Releases: openlayer-ai/openlayer-python
Releases · openlayer-ai/openlayer-python
v0.3.0a0
Added
- A
Project
helper class. - A convenience method
create_or_load_project
which loads in a project in if it is already created.
Changed
- Models and datasets must be added to projects.
- Deprecates
categorical_features_map
in favor ofcategorical_feature_names
for model and dataset uploads. - Moved
TaskType
attribute from theModel
level to theProject
level. Creating aProject
now requires specifying theTaskType
. - Removed
name
fromadd_dataset
. - Changed
description
tocommit_message
fromadd_dataset
,add_dataframe
andadd_model
.
v0.2.0a1
Fixed
- Fail early if
custom_model_code
,dependent_dir
orrequirements_txt_file
areNone
when model type isModelType.custom
. - Fail early if
model
is notNone
when model type isModelType.custom
.
v0.2.0a0
Added
- Accepts AZURE as a
DeploymentType
.
Changed
- Compatibility with Unbox backend storage and data refactor.
v0.1.2
Fixed
- Default Unbox server URL (https://api.unbox.ai/).
v0.1.1
Added
- Can specify
ModelType.keras
. - API documentation.
Changed
- Parameter ordering for
add_model
,add_dataset
,add_dataframe
for clarity's sake.
Fixed
- Fix bug when predict function is a list with numpy objects.
- Better error message if return type isn't a list or numpy array.
- Prevents dataset upload when
label_column_name
is also infeature_names
.