You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: doc/overview.rst
+11-11
Original file line number
Diff line number
Diff line change
@@ -1034,7 +1034,7 @@ You can deploy the XGBoost model from the previous example to a SageMaker endpoi
1034
1034
initial_instance_count=1
1035
1035
)
1036
1036
1037
-
For a sample notebook that demonstrates using ``ModelBuilder`` to build a XGBoost model, see `XGBoost example <https://github.com/aws-samples/sagemaker-hosting/SageMaker-Model-Builder/traditional-models/model-builder-xgboost.ipynb>`_.
1037
+
For a sample notebook that demonstrates using ``ModelBuilder`` to build a XGBoost model, see `XGBoost example <https://github.com/aws-samples/sagemaker-hosting/blob/main/SageMaker-Model-Builder/traditional-models/model-builder-xgboost.ipynb>`_.
1038
1038
1039
1039
**Triton models**.
1040
1040
You can use ``ModelBuilder`` to serve PyTorch models on a Triton inference server. Specify the ``model_server`` parameter as ``ModelServer.TRITON``, pass a model, and include a ``SchemaBuilder`` object which requires sample inputs and outputs from the model. The following snippet shows an example.
@@ -1056,7 +1056,7 @@ You can use ``ModelBuilder`` to serve PyTorch models on a Triton inference serve
1056
1056
initial_instance_count=1
1057
1057
)
1058
1058
1059
-
For a sample notebook that demonstrates using ``ModelBuilder`` to build a Triton model, see `Triton example <https://github.com/aws-samples/sagemaker-hosting/SageMaker-Model-Builder/traditional-models/model-builder-triton.ipynb>`_.
1059
+
For a sample notebook that demonstrates using ``ModelBuilder`` to build a Triton model, see `Triton example <https://github.com/aws-samples/sagemaker-hosting/blob/main/SageMaker-Model-Builder/traditional-models/model-builder-triton.ipynb>`_.
1060
1060
1061
1061
1062
1062
**Hugging Face models**.
@@ -1101,7 +1101,7 @@ Create the ``ModelBuilder`` object and deploy the model onto a SageMaker endpoin
1101
1101
instance_type='ml.g5.2xlarge'
1102
1102
)
1103
1103
1104
-
For a sample notebook that demonstrates using ``ModelBuilder`` to build a Hugging Face model, see `Hugging Face example <https://github.com/aws-samples/sagemaker-hosting/SageMaker-Model-Builder/traditional-models/model-builder-huggingface.ipynb>`_.
1104
+
For a sample notebook that demonstrates using ``ModelBuilder`` to build a Hugging Face model, see `Hugging Face example <https://github.com/aws-samples/sagemaker-hosting/blob/main/SageMaker-Model-Builder/traditional-models/model-builder-huggingface.ipynb>`_.
1105
1105
1106
1106
1107
1107
Deploy foundation models to SageMaker Endpoints
@@ -1135,7 +1135,7 @@ For gated models on Hugging Face Hub, request access and pass the associated key
1135
1135
1136
1136
A feature of ``ModelBuilder`` is the ability to run local tuning on the container when you use `LOCAL_CONTAINER` mode. In this case ``ModelBuilder`` tunes the parameter(s) for the underlying model server. This feature can be used by executing `tuned_model=model.tune()`. Before running `tune`, clean up other containers running locally or else you might see an "address already in use" error.
1137
1137
1138
-
For a sample notebook that demonstrates using ``ModelBuilder`` to build a Hugging Face Hub model, see `Hugging Face Hub example <https://github.com/aws-samples/sagemaker-hosting/SageMaker-Model-Builder/model-builder-huggingface-llama2.ipynb>`_.
1138
+
For a sample notebook that demonstrates using ``ModelBuilder`` to build a Hugging Face Hub model, see `Hugging Face Hub example <https://github.com/aws-samples/sagemaker-hosting/blob/main/SageMaker-Model-Builder/model-builder-huggingface-llama2.ipynb>`_.
