change: add integs

Author: benieric

tests/data/modules/params_script/hyperparameters.json

@@ -0,0 +1,14 @@
+{
+    "integer": 1,
+    "boolean": true,
+    "float": 3.14,
+    "string": "Hello World",
+    "list": [1, 2, 3],
+    "dict": {
+        "string": "value",
+        "integer": 3,
+        "list": [1, 2, 3],
+        "dict": {"key": "value"},
+        "boolean": true
+    }
+}

tests/data/modules/params_script/hyperparameters.yaml

@@ -0,0 +1,18 @@
+integer: 1
+boolean: true
+float: 3.14
+string: "Hello World"
+list:
+  - 1
+  - 2
+  - 3
+dict:
+  string: value
+  integer: 3
+  list:
+    - 1
+    - 2
+    - 3
+  dict:
+    key: value
+  boolean: true

tests/integ/sagemaker/modules/train/test_model_trainer.py

@@ -106,3 +106,37 @@ def test_hp_contract_torchrun_script(modules_sagemaker_session):
     )
 
     model_trainer.train()
+
+
+def test_hp_contract_hyperparameter_json(modules_sagemaker_session):
+    source_dir = f"{DATA_DIR}/modules/params_script"
+    source_code = SourceCode(
+        source_dir=source_dir,
+        entry_script="train.py",
+    )
+    model_trainer = ModelTrainer(
+        sagemaker_session=modules_sagemaker_session,
+        training_image=DEFAULT_CPU_IMAGE,
+        hyperparameters_file=f"{source_dir}/hyperparameters.json",
+        source_code=source_code,
+        base_job_name="hp-contract-hyperparameter-json",
+    )
+    assert model_trainer.hyperparameters == EXPECTED_HYPERPARAMETERS
+    model_trainer.train()
+
+
+def test_hp_contract_hyperparameter_yaml(modules_sagemaker_session):
+    source_dir = f"{DATA_DIR}/modules/params_script"
+    source_code = SourceCode(
+        source_dir=source_dir,
+        entry_script="train.py",
+    )
+    model_trainer = ModelTrainer(
+        sagemaker_session=modules_sagemaker_session,
+        training_image=DEFAULT_CPU_IMAGE,
+        hyperparameters_file=f"{source_dir}/hyperparameters.yaml",
+        source_code=source_code,
+        base_job_name="hp-contract-hyperparameter-yaml",
+    )
+    assert model_trainer.hyperparameters == EXPECTED_HYPERPARAMETERS
+    model_trainer.train()