Skip to content

sklab.adapters.logging

sklab.adapters.logging.LoggerProtocol

Bases: Protocol

Context-managed logger for experiment tracking.

The logger itself provides logging methods. start_run() is a context manager that yields self, so usage is:

with logger.start_run(name="exp-1") as run:
    run.log_params({"lr": 0.01})
    run.log_metrics({"accuracy": 0.95})
Source code in src/sklab/adapters/logging.py 
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
@runtime_checkable
class LoggerProtocol(Protocol):
    """Context-managed logger for experiment tracking.

    The logger itself provides logging methods. `start_run()` is a context
    manager that yields `self`, so usage is:

        with logger.start_run(name="exp-1") as run:
            run.log_params({"lr": 0.01})
            run.log_metrics({"accuracy": 0.95})
    """

    def start_run(
        self,
        name: str | None = None,
        config: Params | None = None,
        tags: Tags | None = None,
        nested: bool = False,
    ) -> AbstractContextManager[Self]: ...

    def log_params(self, params: Params) -> None: ...

    def log_metrics(self, metrics: Metrics, step: int | None = None) -> None: ...

    def set_tags(self, tags: Tags) -> None: ...

    def log_artifact(self, path: str, name: str | None = None) -> None: ...

    def log_model(self, model: Any, name: str | None = None) -> None: ...