Skip to content

MLflow setup #4689

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

Merged
merged 7 commits into from
Jul 2, 2024
Merged

MLflow setup #4689

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
afc684a
Add SageMaker MLflow examples
bobbywlindsey Jun 18, 2024
a4ef61a
Add badges
bobbywlindsey Jun 18, 2024
ede83f4
Merge branch 'aws:main' into master
bobbywlindsey Jun 26, 2024
2b18595
Add MLflow setup notebook; upgrade SageMaker Python SDK for deploymen…
bobbywlindsey Jun 26, 2024
4b777db
Linting
bobbywlindsey Jul 1, 2024
9cf9c13
Merge branch 'aws:main' into master
bobbywlindsey Jul 1, 2024
a6ad9bb
More linting changes
bobbywlindsey Jul 2, 2024
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
55 changes: 32 additions & 23 deletions sagemaker-mlflow/sagemaker_deployment_mlflow.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,6 +23,22 @@
"## Setup environment"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Upgrade SageMaker Python SDK"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"!pip install --upgrade --quiet sagemaker>=2.215.0"
]
},
{
"cell_type": "markdown",
"metadata": {},
Expand Down Expand Up @@ -86,10 +102,10 @@
"region = sagemaker_session.boto_region_name\n",
"\n",
"# S3 prefix for the training dataset to be uploaded to\n",
"prefix = 'DEMO-scikit-iris'\n",
"prefix = \"DEMO-scikit-iris\"\n",
"\n",
"# Provide the ARN of the Tracking Server that you want to track your training job with\n",
"tracking_server_arn = 'your tracking server arn here'"
"tracking_server_arn = \"your tracking server arn here\""
]
},
{
Expand Down Expand Up @@ -125,13 +141,13 @@
"\n",
"s3_client = boto3.client(\"s3\")\n",
"s3_client.download_file(\n",
" f\"sagemaker-example-files-prod-{region}\", 'datasets/tabular/iris/iris.data', './data/iris.csv'\n",
" f\"sagemaker-example-files-prod-{region}\", \"datasets/tabular/iris/iris.data\", \"./data/iris.csv\"\n",
")\n",
"\n",
"df_iris = pd.read_csv('./data/iris.csv', header=None)\n",
"df_iris[4] = df_iris[4].map({\"Iris-setosa\": 0, 'Iris-versicolor': 1, 'Iris-virginica': 2})\n",
"df_iris = pd.read_csv(\"./data/iris.csv\", header=None)\n",
"df_iris[4] = df_iris[4].map({\"Iris-setosa\": 0, \"Iris-versicolor\": 1, \"Iris-virginica\": 2})\n",
"iris = df_iris[[4, 0, 1, 2, 3]].to_numpy()\n",
"np.savetxt('./data/iris.csv', iris, delimiter=',', fmt='%1.1f, %1.3f, %1.3f, %1.3f, %1.3f')"
"np.savetxt(\"./data/iris.csv\", iris, delimiter=\",\", fmt=\"%1.1f, %1.3f, %1.3f, %1.3f, %1.3f\")"
]
},
{
Expand All @@ -147,10 +163,10 @@
"metadata": {},
"outputs": [],
"source": [
"WORK_DIRECTORY = 'data'\n",
"WORK_DIRECTORY = \"data\"\n",
"\n",
"train_input = sagemaker_session.upload_data(\n",
" WORK_DIRECTORY, key_prefix='{}/{}'.format(prefix, WORK_DIRECTORY)\n",
" WORK_DIRECTORY, key_prefix=\"{}/{}\".format(prefix, WORK_DIRECTORY)\n",
")"
]
},
Expand Down Expand Up @@ -278,17 +294,15 @@
"outputs": [],
"source": [
"sklearn = SKLearn(\n",
" entry_point='train.py',\n",
" source_dir='training_code',\n",
" framework_version='1.2-1',\n",
" instance_type='ml.c4.xlarge',\n",
" entry_point=\"train.py\",\n",
" source_dir=\"training_code\",\n",
" framework_version=\"1.2-1\",\n",
" instance_type=\"ml.c4.xlarge\",\n",
" role=role,\n",
