![]() ![]() Section of the model’s conda environment ( conda.yaml) file.Įxperimental: This parameter may change or be removed in a future Both requirements andĬonstraints are automatically parsed and written to requirements.txt and constraints.txtįiles, respectively, and stored as part of the model. Generated automatically based on the user’s current software environment. Written to the pip section of the model’s conda environment ( conda.yaml) file.Įither an iterable of pip requirement stringsĭescribes additional pip requirements that are appended to a default set of pip requirements If the requirement inference fails, it falls back to using get_default_pip_requirements().īoth requirements and constraints are automatically parsed and written to requirements.txt andĬonstraints.txt files, respectively, and stored as part of the model. Is inferred by _pip_requirements() from the current software environment. If provided, thisĭescribes the environment this model should be run in. ) or the string path toĪ pip requirements file on the local filesystem (e.g. Pip_requirements – Either an iterable of pip requirement strings Bytes areĪwait_registration_for – Number of seconds to wait for the model version to finishīeing created and is in READY status. Serialized to json using the Pandas split-oriented format. The given example will be converted to a Pandas DataFrame and then The example can be used as a hint of what data to feed the Input_example – Input example provides one or several instances of valid # compute model predictions signature = infer_signature ( train, predictions ) drop_column ( "target_label" ) predictions =. model predictions generated onįrom import infer_signature train = df. the training dataset with targetĬolumn omitted) and valid model output (e.g. In some cases not all arrays will be set to None.įrom datasets with valid model input (e.g. If True, then all arrays with length nobs are set to None before If False (default), then the instance is pickled without changes. Registered_model_name, also creating a registered model if one Registered_model_name – If given, create a model version under The following is an example dictionary representation of a conda environment:Ĭode_paths – A list of local filesystem paths to Python file dependencies (or directoriesĬontaining file dependencies). Requirements.txt file and the full conda environment is written to conda.yaml. pip requirements from conda_env are written to a pip If the requirement inference fails, it falls back to using If None, a condaĮnvironment with pip requirements inferred by _pip_requirements() is added Should specify the dependencies contained in get_default_conda_env(). ![]() If provided, this describes the environment this model should be run in. Statsmodels_model – statsmodels model (an instance of )Īrtifact_path – Run-relative artifact path.Įither a dictionary representation of a Conda environment or the path to a conda environment yamlįile. Log a statsmodels model as an MLflow artifact for the current run. log_model ( statsmodels_model, artifact_path, conda_env = None, code_paths = None, registered_model_name = None, remove_data : bool = False, signature : = None, input_example : Union = None, await_registration_for = 300, pip_requirements = None, extra_pip_requirements = None, metadata = None, ** kwargs ) ![]() If unspecified, a local outputĪ statsmodels model (an instance of ). Autologging is known to be compatible with the following package versions: 0.11.1 /run-relative/path/to/modelįor more information about supported URI schemes, seeĭst_path – The local filesystem path to which to download the model artifact.
0 Comments
Leave a Reply. |