mlrun 1.9.2rc1__py3-none-any.whl → 1.9.2rc2__py3-none-any.whl

mlrun/frameworks/tf_keras/mlrun_interface.py

@@ -107,14 +107,10 @@ class TFKerasMLRunInterface(MLRunInterface, ABC):
             )
 
         # Call the pre compile method:
-        (optimizer, experimental_run_tf_function) = self._pre_compile(
-            optimizer=kwargs["optimizer"]
-        )
+        optimizer = self._pre_compile(optimizer=kwargs["optimizer"])
 
         # Assign parameters:
         kwargs["optimizer"] = optimizer
-        if experimental_run_tf_function is not None:
-            kwargs["experimental_run_tf_function"] = experimental_run_tf_function
 
         # Call the original compile method:
         return self.original_compile(*args, **kwargs)
@@ -235,23 +231,20 @@ class TFKerasMLRunInterface(MLRunInterface, ABC):
         """
         self._RANK_0_ONLY_CALLBACKS.add(callback_name)
 
-    def _pre_compile(self, optimizer: Optimizer) -> tuple[Optimizer, Union[bool, None]]:
+    def _pre_compile(self, optimizer: Optimizer) -> Optimizer:
         """
         Method to call before calling 'compile' to setup the run and inputs for using horovod.
 
         :param optimizer: The optimzier to compile. It will be wrapped in horovod's distributed optimizer:
                           'hvd.DistributedOptimizer'.
 
-        :return: The updated parameters:
-                 [0] = Wrapped optimizer.
-                 [1] = The 'experimental_run_tf_function' parameter for 'compile' kwargs or 'None' if horovod should not
-                       be used.
+        :return: The updated Wrapped optimizer.
 
         :raise MLRunInvalidArgumentError: In case the optimizer was passed as a string.
         """
         # Check if needed to run with horovod:
         if self._hvd is None:
-            return optimizer, None
+            return optimizer
 
         # Validate the optimizer input:
         if isinstance(optimizer, str):
@@ -288,11 +281,7 @@ class TFKerasMLRunInterface(MLRunInterface, ABC):
         # Wrap the optimizer in horovod's distributed optimizer: 'hvd.DistributedOptimizer'.
         optimizer = self._hvd.DistributedOptimizer(optimizer)
 
-        # Compile the model with `experimental_run_tf_function=False` to ensure Tensorflow uses the distributed
-        # optimizer to compute the gradients:
-        experimental_run_tf_function = False
-
-        return optimizer, experimental_run_tf_function
+        return optimizer
 
     def _pre_fit(
         self,

mlrun/utils/version/version.json

@@ -1,4 +1,4 @@
 {
-  "git_commit": "8c4abef3f59c6e019d0584b32d1c12341b81c4c8",
-  "version": "1.9.2-rc1"
+  "git_commit": "96b5c672c395084970c71c8ea780f3fd00d7547f",
+  "version": "1.9.2-rc2"
 }

{mlrun-1.9.2rc1.dist-info → mlrun-1.9.2rc2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mlrun
-Version: 1.9.2rc1
+Version: 1.9.2rc2
 Summary: Tracking and config of machine learning runs
 Home-page: https://github.com/mlrun/mlrun
 Author: Yaron Haviv
@@ -32,7 +32,7 @@ Requires-Dist: ipython~=8.10
 Requires-Dist: nuclio-jupyter~=0.11.1
 Requires-Dist: numpy<1.27.0,>=1.26.4
 Requires-Dist: pandas<2.2,>=1.2
-Requires-Dist: pyarrow<17,>=10.0
+Requires-Dist: pyarrow<18,>=10.0
 Requires-Dist: pyyaml<7,>=6.0.2
 Requires-Dist: requests~=2.32
 Requires-Dist: tabulate~=0.8.6