" sagemaker_session=sagemaker_session,\n",
" hyperparameters={'max_leaf_nodes': 30},\n",
" hyperparameters={\"max_leaf_nodes\": 30},\n",
" keep_alive_period_in_seconds=3600,\n",
" environment={\n",
" 'MLFLOW_TRACKING_ARN': tracking_server_arn\n",
" }\n",
" environment={\"MLFLOW_TRACKING_ARN\": tracking_server_arn},\n",
")"
]
},
Expand Down Expand Up @@ -394,9 +408,7 @@
" mode=Mode.SAGEMAKER_ENDPOINT,\n",
" schema_builder=sklearn_schema_builder,\n",
" role_arn=role,\n",
" model_metadata={\n",
" \"MLFLOW_MODEL_PATH\": source_path\n",
" }\n",
" model_metadata={\"MLFLOW_MODEL_PATH\": source_path},\n",
")"
]
},
Expand All @@ -415,10 +427,7 @@
"metadata": {},
"outputs": [],
"source": [
"predictor = built_model.deploy(\n",
" initial_instance_count=1,\n",
" instance_type=\"ml.m5.large\"\n",
")"
"predictor = built_model.deploy(initial_instance_count=1, instance_type=\"ml.m5.large\")"
]
},
{
Expand Down
30 changes: 12 additions & 18 deletions sagemaker-mlflow/sagemaker_hpo_mlflow.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -109,11 +109,11 @@
"bucket = sagemaker_session.default_bucket()\n",
"\n",
"# S3 prefix for the training dataset to be uploaded to\n",
"prefix = 'DEMO-pytorch-mnist'\n",
"prefix = \"DEMO-pytorch-mnist\"\n",
"\n",
"# MLflow (replace these values with your own)\n",
"tracking_server_arn = 'your tracking server arn'\n",
"experiment_name = 'MNIST'"
"tracking_server_arn = \"your tracking server arn\"\n",
"experiment_name = \"MNIST\""
]
},
{
Expand Down Expand Up @@ -149,9 +149,9 @@
"metadata": {},
"outputs": [],
"source": [
"local_dir = 'data'\n",
"local_dir = \"data\"\n",
"MNIST.mirrors = [\n",
" f'https://sagemaker-example-files-prod-{region}.s3.amazonaws.com/datasets/image/MNIST/'\n",
" f\"https://sagemaker-example-files-prod-{region}.s3.amazonaws.com/datasets/image/MNIST/\"\n",
"]\n",
"MNIST(\n",
" local_dir,\n",
Expand All @@ -177,7 +177,7 @@
"metadata": {},
"outputs": [],
"source": [
"train_input = sagemaker_session.upload_data(path='data', bucket=bucket, key_prefix=prefix)"
"train_input = sagemaker_session.upload_data(path=\"data\", bucket=bucket, key_prefix=prefix)"
]
},
{
Expand Down Expand Up @@ -577,10 +577,7 @@
"\n",
"objective_metric_name = \"average test loss\"\n",
"objective_type = \"Minimize\"\n",
"metric_definitions = [\n",
" {\"Name\": \"average test loss\",\n",
" \"Regex\": \"Test set: Average loss: ([0-9\\\\.]+)\"}\n",
"]"
"metric_definitions = [{\"Name\": \"average test loss\", \"Regex\": \"Test set: Average loss: ([0-9\\\\.]+)\"}]"
]
},
{
Expand Down Expand Up @@ -612,17 +609,14 @@
" framework_version=\"1.13\",\n",
" instance_count=1,\n",
" instance_type=\"ml.c5.2xlarge\",\n",
" hyperparameters={\n",
" \"epochs\": 5,\n",
" \"backend\": \"gloo\"\n",
" },\n",
" hyperparameters={\"epochs\": 5, \"backend\": \"gloo\"},\n",
" environment={\n",
" 'MLFLOW_TRACKING_URI':tracking_server_arn,\n",
" 'MLFLOW_EXPERIMENT_NAME':experiment.name,\n",
" 'MLFLOW_PARENT_RUN_ID':run.info.run_id\n",
" \"MLFLOW_TRACKING_URI\": tracking_server_arn,\n",
" \"MLFLOW_EXPERIMENT_NAME\": experiment.name,\n",
" \"MLFLOW_PARENT_RUN_ID\": run.info.run_id,\n",
" },\n",
" )\n",
" \n",
"\n",
" tuner = HyperparameterTuner(\n",
" estimator,\n",
" objective_metric_name,\n",
Expand Down
Loading