Comments (3)
I'm not sure what you mean by this. Is option optional then? Or you ignore it if the output of the model doesn't actually match the signature and return it anyway?
Can you provide some examples?
from multi-model-server.
For example:
In SSD example, output shape in signature file is [1, 6132, 6].
This is actually the shape of NDArray that we get after forward pass (inference).
However, final output from the SSD server is a list of tuples - [(class, x1,y1,x2,y2)..] depending on how many objects were detected.
A new user can be confused thinking input and output in signature talks about input and output to the service rather than knowing it is the signature of MXNet model.
So we need to clarify somehow in the documentation
from multi-model-server.
The signature.json file contains a mix of two: (1) Model input and output shapes (in inputs and outputs) (2) Service input and output data types (MIME types) (in input_type and output_type)
This can indeed be confusing for users and we need to clearly document it.
@aaronmarkham can you please address this confusion?
from multi-model-server.
Related Issues (20)
- Is it supporting Apple M1 Chip ?
- Preloading models on Sagemaker multi-model endpoint doesn't work
- python version HOT 1
- Model specific custom python package installation
- Validate MMS process during start
- Invalid Response Headers Set for Non MME Inference Scenario (scikit learn container)
- Config option to increase or disable model load timeout HOT 1
- Improve MMS model loading exception handling
- Streaming support on MMS HOT 1
- Update documentation to establish the difference between backend time and backend response time
- readAddress(..) failed: Connection reset by peer HOT 1
- Overriding the model routing logic
- How to handle invalid input! HOT 1
- MMS Server getting stuck while registering the model and says "worker pid is not available yet"
- command not found: multi-model-server HOT 1
- log4j2 metrics JsonLayout / QLogLayout logger broken
- Question: Can I use cuda 11, for gpu inference on mms HOT 4
- how to change url format
- mms-gpu - cuda error - No kernel image is available for execution on the device
- Issue: Memory Leak when serving multiple models HOT 4
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from multi-model-server.