{mlrun-1.9.2rc1.dist-info → mlrun-1.9.2rc2.dist-info}/RECORD

@@ -199,7 +199,7 @@ mlrun/frameworks/sklearn/mlrun_interface.py,sha256=JzHMBQM4sPBJqzb8P-rsG_2RQ_QrX
 mlrun/frameworks/sklearn/model_handler.py,sha256=n0vpsQznva_WVloz7GTnfMGcMDQU_f1bHhUAJ_qxjfE,4753
 mlrun/frameworks/sklearn/utils.py,sha256=Cg_pSxUMvKe8vBSLQor6JM8u9_ccKJg4Rk5EPDzTsVo,1209
 mlrun/frameworks/tf_keras/__init__.py,sha256=M2sMbYHLrlF-KFR5kvA9mevRo3Nf8U0B5a_DM9rzwCY,10484
-mlrun/frameworks/tf_keras/mlrun_interface.py,sha256=_LVPKsPYopG8osoWfuxeKOkPcchb_oiL_IufWbnpIBI,16753
+mlrun/frameworks/tf_keras/mlrun_interface.py,sha256=ZKBqlkPY9Kk0Mhd1TLEamWCAOGTaiCyYytSCOlbYxXQ,16110
 mlrun/frameworks/tf_keras/model_handler.py,sha256=2PZgxgB0XLIM6V2pLTWpU6UVKrjgGdSgcQNPz2kwFnU,32170
 mlrun/frameworks/tf_keras/model_server.py,sha256=PZW6OBGTJ6bSfHedAhhW8HATbJyp2VaAzSDC02zjyKk,9653
 mlrun/frameworks/tf_keras/utils.py,sha256=Z8hA1CgpSJWLC_T6Ay7xZKVyWlX9B85MSmQr2biXRag,4582
@@ -341,11 +341,11 @@ mlrun/utils/notifications/notification/mail.py,sha256=ZyJ3eqd8simxffQmXzqd3bgbAq
 mlrun/utils/notifications/notification/slack.py,sha256=kfhogR5keR7Zjh0VCjJNK3NR5_yXT7Cv-x9GdOUW4Z8,7294
 mlrun/utils/notifications/notification/webhook.py,sha256=zxh8CAlbPnTazsk6r05X5TKwqUZVOH5KBU2fJbzQlG4,5330
 mlrun/utils/version/__init__.py,sha256=7kkrB7hEZ3cLXoWj1kPoDwo4MaswsI2JVOBpbKgPAgc,614
-mlrun/utils/version/version.json,sha256=klaCU-7EeZjL6BRB4x4M20KH9-JhfypJswvqxMbaTls,88
+mlrun/utils/version/version.json,sha256=h2GhQN_dPdcVCBljnGSQyo8yO7qQRFJpm2Y-AyD8pvQ,88
 mlrun/utils/version/version.py,sha256=eEW0tqIAkU9Xifxv8Z9_qsYnNhn3YH7NRAfM-pPLt1g,1878
-mlrun-1.9.2rc1.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-mlrun-1.9.2rc1.dist-info/METADATA,sha256=saGsXYF-iRbyTmoCPaRvAjGsoC7lWyStRm3L0upzA-E,25759
-mlrun-1.9.2rc1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-mlrun-1.9.2rc1.dist-info/entry_points.txt,sha256=1Owd16eAclD5pfRCoJpYC2ZJSyGNTtUr0nCELMioMmU,46
-mlrun-1.9.2rc1.dist-info/top_level.txt,sha256=NObLzw3maSF9wVrgSeYBv-fgnHkAJ1kEkh12DLdd5KM,6
-mlrun-1.9.2rc1.dist-info/RECORD,,
+mlrun-1.9.2rc2.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+mlrun-1.9.2rc2.dist-info/METADATA,sha256=R_67NXEjUDMY7b-VF9FkZD_q6Ms7MHd2IyKxCZnWS-g,25759
+mlrun-1.9.2rc2.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+mlrun-1.9.2rc2.dist-info/entry_points.txt,sha256=1Owd16eAclD5pfRCoJpYC2ZJSyGNTtUr0nCELMioMmU,46
+mlrun-1.9.2rc2.dist-info/top_level.txt,sha256=NObLzw3maSF9wVrgSeYBv-fgnHkAJ1kEkh12DLdd5KM,6
+mlrun-1.9.2rc2.dist-info/RECORD,,