mlrun 1.7.0rc4__py3-none-any.whl → 1.7.2__py3-none-any.whl

mlrun/utils/notifications/notification_pusher.py

@@ -15,10 +15,17 @@
 import asyncio
 import datetime
 import os
+import re
 import traceback
 import typing
 from concurrent.futures import ThreadPoolExecutor
 
+import mlrun_pipelines.common.ops
+import mlrun_pipelines.models
+import mlrun_pipelines.utils
+
+import mlrun.common.constants as mlrun_constants
+import mlrun.common.runtimes.constants
 import mlrun.common.schemas
 import mlrun.config
 import mlrun.db.base
@@ -161,6 +168,11 @@ class NotificationPusher(_NotificationPusherBase):
                     logger.warning(
                         "Failed to push notification async",
                         error=mlrun.errors.err_to_str(result),
+                        traceback=traceback.format_exception(
+                            etype=type(result),
+                            value=result,
+                            tb=result.__traceback__,
+                        ),
                     )
 
         logger.debug(
@@ -233,25 +245,12 @@ class NotificationPusher(_NotificationPusherBase):
         resource = "Run"
         runs = [run.to_dict()]
 
-        if "workflow" in run.metadata.labels:
-            resource = "Workflow"
+        if mlrun_constants.MLRunInternalLabels.workflow in run.metadata.labels:
+            resource = mlrun_constants.MLRunInternalLabels.workflow
             custom_message = (
                 f" (workflow: {run.metadata.labels['workflow']}){custom_message}"
             )
-            db = mlrun.get_run_db()
-
-            workflow_id = run.status.results.get("workflow_id", None)
-            if workflow_id:
-                workflow_runs = db.list_runs(
-                    project=run.metadata.project,
-                    labels=f"workflow={workflow_id}",
-                )
-                logger.debug(
-                    "Found workflow runs, extending notification runs",
-                    workflow_id=workflow_id,
-                    workflow_runs_amount=len(workflow_runs),
-                )
-                runs.extend(workflow_runs)
+            runs.extend(self.get_workflow_steps(run))
 
         message = (
             self.messages.get(run.state(), "").format(resource=resource)
@@ -395,6 +394,131 @@ class NotificationPusher(_NotificationPusherBase):
             mask_params=False,
         )
 
+    def get_workflow_steps(self, run: mlrun.model.RunObject) -> list:
+        steps = []
+        db = mlrun.get_run_db()
+
+        def _add_run_step(_step: mlrun_pipelines.models.PipelineStep):
+            try:
+                _run = db.list_runs(
+                    project=run.metadata.project,
+                    labels=f"{mlrun_constants.MLRunInternalLabels.runner_pod}={_step.node_name}",
+                )[0]
+            except IndexError:
+                _run = {
+                    "metadata": {
+                        "name": _step.display_name,
+                        "project": run.metadata.project,
+                    },
+                }
+            _run["step_kind"] = _step.step_type
+            if _step.skipped:
+                _run.setdefault("status", {})["state"] = (
+                    mlrun.common.runtimes.constants.RunStates.skipped
+                )
+            steps.append(_run)
+
+        def _add_deploy_function_step(_step: mlrun_pipelines.models.PipelineStep):
+            project, name, hash_key = self._extract_function_uri(
+                _step.get_annotation("mlrun/function-uri")
+            )
+            if name:
+                try:
+                    function = db.get_function(
+                        project=project, name=name, hash_key=hash_key
+                    )
+                except mlrun.errors.MLRunNotFoundError:
+                    # If the function is not found (if build failed for example), we will create a dummy
+                    # function object for the notification to display the function name
+                    function = {
+                        "metadata": {
+                            "name": name,
+                            "project": project,
+                            "hash_key": hash_key,
+                        },
+                    }
+                pod_phase = _step.phase
+                if _step.skipped:
+                    state = mlrun.common.schemas.FunctionState.skipped
+                else:
+                    state = mlrun.common.runtimes.constants.PodPhases.pod_phase_to_run_state(
+                        pod_phase
+                    )
+                function["status"] = {"state": state}
+                if isinstance(function["metadata"].get("updated"), datetime.datetime):
+                    function["metadata"]["updated"] = function["metadata"][
+                        "updated"
+                    ].isoformat()
+                function["step_kind"] = _step.step_type
+                steps.append(function)
+
+        step_methods = {
+            mlrun_pipelines.common.ops.PipelineRunType.run: _add_run_step,
+            mlrun_pipelines.common.ops.PipelineRunType.build: _add_deploy_function_step,
+            mlrun_pipelines.common.ops.PipelineRunType.deploy: _add_deploy_function_step,
+        }
+
+        workflow_id = run.status.results.get("workflow_id", None)
+        if not workflow_id:
+            return steps
+
+        workflow_manifest = self._get_workflow_manifest(workflow_id)
+        if not workflow_manifest:
+            return steps
+
+        try:
+            for step in workflow_manifest.get_steps():
+                step_method = step_methods.get(step.step_type)
+                if step_method:
+                    step_method(step)
+            return steps
+        except Exception:
+            # If we fail to read the pipeline steps, we will return the list of runs that have the same workflow id
+            logger.warning(
+                "Failed to extract workflow steps from workflow manifest, "
+                "returning all runs with the workflow id label",
+                workflow_id=workflow_id,
+                traceback=traceback.format_exc(),
+            )
+            return db.list_runs(
+                project=run.metadata.project,
+                labels=f"workflow={workflow_id}",
+            )
+
+    @staticmethod
+    def _get_workflow_manifest(
+        workflow_id: str,
+    ) -> typing.Optional[mlrun_pipelines.models.PipelineManifest]:
+        kfp_client = mlrun_pipelines.utils.get_client(mlrun.mlconf.kfp_url)
+
+        # arbitrary timeout of 5 seconds, the workflow should be done by now
+        kfp_run = kfp_client.wait_for_run_completion(workflow_id, 5)
+        if not kfp_run:
+            return None
+
+        kfp_run = mlrun_pipelines.models.PipelineRun(kfp_run)
+        return kfp_run.workflow_manifest()
+
+    def _extract_function_uri(self, function_uri: str) -> tuple[str, str, str]:
+        """
+        Extract the project, name, and hash key from a function uri.
+        Examples:
+            - "project/name@hash_key" returns project, name, hash_key
+            - "project/name returns" project, name, ""
+        """
+        project, name, hash_key = None, None, None
+        hashed_pattern = r"^(.+)/(.+)@(.+)$"
+        pattern = r"^(.+)/(.+)$"
+        match = re.match(hashed_pattern, function_uri)
+        if match:
+            project, name, hash_key = match.groups()
+        else:
+            match = re.match(pattern, function_uri)
+            if match:
+                project, name = match.groups()
+                hash_key = ""
+        return project, name, hash_key
+
 
