Enabling passing context into Model.forecast v2 #888
Conversation
|
🚀 Deployed on https://deploy-preview-888--etna-docs.netlify.app |
Codecov Report
@@ Coverage Diff @@
## inference #888 +/- ##
============================================
Coverage ? 49.58%
============================================
Files ? 130
Lines ? 7501
Branches ? 0
============================================
Hits ? 3719
Misses ? 3782
Partials ? 0 📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more |
| self.model = cast(ContextRequiredModelType, self.model) | ||
| if isinstance(self.model, DeepBaseModel): | ||
| current_ts_forecast = current_ts.make_future( | ||
| future_steps=self.model.decoder_length, tail_steps=self.model.encoder_length |
There was a problem hiding this comment.
self.model.context_size
There was a problem hiding this comment.
For DeepBaseModel they are the same.
There was a problem hiding this comment.
Why do we use extra branch in this case?
There was a problem hiding this comment.
Because of self.model.decoder_length.
There was a problem hiding this comment.
Ok, we should fix it in the future
|
|
||
| if isinstance(self.model, get_args(ContextRequiredModelType)): | ||
| self.model = cast(ContextRequiredModelType, self.model) | ||
| if isinstance(self.model, DeepBaseModel): |
There was a problem hiding this comment.
Nested ifs are not readable. Better to use plain if cases.
Before submitting (must do checklist)
Proposed Changes
Closing issues
Closes #845.