fix: Huggingface glue failing tests (#4367)

* fix: Huggingface glue failing tests

* fix: Sphinx doc build failure

* fix: Huggingface glue failing tests

* fix: failing sphinx tests

* fix: failing sphinx tests

* fix: failing black check

* fix: sphinx doc errors

* fix: sphinx doc errors

* sphinx

* black-format

* sphinx

* sphinx

* sphinx

---------

Co-authored-by: Mufaddal Rohawala <[email protected]>
Co-authored-by: Erick Benitez-Ramos <[email protected]>

Author: 3 people

doc/conf.py

@@ -94,7 +94,7 @@
 }
 
 # Example configuration for intersphinx: refer to the Python standard library.
-intersphinx_mapping = {"http://docs.python.org/": None}
+intersphinx_mapping = {"python": ("http://docs.python.org/", None)}
 
 # -- Options for autodoc ----------------------------------------------------
 # https://www.sphinx-doc.org/en/master/usage/extensions/autodoc.html#configuration

doc/requirements.txt

@@ -1,4 +1,4 @@
-sphinx==3.1.2
+sphinx==3.4.3
 sphinx-rtd-theme==0.5.0
 docutils==0.15.2
 packaging==20.9

src/sagemaker/feature_store/feature_processor/feature_processor.py

@@ -79,7 +79,7 @@ def transform(input_feature_group, input_csv):
             return ...
 
     Args:
-        inputs (Sequence[Union[FeatureGroupDataSource, CSVDataSource, ParquetDataSource,
+        inputs (Sequence[Union[FeatureGroupDataSource, CSVDataSource, ParquetDataSource,\
             BaseDataSource]]): A list of data sources.
         output (str): A Feature Group ARN to write results of this function to.
         target_stores (Optional[list[str]], optional): A list containing at least one of

src/sagemaker/jumpstart/estimator.py

@@ -251,8 +251,8 @@ def __init__(
                 (Default: None).
             model_channel_name (Optional[Union[str, PipelineVariable]]): Name of the channel where
                 'model_uri' will be downloaded. (Default: None).
-            metric_definitions (Optional[Union[list[dict[str, str], list[dict[str,
-                PipelineVariable]]]]): A list of dictionaries that defines the metric(s)
+            metric_definitions (Optional[list[dict[str, Union[str, PipelineVariable]]]]):
+                A list of dictionaries that defines the metric(s)
                 used to evaluate the training jobs. Each dictionary contains two keys: 'Name'
                 for the name of the metric, and 'Regex' for the regular expression used to extract
                 the metric from the logs. This should be defined only for jobs that
@@ -292,8 +292,8 @@ def __init__(
                 SageMaker Debugger rules for real-time analysis
                 (Default: None). For more information,
                 see `Continuous analyses through rules
-                <https://sagemaker.readthedocs.io/en/stable/amazon_sagemaker_debugger.html
-                #continuous-analyses-through-rules)>`_.
+                <https://sagemaker.readthedocs.io/en/stable/amazon_sagemaker_debugger.html#
+                continuous-analyses-through-rules)>`_.
                 (Default: None).
             debugger_hook_config (Optional[Union[DebuggerHookConfig, bool]]):
                 Configuration for how debugging information is emitted with

src/sagemaker/session.py

@@ -4565,20 +4565,18 @@ def update_inference_component(
         Args:
             inference_component_name (str): Name of the Amazon SageMaker ``InferenceComponent``.
             specification ([dict[str,int]]): Resource configuration. Optional.
-            Example: {
+                Example: {
                 "MinMemoryRequiredInMb": 1024,
                 "NumberOfCpuCoresRequired": 1,
                 "NumberOfAcceleratorDevicesRequired": 1,
                 "MaxMemoryRequiredInMb": 4096,
-            },
-
+                },
             runtime_config ([dict[str,int]]): Number of copies. Optional.
-            Default: {
+                Default: {
                 "copyCount": 1
-            }
-
+                }
             wait: Wait for inference component to be created before return. Optional. Default is
-            True.
+                True.
 
         Return:
             str: inference component name

tests/data/huggingface/requirements.txt

@@ -0,0 +1 @@
+datasets==2.16.1

tests/data/huggingface_byoc/requirements.txt

@@ -1,2 +1,2 @@
 transformers<=4.28.1
-datasets<=2.12.0
+datasets==2.16.1

tests/integ/test_huggingface.py

@@ -71,7 +71,8 @@ def test_huggingface_training(
 
         hf = HuggingFace(
             py_version=huggingface_pytorch_latest_training_py_version,
-            entry_point=os.path.join(data_path, "run_glue.py"),
+            source_dir=data_path,
+            entry_point="run_glue.py",
             role="SageMakerRole",
             transformers_version=huggingface_training_latest_version,
             pytorch_version=huggingface_training_pytorch_latest_version,

tests/integ/test_huggingface_torch_distributed.py

@@ -24,10 +24,10 @@ def test_huggingface_torch_distributed_g5_glue(
     huggingface_pytorch_latest_training_py_version,
 ):
     with timeout.timeout(minutes=TRAINING_DEFAULT_TIMEOUT_MINUTES):
-        data_path = os.path.join(DATA_DIR, "huggingface")
         estimator = HuggingFace(
             py_version=huggingface_pytorch_latest_training_py_version,
-            entry_point=os.path.join(data_path, "run_glue.py"),
+            source_dir=os.path.join(DATA_DIR, "huggingface"),
+            entry_point="run_glue.py",
             role="SageMakerRole",
             transformers_version=huggingface_training_latest_version,
             pytorch_version=huggingface_training_pytorch_latest_version,