 class CustomNotificationPusher(_NotificationPusherBase):
     def __init__(self, notification_types: list[str] = None):
@@ -413,6 +537,12 @@ class CustomNotificationPusher(_NotificationPusherBase):
             if notification.is_async
         }
 
+    @property
+    def notifications(self):
+        notifications = self._sync_notifications.copy()
+        notifications.update(self._async_notifications)
+        return notifications
+
     def push(
         self,
         message: str,

mlrun/utils/regex.py

@@ -92,3 +92,12 @@ artifact_key = [r"[^\/\\]+$"]
 # must be alphanumeric or _
 # max 256 length
 v3io_stream_consumer_group = [r"^(?!_)[a-zA-Z0-9_]{1,256}$"]
+
+# URI patterns
+run_uri_pattern = r"^(?P<project>.*)@(?P<uid>.*)\#(?P<iteration>.*?)(:(?P<tag>.*))?$"
+
+artifact_uri_pattern = r"^((?P<project>.*)/)?(?P<key>.*?)(\#(?P<iteration>.*?))?(:(?P<tag>.*?))?(@(?P<tree>.*))?$"
+
+artifact_producer_uri_pattern = (
+    r"^((?P<project>.*)/)?(?P<uid>.*?)(\-(?P<iteration>.*?))?$"
+)

mlrun/utils/retryer.py

@@ -117,7 +117,7 @@ class Retryer:
         self._raise_last_exception()
 
     def _prepare(self):
-        self.start_time = time.time()
+        self.start_time = time.monotonic()
         self.last_exception = None
 
         # Check if backoff is just a simple interval
@@ -138,6 +138,7 @@ class Retryer:
         except mlrun.errors.MLRunFatalFailureError as exc:
             raise exc.original_exception
         except Exception as exc:
+            self.last_exception = exc
             return (
                 None,
                 self.last_exception,
@@ -172,7 +173,7 @@ class Retryer:
         ) from self.last_exception
 
     def _timeout_exceeded(self, next_interval=None):
-        now = time.time()
+        now = time.monotonic()
         if next_interval:
             now = now + next_interval
         return self.timeout is not None and now >= self.start_time + self.timeout

mlrun/utils/v3io_clients.py

@@ -11,10 +11,9 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-#
 
 from v3io.dataplane import Client as V3IOClient
-from v3io_frames import Client as get_client
+from v3io_frames import Client as V3IOFramesClient
 from v3io_frames.client import ClientBase
 
