Edit on GitHub
Live.log_params()
def log_params(params: Dict[ParamLike]):
Also see Live.log_param()
.
Usage
from dvclive import Live
live = Live()
params = {
"num_classes": 10,
"metrics": ["accuracy", "mae"],
"training": {
"epochs": 42,
"batch_size": 128,
"optimizer": {
"name": "adam",
"loss": "categorical_crossentropy",
"learning_rate": 1e-3,
},
},
}
live.log_params(params)
Description
On each Live.log_params(params)
call, DVCLive will write keys/values pairs in
the params
dict to {Live.dir}/params.yaml
:
lr: 1e-3
n_iter: 50
The logged params can be visualized with dvc params
:
dvc params diff dvclive/params.yaml
If you use DVC pipelines, parameter dependencies are tracked automatically, and you can skip logging them with DVCLive.
Parameters
params
- dictionary with name/value pairs of parameters to be logged.
Exceptions
-
dvclive.error.InvalidParameterType
- thrown if any of the parameter values in theparams
dict is not among supported types. Supported types include:ParamLike = Union[ int, float, str, bool, List["ParamLike"], Dict[str, "ParamLike"] ]