Skip to main content

Setting Trace Tags

Tags are mutable key-value pairs that you can attach to traces to add valuable labels and context for grouping and filtering traces. For example, you can tag traces based on the topic of the user's input or the type of request being processed and group them together for analysis and quality evaluation.

Setting Tags via the MLflow UI​

Setting Tags on Ongoing Traces​

Use mlflow.update_current_trace() to add tags during trace execution.

python
import mlflow


@mlflow.trace
def my_func(x):
    mlflow.update_current_trace(tags={"fruit": "apple"})
    return x + 1


result = my_func(5)
note

If the key is already present, the mlflow.update_current_trace() function will update the key with the new value.

Setting Tags on Completed Traces​

Add or modify tags on traces that have already been completed and logged.

python
import mlflow

mlflow.set_trace_tag(trace_id="your-trace-id", key="tag_key", value="tag_value")

Retrieving Tags​

Tags are stored on the info.tags attribute of the trace object.

python
import mlflow

trace = mlflow.get_trace(trace_id="your-trace-id")
print(trace.info.tags)
# Output: {'tag_key': 'tag_value'}

Searching and Filtering with Tags​

Use tags to find specific traces quickly and efficiently.

python
# Search for traces with tag 'environment' set to 'production'
traces = mlflow.search_traces(filter_string="tags.environment = 'production'")

You can also use pattern matching to find traces by tag value.

python
# Search for traces with tag that contains the word 'mlflow'
traces = mlflow.search_traces(filter_string="tags.topic LIKE '%mlflow%'")

View the full list of supported filter syntax in the Search Traces guide.