 _v3io_clients: dict[frozenset, V3IOClient] = {}
@@ -25,7 +24,7 @@ def get_frames_client(**kwargs) -> ClientBase:
     global _frames_clients
     kw_set = frozenset(kwargs.items())
     if kw_set not in _frames_clients:
-        _frames_clients[kw_set] = get_client(**kwargs)
+        _frames_clients[kw_set] = V3IOFramesClient(**kwargs)
 
     return _frames_clients[kw_set]
 

mlrun/utils/version/version.json

@@ -1,4 +1,4 @@
 {
-  "git_commit": "cb2750f25e202a321723af3465359944445dfda7",
-  "version": "1.7.0-rc4"
+  "git_commit": "bb224abaed1b5d221269586c9497124a526615e8",
+  "version": "1.7.2"
 }

mlrun-1.7.2.dist-info/METADATA

@@ -0,0 +1,390 @@
+Metadata-Version: 2.2
+Name: mlrun
+Version: 1.7.2
+Summary: Tracking and config of machine learning runs
+Home-page: https://github.com/mlrun/mlrun
+Author: Yaron Haviv
+Author-email: yaronh@iguazio.com
+License: Apache License 2.0
+Keywords: mlrun,mlops,data-science,machine-learning,experiment-tracking
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Operating System :: POSIX :: Linux
+Classifier: Operating System :: Microsoft :: Windows
+Classifier: Operating System :: MacOS
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Classifier: Topic :: Software Development :: Libraries
+Requires-Python: >=3.9, <3.12
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: urllib3<1.27,>=1.26.9
+Requires-Dist: GitPython>=3.1.41,~=3.1
+Requires-Dist: aiohttp<3.11,>=3.9
+Requires-Dist: aiohttp-retry~=2.8.0
+Requires-Dist: click~=8.1
+Requires-Dist: nest-asyncio~=1.0
+Requires-Dist: ipython~=8.10
+Requires-Dist: nuclio-jupyter~=0.10.4
+Requires-Dist: numpy<1.27.0,>=1.16.5
+Requires-Dist: pandas<2.2,>=1.2
+Requires-Dist: pyarrow<18,>=10.0
+Requires-Dist: pyyaml<7,>=5.4.1
+Requires-Dist: requests~=2.32
+Requires-Dist: tabulate~=0.8.6
+Requires-Dist: v3io~=0.6.9
+Requires-Dist: pydantic<1.10.15,>=1.10.8
+Requires-Dist: mergedeep~=1.3
+Requires-Dist: v3io-frames~=0.10.14
+Requires-Dist: semver~=3.0
+Requires-Dist: dependency-injector~=4.41
+Requires-Dist: fsspec<2024.7,>=2023.9.2
+Requires-Dist: v3iofs~=0.1.17
+Requires-Dist: storey~=1.7.50
+Requires-Dist: inflection~=0.5.0
+Requires-Dist: python-dotenv~=0.17.0
+Requires-Dist: setuptools~=71.0
+Requires-Dist: deprecated~=1.2
+Requires-Dist: jinja2>=3.1.3,~=3.1
+Requires-Dist: orjson<4,>=3.9.15
+Requires-Dist: mlrun-pipelines-kfp-common~=0.1.9
+Requires-Dist: mlrun-pipelines-kfp-v1-8~=0.1.6
+Provides-Extra: s3
+Requires-Dist: boto3<1.36,>=1.28.0; extra == "s3"
+Requires-Dist: aiobotocore<2.16,>=2.5.0; extra == "s3"
+Requires-Dist: s3fs<2024.7,>=2023.9.2; extra == "s3"
+Provides-Extra: azure-blob-storage
+Requires-Dist: msrest~=0.6.21; extra == "azure-blob-storage"
+Requires-Dist: azure-core~=1.24; extra == "azure-blob-storage"
+Requires-Dist: adlfs==2023.9.0; extra == "azure-blob-storage"
+Requires-Dist: pyopenssl>=23; extra == "azure-blob-storage"
+Provides-Extra: azure-key-vault
+Requires-Dist: azure-identity~=1.5; extra == "azure-key-vault"
+Requires-Dist: azure-keyvault-secrets~=4.2; extra == "azure-key-vault"
+Requires-Dist: pyopenssl>=23; extra == "azure-key-vault"
+Provides-Extra: bokeh
+Requires-Dist: bokeh>=2.4.2,~=2.4; extra == "bokeh"
+Provides-Extra: plotly
+Requires-Dist: plotly~=5.23; extra == "plotly"
+Provides-Extra: graphviz
+Requires-Dist: graphviz~=0.20.0; extra == "graphviz"
+Provides-Extra: google-cloud
+Requires-Dist: google-cloud-storage==2.14.0; extra == "google-cloud"
+Requires-Dist: google-cloud-bigquery[bqstorage,pandas]==3.14.1; extra == "google-cloud"
+Requires-Dist: google-cloud-bigquery-storage~=2.17; extra == "google-cloud"
+Requires-Dist: google-cloud==0.34; extra == "google-cloud"
+Requires-Dist: gcsfs<2024.7,>=2023.9.2; extra == "google-cloud"
+Provides-Extra: kafka
+Requires-Dist: kafka-python~=2.0; extra == "kafka"
+Requires-Dist: avro~=1.11; extra == "kafka"
+Provides-Extra: redis
+Requires-Dist: redis~=4.3; extra == "redis"
+Provides-Extra: mlflow
+Requires-Dist: mlflow~=2.8; extra == "mlflow"
+Provides-Extra: databricks-sdk
+Requires-Dist: databricks-sdk~=0.13.0; extra == "databricks-sdk"
+Provides-Extra: sqlalchemy
+Requires-Dist: sqlalchemy~=1.4; extra == "sqlalchemy"
