Move user to tag

docs/source/rest-api.rst

@@ -353,6 +353,8 @@ Request Structure
 | experiment_id    | ``INT64``                       | ID of the associated experiment.                                                               |
 +------------------+---------------------------------+------------------------------------------------------------------------------------------------+
 | user_id          | ``STRING``                      | ID of the user executing the run.                                                              |
+|                  |                                 | This field is deprecated and will be removed in a later MLflow version. Use the                |
+|                  |                                 | ``mlflow.user`` run tag instead.                                                               |
 +------------------+---------------------------------+------------------------------------------------------------------------------------------------+
 | run_name         | ``STRING``                      | Human readable name for the run.                                                               |
 |                  |                                 | This field is deprecated and will be removed in MLflow 1.0. Use the ``mlflow.runName`` run tag |
@@ -1279,6 +1281,8 @@ Metadata of a single run.
 |                  |                         | tag instead.                                                                                   |
 +------------------+-------------------------+------------------------------------------------------------------------------------------------+
 | user_id          | ``STRING``              | User who initiated the run.                                                                    |
+|                  |                         | This field is deprecated and will be removed in a later MLflow version. Use the                |
+|                  |                         | ``mlflow.user`` run tag instead.                                                               |
 +------------------+-------------------------+------------------------------------------------------------------------------------------------+
 | status           | :ref:`mlflowrunstatus`  | Current status of the run.                                                                     |
 +------------------+-------------------------+------------------------------------------------------------------------------------------------+

examples/rest_api/mlflow_tracking_rest_api.py

@@ -26,6 +26,7 @@ def __init__(self, hostname, port, experiment_id):
 	def create_run(self):
 		"""Create a new run for tracking."""
 		url = self.base_url + '/runs/create'
+		# user_id is deprecated and will be removed from the API in a future release  
 		payload = {'experiment_id': self.experiment_id, 'start_time': int(time.time() * 1000), 'user_id': _get_user_id()}
 		r = requests.post(url, json=payload)
 		run_id = None

mlflow/R/mlflow/R/tracking-runs.R

@@ -26,18 +26,18 @@ mlflow_log_metric <- function(key, value, timestamp = NULL, run_id = NULL, clien
   invisible(value)
 }
 
+mlflow_create_run <- function(start_time = NULL, tags = NULL, experiment_id = NULL, client) {
+  experiment_id <- resolve_experiment_id(experiment_id)
 
+  # Read user_id from tags
+  # user_id is deprecated and will be removed from a future release
+  user_id <- tags[[MLFLOW_TAGS$MLFLOW_USER]] %||% "unknown"
 
-
-mlflow_create_run <- function(user_id = NULL, start_time = NULL, tags = NULL,
-                              experiment_id = NULL, client) {
-  experiment_id <- resolve_experiment_id(experiment_id)
   tags <- if (!is.null(tags)) tags %>%
     purrr::imap(~ list(key = .y, value = .x)) %>%
     unname()
 
   start_time <- start_time %||% current_time()
-  user_id <- user_id %||% mlflow_user()
 
   response <- mlflow_rest(
     "runs", "create",
@@ -310,7 +310,6 @@ mlflow_set_terminated <- function(status, end_time, run_id, client) {
   mlflow_get_run(client = client, run_id = response$run_info$run_uuid)
 }
 
-
 #' Download Artifacts
 #'
 #' Download an artifact file or directory from a run to a local directory if applicable,
@@ -341,7 +340,6 @@ mlflow_download_artifacts <- function(path, run_id = NULL, client = NULL) {
   gsub("\n", "", result$stdout)
 }
 
-
 # ' Download Artifacts from URI.
 mlflow_download_artifacts_from_uri <- function(artifact_uri, client = mlflow_client()) {
   result <- mlflow_cli("artifacts", "download-from-uri", "-a", artifact_uri, echo = FALSE,
@@ -444,12 +442,11 @@ mlflow_log_artifact <- function(path, artifact_path = NULL, run_id = NULL, clien
   mlflow_list_artifacts(run_id = run_id, path = artifact_path, client = client)
 }
 
-
 #' Start Run
 #'
 #' Starts a new run. If `client` is not provided, this function infers contextual information such as
 #'   source name and version, and also registers the created run as the active run. If `client` is provided,
-#'   no inference is done, and additional arguments such as `user_id` and `start_time` can be provided.
+#'   no inference is done, and additional arguments such as `start_time` can be provided.
 #'
 #' @param run_id If specified, get the run with the specified UUID and log metrics
 #'   and params under that run. The run's end time is unset and its status is set to
@@ -462,7 +459,6 @@ mlflow_log_artifact <- function(path, artifact_path = NULL, run_id = NULL, clien
 #' @param source_version Optional Git commit hash to associate with the run.
 #' @param entry_point_name Optional name of the entry point for to the current run.
 #' @param source_type Integer enum value describing the type of the run  ("local", "project", etc.).
-#' @param user_id User ID or LDAP for the user executing the run. Only used when `client` is specified.
 #' @param start_time Unix timestamp of when the run started in milliseconds. Only used when `client` is specified.
 #' @param tags Additional metadata for run in key-value pairs. Only used when `client` is specified.
 #' @template roxlate-client
@@ -475,21 +471,19 @@ mlflow_log_artifact <- function(path, artifact_path = NULL, run_id = NULL, clien
 #' }
 #'
 #' @export
-mlflow_start_run <- function(run_id = NULL, experiment_id = NULL, user_id = NULL,
-                             start_time = NULL, tags = NULL, client = NULL) {
+mlflow_start_run <- function(run_id = NULL, experiment_id = NULL, start_time = NULL, tags = NULL, client = NULL) {
 
