command to start sagemaker training

include sample training

src/config.py

@@ -1,7 +1,8 @@
 from enum import Enum
-from typing import Literal
+from typing import Any, Literal, TypedDict
 
 from mypy_boto3_s3.literals import BucketLocationConstraintType
+from mypy_boto3_sagemaker.literals import TrainingInstanceTypeType
 from pydantic import BaseModel, Field, model_validator
 
 
@@ -17,10 +18,19 @@ class MlflowServerSize(str, Enum):
     large = "Large"
 
 
+class Boto3SessionKwargs(TypedDict):
+    profile_name: str
+    region_name: str
+
+
 class AwsConfig(BaseModel):
     region: BucketLocationConstraintType | Literal["us-east-1"] = "us-east-1"
     profile: str = "default"
 
+    @property
+    def boto3_session(self) -> Boto3SessionKwargs:
+        return {"profile_name": self.profile, "region_name": self.region}
+
 
 class S3Config(BaseModel):
     bucket: str = "my-qc-mlops-bucket"
@@ -28,8 +38,18 @@ class S3Config(BaseModel):
     model_prefix: str = "models/"
 
 
+class TrainingConfig(BaseModel):
+    instance_type: TrainingInstanceTypeType = "ml.m5.xlarge"
+    instance_count: int = 1
+    image_uri: str = ""
+    entry_point: str | None = None
+    source_dir: str | None = None
+    hyperparameters: dict[str, Any] = Field(default_factory=dict)
+
+
 class SageMakerConfig(BaseModel):
     role_name: str = "qc-cli-sagemaker-role"
+    training: TrainingConfig = Field(default_factory=TrainingConfig)
 
 
 class MlflowConfig(BaseModel):