+Provides-Extra: dask
+Requires-Dist: dask~=2023.12.1; extra == "dask"
+Requires-Dist: distributed~=2023.12.1; extra == "dask"
+Provides-Extra: alibaba-oss
+Requires-Dist: ossfs==2023.12.0; extra == "alibaba-oss"
+Requires-Dist: oss2==2.18.1; extra == "alibaba-oss"
+Provides-Extra: tdengine
+Requires-Dist: taos-ws-py==0.3.2; extra == "tdengine"
+Requires-Dist: taoswswrap~=0.2.0; extra == "tdengine"
+Provides-Extra: snowflake
+Requires-Dist: snowflake-connector-python~=3.7; extra == "snowflake"
+Provides-Extra: api
+Requires-Dist: uvicorn~=0.27.1; extra == "api"
+Requires-Dist: dask-kubernetes~=0.11.0; extra == "api"
+Requires-Dist: apscheduler==3.10.3; extra == "api"
+Requires-Dist: objgraph~=3.6; extra == "api"
+Requires-Dist: igz-mgmt~=0.2.0; extra == "api"
+Requires-Dist: humanfriendly~=10.0; extra == "api"
+Requires-Dist: fastapi~=0.110.0; extra == "api"
+Requires-Dist: sqlalchemy~=1.4; extra == "api"
+Requires-Dist: pymysql~=1.0; extra == "api"
+Requires-Dist: alembic~=1.9; extra == "api"
+Requires-Dist: timelength~=1.1; extra == "api"
+Requires-Dist: memray~=1.12; sys_platform != "win32" and extra == "api"
+Provides-Extra: all
+Requires-Dist: adlfs==2023.9.0; extra == "all"
+Requires-Dist: aiobotocore<2.16,>=2.5.0; extra == "all"
+Requires-Dist: avro~=1.11; extra == "all"
+Requires-Dist: azure-core~=1.24; extra == "all"
+Requires-Dist: azure-identity~=1.5; extra == "all"
+Requires-Dist: azure-keyvault-secrets~=4.2; extra == "all"
+Requires-Dist: bokeh>=2.4.2,~=2.4; extra == "all"
+Requires-Dist: boto3<1.36,>=1.28.0; extra == "all"
+Requires-Dist: dask~=2023.12.1; extra == "all"
+Requires-Dist: databricks-sdk~=0.13.0; extra == "all"
+Requires-Dist: distributed~=2023.12.1; extra == "all"
+Requires-Dist: gcsfs<2024.7,>=2023.9.2; extra == "all"
+Requires-Dist: google-cloud-bigquery-storage~=2.17; extra == "all"
+Requires-Dist: google-cloud-bigquery[bqstorage,pandas]==3.14.1; extra == "all"
+Requires-Dist: google-cloud-storage==2.14.0; extra == "all"
+Requires-Dist: google-cloud==0.34; extra == "all"
+Requires-Dist: graphviz~=0.20.0; extra == "all"
+Requires-Dist: kafka-python~=2.0; extra == "all"
+Requires-Dist: mlflow~=2.8; extra == "all"
+Requires-Dist: msrest~=0.6.21; extra == "all"
+Requires-Dist: oss2==2.18.1; extra == "all"
+Requires-Dist: ossfs==2023.12.0; extra == "all"
+Requires-Dist: plotly~=5.23; extra == "all"
+Requires-Dist: pyopenssl>=23; extra == "all"
+Requires-Dist: redis~=4.3; extra == "all"
+Requires-Dist: s3fs<2024.7,>=2023.9.2; extra == "all"
+Requires-Dist: snowflake-connector-python~=3.7; extra == "all"
+Requires-Dist: sqlalchemy~=1.4; extra == "all"
+Requires-Dist: taos-ws-py==0.3.2; extra == "all"
+Requires-Dist: taoswswrap~=0.2.0; extra == "all"
+Provides-Extra: complete
+Requires-Dist: adlfs==2023.9.0; extra == "complete"
+Requires-Dist: aiobotocore<2.16,>=2.5.0; extra == "complete"
+Requires-Dist: avro~=1.11; extra == "complete"
+Requires-Dist: azure-core~=1.24; extra == "complete"
+Requires-Dist: azure-identity~=1.5; extra == "complete"
+Requires-Dist: azure-keyvault-secrets~=4.2; extra == "complete"
+Requires-Dist: boto3<1.36,>=1.28.0; extra == "complete"
+Requires-Dist: dask~=2023.12.1; extra == "complete"
+Requires-Dist: databricks-sdk~=0.13.0; extra == "complete"
+Requires-Dist: distributed~=2023.12.1; extra == "complete"
+Requires-Dist: gcsfs<2024.7,>=2023.9.2; extra == "complete"
+Requires-Dist: google-cloud-bigquery-storage~=2.17; extra == "complete"
+Requires-Dist: google-cloud-bigquery[bqstorage,pandas]==3.14.1; extra == "complete"
+Requires-Dist: google-cloud-storage==2.14.0; extra == "complete"
+Requires-Dist: google-cloud==0.34; extra == "complete"
+Requires-Dist: graphviz~=0.20.0; extra == "complete"
+Requires-Dist: kafka-python~=2.0; extra == "complete"
+Requires-Dist: mlflow~=2.8; extra == "complete"
+Requires-Dist: msrest~=0.6.21; extra == "complete"
+Requires-Dist: oss2==2.18.1; extra == "complete"
+Requires-Dist: ossfs==2023.12.0; extra == "complete"
+Requires-Dist: plotly~=5.23; extra == "complete"
+Requires-Dist: pyopenssl>=23; extra == "complete"
+Requires-Dist: redis~=4.3; extra == "complete"
+Requires-Dist: s3fs<2024.7,>=2023.9.2; extra == "complete"
