Comments (6)
We only have ".mlem.yaml" inside ".mlem" dir, so we can just rename them ".yaml".
Other model/dataset metafiles are just "mlem.yaml". We can rename them into ".mlem" or "meta.mlem" of course, but not sure if we should.
Another feature we can get: name those files <object_type>.mlem
- this way we would know which type of object is there without reading metadata (but this will make checking if dir is mlem object a bit harder)
from mlem.
- The other ➕ is faster repo scanning for example. I think rn we dont have enought arguments, issues like #37 could add something to the table. For now lets remove this from alpha scope.
- Yep, and we can just leave
.yaml
for now. As for pydantic, yes, if it is possible in jsonschema, and we will try to contribute our polymorphic stuff to pydantic, we should implement this too.
from mlem.
As for jsonschema, there is a big problem: since pydantic does not support polymorphism, the schema it generates lacks this also. And I can't say for sure if it is even possible: we need to create jsonschema definitions for every implementation of each polymorphic hierarchy and somehow configure fields to be "one of the subclasses of type X".
from mlem.
Another feature we can get: name those files
<object_type>.mlem
.
On one hand, I would say the bigger ➕ may be for users as those who new to a particular repo can easier understand what is a particular folder: data or model. On the other, I suppose it's common to have those two in specific parent folders which names explain this good enough. Considering this, it's not clear for me, whether this worth the efforts. What do you think?
As for jsonschema, there is a big problem
Got it. So, the only thing we can do here for now is to write some generic rules, like "treat .mlem
just as .yaml
. Btw, do you think there is a possibility to fix this by contributing to official pydantic repo?
from mlem.
I think I am happy with the way things are right now (after #114 the thought of adding different names for metafiles gives me nightmares). @aguschin WDYT? should we close this?
from mlem.
Agreed, let's close this!
from mlem.
Related Issues (20)
- flyctl deployment failure: org slug must be specified when not running interactively HOT 5
- `link` won't accept --rev flag / unexpected 404 HOT 5
- Expose custom information in FastAPI's `interface.json` HOT 1
- Expose GTO's model version in FastAPI's `interface.json` HOT 2
- Requirements aren't found when using `FunctionTransformer` HOT 1
- Confusing error in `deploy run flyio` when app exists HOT 2
- Bug: Validation error when loading linked model
- Pin flyctl version to <0.1.0 in MLEM
- Bug: image_name is missing from command arguments
- Use `classify-imports` instead of `isort` for import classification HOT 1
- Bug: Mlem couldn’t fetch your accounts information HOT 2
- Get started feedback
- How to use callbacks HOT 1
- fast.ai support HOT 2
- Fixing version collection HOT 2
- No clear way to overwrite requirements inferred by MLEM
- Uploading docker image is suboptimal
- Inferencing types at saving is not feasible sometimes
- Right way to add authentification middleware
- NameError: name 'UUID' is not defined HOT 3
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 mlem.