a bit the speedups we can have in a end-to-end case with TorchServe to deploy the model on a cloud instance (AWS EC2 g4dn, using one T4 GPU): https://huggingface.co/spaces/fxmarty/bettertransformer-demo
-
[D] How to get the fastest PyTorch inference and what is the "best" model serving framework?
For 2), I am aware of a few options. Triton inference server is an obvious one as is the ‘transformer-deploy’ version from LDS. My only reservation here is that they require the model compilation or are architecture specific. I am aware of others like Bento, Ray serving and TorchServe. Ideally I would have something that allows any (PyTorch model) to be used without the extra compilation effort (or at least optionally) and has some convenience things like ease of use, easy to deploy, easy to host multiple models and can perform some dynamic batching. Anyway, I am really interested to hear people's experience here as I know there are now quite a few options! Any help is appreciated! Disclaimer - I have no affiliation or are connected in any way with the libraries or companies listed here. These are just the ones I know of. Thanks in advance.
-
how to integrate a deep learning model into a Django webapp!?
If you built the model using pytorch or tensorflow, I'd suggest using torchserve or TF serving to serve the model as its own "microservice," then query it from your django app. Among other things, it will make retraining and updating your model a lot easier.
- Choose JavaScript ðŸ§
-
Popular Machine Learning Deployment Tools
GitHub
ML-Workspace
-
[D] I recently quit my job to start a ML company. Would really appreciate feedback on what we're working on.
Also check out: https://github.com/ml-tooling/ml-workspace, it a nice open source project with lots of packages ready to use.
- ML-Workspace
-
Coding for machine learning on Tab S8?
The other option - no reason why you couldn't host something on the desktop machine - web based IDE like R-Studio or Python - have a look at ml-workspace - https://github.com/ml-tooling/ml-workspace that runs in Docker and would provide interfaces for both Python and R, VSCode as well as a GPU accelerated variant for doing Tensorflow etc - either Windows or Linux can support Docker containers (Linux is less trouble apparently - I only have played with it in Linux personally)
-
Dynamically spin up VM (based on specific HTTPS request) and stop it once session is over?
It will be a web based IDE dev kit (like Jupyter Hub, or JupyterLab) if you are familiar with them)
- All-in-One Docker Based IDE for Data Science and ML
- Visual Studio Code now available as Web based editor for GitHub repos
-
[P] Install or update CUDA, NVIDIA Drivers, Pytorch, Tensorflow, and CuDNN with a single command: Lambda Stack
I'll stick with https://github.com/ml-tooling/ml-workspace, is a docker with all tools installed, also the option of using GPU, so I think is better than only for debian. This way anyone can use it.
What are some alternatives?
server - The Triton Inference Server provides an optimized cloud and edge inferencing solution.
JupyterLab - JupyterLab computational environment.
serving - A flexible, high-performance serving system for machine learning models
Gitpod - DEPRECATED since Gitpod 0.5.0; use https://github.com/gitpod-io/gitpod/tree/master/chart and https://github.com/gitpod-io/gitpod/tree/master/install/helm
JavaScriptClassifier - [Moved to: https://github.com/JonathanSum/JavaScriptClassifier]
keytotext - Keywords to Sentences
pinferencia - Python + Inference - Model Deployment library in Python. Simplest model inference server ever.
self-hosted - Sentry, feature-complete and packaged up for low-volume deployments and proofs-of-concept
kernl - Kernl lets you run PyTorch transformer models several times faster on GPU with a single line of code, and is designed to be easily hackable.
Code-Server - VS Code in the browser
deepsparse - Sparsity-aware deep learning inference runtime for CPUs
cocalc-docker - DEPRECATED (was -- Docker setup for running CoCalc as downloadable software on your own computer)