+Requires-Dist: snowflake-connector-python~=3.7; extra == "complete"
+Requires-Dist: sqlalchemy~=1.4; extra == "complete"
+Requires-Dist: taos-ws-py==0.3.2; extra == "complete"
+Requires-Dist: taoswswrap~=0.2.0; extra == "complete"
+Provides-Extra: complete-api
+Requires-Dist: adlfs==2023.9.0; extra == "complete-api"
+Requires-Dist: aiobotocore<2.16,>=2.5.0; extra == "complete-api"
+Requires-Dist: alembic~=1.9; extra == "complete-api"
+Requires-Dist: apscheduler==3.10.3; extra == "complete-api"
+Requires-Dist: avro~=1.11; extra == "complete-api"
+Requires-Dist: azure-core~=1.24; extra == "complete-api"
+Requires-Dist: azure-identity~=1.5; extra == "complete-api"
+Requires-Dist: azure-keyvault-secrets~=4.2; extra == "complete-api"
+Requires-Dist: boto3<1.36,>=1.28.0; extra == "complete-api"
+Requires-Dist: dask-kubernetes~=0.11.0; extra == "complete-api"
+Requires-Dist: dask~=2023.12.1; extra == "complete-api"
+Requires-Dist: databricks-sdk~=0.13.0; extra == "complete-api"
+Requires-Dist: distributed~=2023.12.1; extra == "complete-api"
+Requires-Dist: fastapi~=0.110.0; extra == "complete-api"
+Requires-Dist: gcsfs<2024.7,>=2023.9.2; extra == "complete-api"
+Requires-Dist: google-cloud-bigquery-storage~=2.17; extra == "complete-api"
+Requires-Dist: google-cloud-bigquery[bqstorage,pandas]==3.14.1; extra == "complete-api"
+Requires-Dist: google-cloud-storage==2.14.0; extra == "complete-api"
+Requires-Dist: google-cloud==0.34; extra == "complete-api"
+Requires-Dist: graphviz~=0.20.0; extra == "complete-api"
+Requires-Dist: humanfriendly~=10.0; extra == "complete-api"
+Requires-Dist: igz-mgmt~=0.2.0; extra == "complete-api"
+Requires-Dist: kafka-python~=2.0; extra == "complete-api"
+Requires-Dist: memray~=1.12; sys_platform != "win32" and extra == "complete-api"
+Requires-Dist: mlflow~=2.8; extra == "complete-api"
+Requires-Dist: msrest~=0.6.21; extra == "complete-api"
+Requires-Dist: objgraph~=3.6; extra == "complete-api"
+Requires-Dist: oss2==2.18.1; extra == "complete-api"
+Requires-Dist: ossfs==2023.12.0; extra == "complete-api"
+Requires-Dist: plotly~=5.23; extra == "complete-api"
+Requires-Dist: pymysql~=1.0; extra == "complete-api"
+Requires-Dist: pyopenssl>=23; extra == "complete-api"
+Requires-Dist: redis~=4.3; extra == "complete-api"
+Requires-Dist: s3fs<2024.7,>=2023.9.2; extra == "complete-api"
+Requires-Dist: snowflake-connector-python~=3.7; extra == "complete-api"
+Requires-Dist: sqlalchemy~=1.4; extra == "complete-api"
+Requires-Dist: taos-ws-py==0.3.2; extra == "complete-api"
+Requires-Dist: taoswswrap~=0.2.0; extra == "complete-api"
+Requires-Dist: timelength~=1.1; extra == "complete-api"
+Requires-Dist: uvicorn~=0.27.1; extra == "complete-api"
+Dynamic: author
+Dynamic: author-email
+Dynamic: classifier
+Dynamic: description
+Dynamic: description-content-type
+Dynamic: home-page
+Dynamic: keywords
+Dynamic: license
+Dynamic: provides-extra
+Dynamic: requires-dist
+Dynamic: requires-python
+Dynamic: summary
+
+<a id="top"></a>
+[![Build Status](https://github.com/mlrun/mlrun/actions/workflows/build.yaml/badge.svg?branch=development)](https://github.com/mlrun/mlrun/actions/workflows/build.yaml?query=branch%3Adevelopment)
+[![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
+[![PyPI version fury.io](https://badge.fury.io/py/mlrun.svg)](https://pypi.python.org/pypi/mlrun/)
+[![Documentation](https://readthedocs.org/projects/mlrun/badge/?version=latest)](https://mlrun.readthedocs.io/en/latest/?badge=latest)
+[![Ruff](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/astral-sh/ruff/main/assets/badge/v2.json)](https://github.com/astral-sh/ruff)
+![GitHub commit activity](https://img.shields.io/github/commit-activity/w/mlrun/mlrun)
+![GitHub release (latest SemVer)](https://img.shields.io/github/v/release/mlrun/mlrun?sort=semver)
+[![Join MLOps Live](https://img.shields.io/badge/slack-join_chat-white.svg?logo=slack&style=social)](https://mlopslive.slack.com)
+
+<p align="left"><img src="https://github.com/mlrun/mlrun/raw/development/docs/_static/images/MLRun-logo.png" alt="MLRun logo" width="150"/></p>
+
+# Using MLRun 
+
