Skip to content

bump tensorflow (do not merge) #125

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 2 commits into
base: main
Choose a base branch
from
Open

bump tensorflow (do not merge) #125

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
67d3331
bump tensorflow
MichaelSkralivetsky Jun 19, 2025
4a3ad57
Merge branch 'main' into bump_tensorflow
MichaelSkralivetsky Jun 19, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
36 changes: 28 additions & 8 deletions test-horovod-tensorflow/test-horovod-tensorflow.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -44,8 +44,8 @@
},
"outputs": [],
"source": [
"# per ML-3824 need to install tensorflow and mlrun in the same command\n",
"!pip install plotly tensorflow==2.15.1 mlrun # TODO: remove 2.15.1 here and in functions requirements after ML-10271 fix"
"# with tensorflow==2.19.1 horovod build fails\n",
"!pip install plotly tensorflow==2.18.1 mlrun"
]
},
{
Expand Down Expand Up @@ -118,7 +118,7 @@
"def get_model() -> tf.keras.Model:\n",
" # Build the model architecture:\n",
" inputs = tf.keras.Input(shape=(28, 28))\n",
" x = tf.keras.layers.experimental.preprocessing.Rescaling(1.0 / 255)(inputs)\n",
" x = tf.keras.layers.Rescaling(1.0 / 255)(inputs)\n",
" x = tf.keras.layers.Flatten()(x)\n",
" x = tf.keras.layers.Dense(128, activation=\"relu\")(x)\n",
" x = tf.keras.layers.Dense(128, activation=\"relu\")(x)\n",
Expand Down Expand Up @@ -147,7 +147,7 @@
"\n",
" # Compile the model:\n",
" model.compile(\n",
" optimizer=tf.keras.optimizers.legacy.SGD(learning_rate=0.1, momentum=0.9),\n",
" optimizer = tf.keras.optimizers.SGD(learning_rate=0.1, momentum=0.9),\n",
" loss=\"sparse_categorical_crossentropy\",\n",
" metrics=[\"accuracy\"]\n",
" )\n",
Expand Down Expand Up @@ -221,7 +221,7 @@
"outputs": [],
"source": [
"# Create the job function:\n",
"job_function = project.set_function(\"./func.py\", name=\"train_job\", kind=\"job\", image=\"mlrun/mlrun\", handler=\"train\", requirements=[\"tensorflow==2.15.1\"])\n",
"job_function = project.set_function(\"./func.py\", name=\"train_job\", kind=\"job\", image=\"mlrun/mlrun\", handler=\"train\", requirements=[\"tensorflow==2.18.1\"])\n",
"job_function.apply(mlrun.auto_mount())\n",
"job_function.deploy()"
]
Expand All @@ -238,11 +238,15 @@
"outputs": [],
"source": [
"# Create the open mpi function:\n",
"mpijob_function = project.set_function(\"./func.py\", name=\"train_mpijob\", kind=\"mpijob\", image=\"mlrun/mlrun\", handler=\"train\", requirements=[\"horovod[tensorflow]\"])\n",
"mpijob_function = project.set_function(\"./func.py\", name=\"train_mpijob\", kind=\"mpijob\", image=\"mlrun/mlrun\", handler=\"train\")\n",
"mpijob_function.apply(mlrun.auto_mount())\n",
"mpijob_function.spec.replicas = N_RANKS\n",
"mpijob_function.with_commands([\"pip install tensorflow==2.15.1\"])\n",
"mpijob_function.deploy(builder_env={\"HOROVOD_WITH_TENSORFLOW\": \"1\"})"
"mpijob_function.with_commands([\n",
" \"apt-get update && apt-get install -y build-essential g++ cmake llvm-dev libclang-dev\",\n",
" \"pip install tensorflow==2.18.1\",\n",
" \"HOROVOD_WITH_TENSORFLOW=1 HOROVOD_WITHOUT_XLA=1 HOROVOD_WITHOUT_MLX=1 pip install horovod[tensorflow]\",\n",
"])\n",
"mpijob_function.deploy()\n"
]
},
{
Expand Down Expand Up @@ -384,6 +388,22 @@
"# assert mpijob_time < job_time\n",
"# assert np.isclose(job_accuracy, mpijob_accuracy, atol=0.1)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "31ba0442-a98b-46fc-a680-9dfcdf1dfe9a",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "81d92305-268f-4ead-87df-5ab3a30f294b",
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
Expand Down