1139
1139
1140
1140
1141
1141
**JumpStart**. JumpStart also offers a number of pre-trained foundation models. Again, the model ID is required. Deploying a JumpStart model to a SageMaker endpoint is straightforward, as shown in the following example:
@@ -1154,24 +1154,24 @@ For a sample notebook that demonstrates using ``ModelBuilder`` to build a Huggin
1154
1154
1155
1155
For a list of available JumpStart model IDs, see `Built-in Algorithms with pre-trained Model Table <https://sagemaker.readthedocs.io/en/stable/doc_utils/pretrainedmodels.html>`_.
1156
1156
1157
-
For a sample notebook that demonstrates using ``ModelBuilder`` to build a JumpStart model, see `JumpStart example <https://github.com/aws-samples/sagemaker-hosting/SageMaker-Model-Builder/foundation-models/model-builder-jumpstart-falcon.ipynb>`_.
1157
+
For a sample notebook that demonstrates using ``ModelBuilder`` to build a JumpStart model, see `JumpStart example <https://github.com/aws-samples/sagemaker-hosting/blob/main/SageMaker-Model-Builder/foundation-models/model-builder-jumpstart-falcon.ipynb>`_.
1158
1158
1159
1159
ModelBuilder examples
1160
1160
^^^^^^^^^^^^^^^^^^^^^
1161
1161
1162
1162
For example notebooks that demonstrate the use of ``ModelBuilder`` and its supporting classes, as well as model creation of traditional and foundation models, see the following links:
1163
1163
1164
-
* `Pytorch example <https://github.com/aws-samples/sagemaker-hosting/SageMaker-Model-Builder/traditional-models/model-builder-pytorch.ipynb>`_
1164
+
* `Pytorch example <https://github.com/aws-samples/sagemaker-hosting/blob/main/SageMaker-Model-Builder/traditional-models/model-builder-pytorch.ipynb>`__
1165
1165
1166
-
* `XGBoost example <https://github.com/aws-samples/sagemaker-hosting/SageMaker-Model-Builder/traditional-models/model-builder-xgboost.ipynb>`_
1166
+
* `XGBoost example <https://github.com/aws-samples/sagemaker-hosting/blob/main/SageMaker-Model-Builder/traditional-models/model-builder-xgboost.ipynb>`__
1167
1167
1168
-
* `Triton example <https://github.com/aws-samples/sagemaker-hosting/SageMaker-Model-Builder/traditional-models/model-builder-triton.ipynb>`_
1168
+
* `Triton example <https://github.com/aws-samples/sagemaker-hosting/blob/main/SageMaker-Model-Builder/traditional-models/model-builder-triton.ipynb>`__
1169
1169
1170
-
* `Hugging Face example <https://github.com/aws-samples/sagemaker-hosting/SageMaker-Model-Builder/traditional-models/model-builder-huggingface.ipynb>`_
1170
+
* `Hugging Face example <https://github.com/aws-samples/sagemaker-hosting/blob/main/SageMaker-Model-Builder/traditional-models/model-builder-huggingface.ipynb>`__
1171
1171
1172
-
* `Hugging Face Hub example <https://github.com/aws-samples/sagemaker-hosting/SageMaker-Model-Builder/model-builder-huggingface-llama2.ipynb>`_
1172
+
* `Hugging Face Hub example <https://github.com/aws-samples/sagemaker-hosting/blob/main/SageMaker-Model-Builder/foundation-models/model-builder-huggingface-llama2.ipynb>`__
1173
1173
1174
-
* `JumpStart example <https://github.com/aws-samples/sagemaker-hosting/SageMaker-Model-Builder/foundation-models/model-builder-jumpstart-falcon.ipynb>`_
1174
+
* `JumpStart example <https://github.com/aws-samples/sagemaker-hosting/blob/main/SageMaker-Model-Builder/foundation-models/model-builder-jumpstart-falcon.ipynb>`__
1175
1175
1176
1176
1177
1177
Fine-tune a Model and Deploy to a SageMaker Endpoint
0 commit comments