+MLRun is an open source AI orchestration platform for quickly building and managing continuous (gen) AI applications across their lifecycle. MLRun integrates into your development and CI/CD environment and automates the delivery of production data, ML pipelines, and online applications. 
+MLRun significantly reduces engineering efforts, time to production, and computation resources. 
+With MLRun, you can choose any IDE on your local machine or on the cloud. MLRun breaks the silos between data, ML, software, and DevOps/MLOps teams, enabling collaboration and fast continuous improvements.
+
+Get started with the MLRun [**Tutorials and Examples**](https://docs.mlrun.org/en/stable/tutorials/index.html) and the [**Installation and setup guide**](https://docs.mlrun.org/en/stable/install.html), or read about the [**MLRun Architecture**](https://docs.mlrun.org/en/stable/architecture.html).
+
+This page explains how MLRun addresses the [**gen AI tasks**](#genai-tasks), [**MLOps tasks**](#mlops-tasks), and presents the [**MLRun core components**](#core-components).
+
+See the supported data stores, development tools, services, platforms, etc., supported by MLRun's open architecture in **https://docs.mlrun.org/en/stable/ecosystem.html**.
+
+## Gen AI tasks
+
+<p align="center"><img src="https://github.com/mlrun/mlrun/raw/development/docs/_static/images/ai-tasks.png" alt="ai-tasks" width="800"/></p><br>
+
+Use MLRun to develop, scale, deploy, and monitor your AI model across your enterprise. The [**gen AI development workflow**](https://docs.mlrun.org/en/stable/genai/genai-flow.html) 
+section describes the different tasks and stages in detail.
+
+### Data management
+
+
+MLRun supports batch or realtime data processing at scale, data lineage and versioning, structured and unstructured data, and more. 
+Removing inappropriate data at an early stage saves resources that would otherwise be required later on.
+
+
+**Docs:**
+[Using LLMs to process unstructured data](https://docs.mlrun.org/en/stable/genai/data-mgmt/unstructured-data.html)
+[Vector databases](https://docs.mlrun.org/en/stable/genai/data-mgmt/vector-databases.html)
+[Guardrails for data management](https://docs.mlrun.org/en/stable/genai/data-mgmt/guardrails-data.html)
+**Demo:**
+[Call center demo](https://github.com/mlrun/demo-call-center>`
+**Video:**
+[Call center](https://youtu.be/YycMbxRgLBA>`
+
+### Development
+Use MLRun to build an automated ML pipeline to: collect data, 
+preprocess (prepare) the data, run the training pipeline, and evaluate the model.
+
+**Docs:**
+[Working with RAG](https://docs.mlrun.org/en/stable/genai/development/working-with-rag.html), [Evalating LLMs](https://docs.mlrun.org/en/stable/genai/development/evaluating-llms.html), [Fine tuning LLMS](https://docs.mlrun.org/en/stable/genai/development/fine-tuning-llms.html)
+**Demos:**
+[Call center demo](https://github.com/mlrun/demo-call-center), [Build & deploy custom (fine-tuned) LLM models and applications](https://github.com/mlrun/demo-llm-tuning/blob/main), [Interactive bot demo using LLMs](https://github.com/mlrun/demo-llm-bot/blob/main)
+**Video:**
+[Call center](https://youtu.be/YycMbxRgLBA)
+
+
+### Deployment
+MLRun serving can productize the newly trained LLM as a serverless function using real-time auto-scaling Nuclio serverless functions. 
+The application pipeline includes all the steps from accepting events or data, contextualizing it with a state  preparing the required model features, 
+inferring results using one or more models, and driving actions. 
+
+
+**Docs:**
+[Serving gen AI models](https://docs.mlrun.org/en/stable/genai/deployment/genai_serving.html), GPU utilization](https://docs.mlrun.org/en/stable/genai/deployment/gpu_utilization.html), [Gen AI realtime serving graph](https://docs.mlrun.org/en/stable/genai/deployment/genai_serving_graph.html)
+**Tutorial:**
+[Deploy LLM using MLRun](https://docs.mlrun.org/en/stable/tutorials/genai_01_basic_tutorial.html)
+**Demos:**
+[Call center demo](https://github.com/mlrun/demo-call-center), [Build & deploy custom(fine-tuned)]LLM models and applications <https://github.com/mlrun/demo-llm-tuning/blob/main), [Interactive bot demo using LLMs]<https://github.com/mlrun/demo-llm-bot/blob/main)