   # When `client` is provided, this function acts as a wrapper for `runs/create` and does not register
   #  an active run.
   if (!is.null(client)) {
     if (!is.null(run_id)) stop("`run_id` should not be specified when `client` is specified.", call. = FALSE)
-    run <- mlflow_create_run(client = client, user_id = user_id, start_time = start_time,
+    run <- mlflow_create_run(client = client, start_time = start_time,
                              tags = tags, experiment_id = experiment_id)
     return(run)
   }
 
   # Fluent mode, check to see if extraneous params passed.
 
-  if (!is.null(user_id)) stop("`user_id` should only be specified when `client` is specified.", call. = FALSE)
   if (!is.null(start_time)) stop("`start_time` should only be specified when `client` is specified.", call. = FALSE)
   if (!is.null(tags)) stop("`tags` should only be specified when `client` is specified.", call. = FALSE)
 
@@ -524,13 +518,13 @@ mlflow_start_run <- function(run_id = NULL, experiment_id = NULL, user_id = NULL
   run
 }
 
-
 mlflow_get_run_context <- function(client, ...) {
   UseMethod("mlflow_get_run_context")
 }
 
 mlflow_get_run_context.default <- function(client, experiment_id, ...) {
   tags <- list()
+  tags[[MLFLOW_TAGS$MLFLOW_USER]] <- mlflow_user()
   tags[[MLFLOW_TAGS$MLFLOW_SOURCE_NAME]] <- get_source_name()
   tags[[MLFLOW_TAGS$MLFLOW_SOURCE_VERSION]] <- get_source_version()
   tags[[MLFLOW_TAGS$MLFLOW_SOURCE_TYPE]] <- MLFLOW_SOURCE_TYPE$LOCAL
@@ -580,6 +574,7 @@ mlflow_end_run <- function(status = c("FINISHED", "SCHEDULED", "FAILED", "KILLED
 }
 
 MLFLOW_TAGS <- list(
+  MLFLOW_USER = "mlflow.user",
   MLFLOW_SOURCE_NAME = "mlflow.source.name",
   MLFLOW_SOURCE_VERSION = "mlflow.source.version",
   MLFLOW_SOURCE_TYPE = "mlflow.source.type"

mlflow/R/mlflow/tests/testthat/test-tracking-runs.R

@@ -6,8 +6,7 @@ test_that("mlflow_start_run()/mlflow_get_run() work properly", {
   run <- mlflow_start_run(
     client = client,
     experiment_id = "0",
-    user_id = "user1",
-    tags = list(foo = "bar", foz = "baz")
+    tags = list(foo = "bar", foz = "baz", mlflow.user = "user1")
   )
 
   run <- mlflow_get_run(client = client, run$run_uuid)
@@ -16,7 +15,11 @@ test_that("mlflow_start_run()/mlflow_get_run() work properly", {
 
   expect_true(
     all(purrr::transpose(run$tags[[1]]) %in%
-          list(list(key = "foz", value = "baz"), list(key = "foo", value = "bar"))
+      list(
+        list(key = "foz", value = "baz"),
+        list(key = "foo", value = "bar"),
+        list(key = "mlflow.user", value = "user1")
+      )
     )
   )
 })
@@ -72,7 +75,6 @@ test_that("logging functionality", {
   expect_true(all(difftime(metric_history$timestamp, run_start_time) >= 0))
   expect_true(all(difftime(metric_history$timestamp, run_end_time) <= 0))
 
-
   expect_error(
     mlflow_get_run(),
     "`run_id` must be specified when there is no active run\\."

mlflow/java/client/src/main/java/org/mlflow/tracking/MlflowClient.java

@@ -84,6 +84,8 @@ public RunInfo createRun(String experimentId) {
     CreateRun.Builder request = CreateRun.newBuilder();
     request.setExperimentId(experimentId);
     request.setStartTime(System.currentTimeMillis());
+    // userId is deprecated and will be removed in a future release.
+    // It should be set as the `mlflow.user` tag instead.
     String username = System.getProperty("user.name");
     if (username != null) {
       request.setUserId(System.getProperty("user.name"));
@@ -432,7 +434,6 @@ public void logArtifacts(String runId, File localDir) {
     getArtifactRepository(runId).logArtifacts(localDir);
   }
 
-
   /**
    * Uploads all files within the given local director an artifactPath within the run's root
    * artifact directory. For example, if /my/local/dir/ contains two files "file1" and "file2", then