+**Video:**
+[Call center]<https://youtu.be/YycMbxRgLBA)
+
+
+### Live Ops
+Monitor all resources, data, model and application metrics to ensure performance. Then identify risks, control costs, and measure business KPIs.
+Collect production data, metadata, and metrics to tune the model and application further, and to enable governance and explainability.
+
+
+**Docs:**
+[Model monitoring <monitoring](https://docs.mlrun.org/en/stable/concepts/monitoring.html), [Alerts and notifications](https://docs.mlrun.org/en/stable/concepts/alerts-notifications.html)
+**Tutorials:**
+[Deploy LLM using MLRun](https://docs.mlrun.org/en/stable/tutorials/genai_01_basic_tutorial.html), [Model monitoring using LLM](https://docs.mlrun.org/en/stable/tutorials/genai-02-monitoring-llm.html)
+**Demo:**
+[Build & deploy custom (fine-tuned) LLM models and applications](https://github.com/mlrun/demo-llm-tuning/blob/main)
+
+
+<a id="mlops-tasks"></a>
+## MLOps tasks
+
+<p align="center"><img src="https://github.com/mlrun/mlrun/raw/development/docs/_static/images/mlops-task.png" alt="mlrun-tasks" width="800"/></p><br>
+
+The [**MLOps development workflow**](https://docs.mlrun.org/en/stable/mlops-dev-flow.html) section describes the different tasks and stages in detail.
+MLRun can be used to automate and orchestrate all the different tasks or just specific tasks (and integrate them with what you have already deployed).
+
+### Project management and CI/CD automation
+
+In MLRun the assets, metadata, and services (data, functions, jobs, artifacts, models, secrets, etc.) are organized into projects.
+Projects can be imported/exported as a whole, mapped to git repositories or IDE projects (in PyCharm, VSCode, etc.), which enables versioning, collaboration, and CI/CD. 
+Project access can be restricted to a set of users and roles. 
+
+**Docs:** [Projects and Automation](https://docs.mlrun.org/en/stable/projects/project.html), [CI/CD Integration](https://docs.mlrun.org/en/stable/projects/ci-integration.html)
+**Tutorials:** [Quick start](https://docs.mlrun.org/en/stable/tutorials/01-mlrun-basics.html), [Automated ML Pipeline](https://docs.mlrun.org/en/stable/tutorials/04-pipeline.html)
+**Video:** [Quick start](https://youtu.be/xI8KVGLlj7Q).
+
+### Ingest and process data
+
+MLRun provides abstract interfaces to various offline and online [**data sources**](https://docs.mlrun.org/en/stable/store/datastore.html), supports batch or realtime data processing at scale, data lineage and versioning, structured and unstructured data, and more. 
+In addition, the MLRun [**Feature Store**](https://docs.mlrun.org/en/stable/feature-store/feature-store.html) automates the collection, transformation, storage, catalog, serving, and monitoring of data features across the ML lifecycle and enables feature reuse and sharing. 
+
+See: **Docs:** [Ingest and process data](https://docs.mlrun.org/en/stable/data-prep/index.html), [Feature Store](https://docs.mlrun.org/en/stable/feature-store/feature-store.html), [Data & Artifacts](https://docs.mlrun.org/en/stable/concepts/data.html)
+**Tutorials:** [Quick start](https://docs.mlrun.org/en/stable/tutorials/01-mlrun-basics.html), [Feature Store](https://docs.mlrun.org/en/stable/feature-store/basic-demo.html).
+
+### Develop and train models
+
+MLRun allows you to easily build ML pipelines that take data from various sources or the Feature Store and process it, train models at scale with multiple parameters, test models, tracks each experiments, register, version and deploy models, etc. MLRun provides scalable built-in or custom model training services, integrate with any framework and can work with 3rd party training/auto-ML services. You can also bring your own pre-trained model and use it in the pipeline.
+
+**Docs:** [Develop and train models](https://docs.mlrun.org/en/stable/development/index.html), [Model Training and Tracking](https://docs.mlrun.org/en/stable/development/model-training-tracking.html), [Batch Runs and Workflows](https://docs.mlrun.org/en/stable/concepts/runs-workflows.html)
+**Tutorials:** [Train, compare, and register models](https://docs.mlrun.org/en/stable/tutorials/02-model-training.html), [Automated ML Pipeline](https://docs.mlrun.org/en/stable/tutorials/04-pipeline.html)
+**Video:** [Train and compare models](https://youtu.be/bZgBsmLMdQo).
+
+### Deploy models and applications
+
+MLRun rapidly deploys and manages production-grade real-time or batch application pipelines using elastic and resilient serverless functions. MLRun addresses the entire ML application: intercepting application/user requests, running data processing tasks, inferencing using one or more models, driving actions, and integrating with the application logic.
+
+**Docs:** [Deploy models and applications](https://docs.mlrun.org/en/stable/deployment/index.html), [Realtime Pipelines](https://docs.mlrun.org/en/stable/serving/serving-graph.html), [Batch Inference](https://docs.mlrun.org/en/stable/deployment/batch_inference.html)
+**Tutorials:** [Realtime Serving](https://docs.mlrun.org/en/stable/tutorials/03-model-serving.html), [Batch Inference](https://docs.mlrun.org/en/stable/tutorials/07-batch-infer.html), [Advanced Pipeline](https://docs.mlrun.org/en/stable/tutorials/07-batch-infer.html)
+**Video:** [Serving pre-trained models](https://youtu.be/OUjOus4dZfw).
+
+### Model Monitoring
+
+Observability is built into the different MLRun objects (data, functions, jobs, models, pipelines, etc.), eliminating the need for complex integrations and code instrumentation. With MLRun, you can observe the application/model resource usage and model behavior (drift, performance, etc.), define custom app metrics, and trigger alerts or retraining jobs.
+
+**Docs:** [Model monitoring](https://docs.mlrun.org/en/stable/concepts/model-monitoring.html), [Model Monitoring Overview](https://docs.mlrun.org/en/stable/monitoring/model-monitoring-deployment.html)
+**Tutorials:** [Model Monitoring & Drift Detection](https://docs.mlrun.org/en/stable/tutorials/05-model-monitoring.html).
+
+
+<a id="core-components"></a>
+## MLRun core components
+
+<p align="center"><img src="https://github.com/mlrun/mlrun/raw/development/docs/_static/images/mlops-core.png" alt="mlrun-core" width="800"/></p><br>
+
+
+MLRun includes the following major components:
+
+[**Project Management:**](https://docs.mlrun.org/en/stable/projects/project.html) A service (API, SDK, DB, UI) that manages the different project assets (data, functions, jobs, workflows, secrets, etc.) and provides central control and metadata layer.  
+
+[**Functions:**](https://docs.mlrun.org/en/stable/runtimes/functions.html) automatically deployed software package with one or more methods and runtime-specific attributes (such as image, libraries, command, arguments, resources, etc.).
+
+[**Data & Artifacts:**](https://docs.mlrun.org/en/stable/concepts/data.html) Glueless connectivity to various data sources, metadata management, catalog, and versioning for structures/unstructured artifacts.
+
+[**Batch Runs & Workflows:**](https://docs.mlrun.org/en/stable/concepts/runs-workflows.html) Execute one or more functions with specific parameters and collect, track, and compare all their results and artifacts.
+
+[**Real-Time Serving Pipeline:**](https://docs.mlrun.org/en/stable/serving/serving-graph.html) Rapid deployment of scalable data and ML pipelines using real-time serverless technology, including API handling, data preparation/enrichment, model serving, ensembles, driving and measuring actions, etc.
+
+[**Model monitoring:**](https://docs.mlrun.org/en/stable/monitoring/index.html) monitors data, models, resources, and production components and provides a feedback loop for exploring production data, identifying drift, alerting on anomalies or data quality issues, triggering retraining jobs, measuring business impact, etc.
+
+[**Alerts and notifications:**](https://docs.mlrun.org/en/stable/concepts/model-monitoring.html) Use alerts to identify and inform you of possible problem situations. Use notifications to report status on runs and pipelines.
+
+[**Feature Store:**](https://docs.mlrun.org/en/stable/feature-store/feature-store.html) automatically collects, prepares, catalogs, and serves production data features for development (offline) and real-time (online) deployment using minimal engineering effort.