workbench 0.8.156__tar.gz → 0.8.157__tar.gz

{workbench-0.8.156 → workbench-0.8.157}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: workbench
-Version: 0.8.156
+Version: 0.8.157
 Summary: Workbench: A Dashboard and Python API for creating and deploying AWS SageMaker Model Pipelines
 Author-email: SuperCowPowers LLC <support@supercowpowers.com>
 License-Expression: MIT

{workbench-0.8.156 → workbench-0.8.157}/applications/aws_dashboard/Dockerfile

@@ -7,7 +7,7 @@ EXPOSE 8000
 ENV PYTHONHASHSEED=0
 
 # Define workbench version
-ARG WORKBENCH_VERSION=0.8.155
+ARG WORKBENCH_VERSION=0.8.156
 
 # Install system packages and gunicorn in one layer
 RUN apt-get update && apt-get install -y --no-install-recommends \
@@ -21,15 +21,15 @@ RUN pip install --no-cache-dir "workbench[ui]==${WORKBENCH_VERSION}" && \
     pip uninstall -y workbench && \
     pip cache purge
 
-# Copy app code and configuration
+# Install workbench[ui] in separate layer (dependencies already satisfied)
+RUN pip install --no-cache-dir "workbench[ui]==${WORKBENCH_VERSION}"
+
+# Copy app code and configuration (after pip installs to avoid cache invalidation)
 COPY . /app
 ARG WORKBENCH_CONFIG
 COPY $WORKBENCH_CONFIG /app/workbench_config.json
 ENV WORKBENCH_CONFIG=/app/workbench_config.json
 
-# Install workbench[ui] in separate layer (dependencies already satisfied)
-RUN pip install --no-cache-dir "workbench[ui]==${WORKBENCH_VERSION}"
-
 # Configure Nginx and Supervisor
 COPY nginx.conf /etc/nginx/sites-available/default
 COPY supervisord.conf /etc/supervisor/conf.d/supervisord.conf

{workbench-0.8.156 → workbench-0.8.157}/aws_setup/aws_account_check.py

@@ -45,36 +45,6 @@ class AWSAccountCheck:
                 self.log.error(f"Unexpected error: {e}")
                 sys.exit(1)
 
-    def check_s3_bucket_subfolders(self):
-        """Check if the Workbench S3 Bucket is set up and has the correct sub-folders"""
-
-        self.log.info("*** AWS Workbench Bucket Check ***")
-        s3 = self.aws_clamp.boto3_session.resource("s3")
-        bucket = s3.Bucket(self.workbench_bucket)
-
-        # Check if the bucket exists
-        if bucket.creation_date is None:
-            self.log.critical(f"The {self.workbench_bucket} bucket does not exist")
-            sys.exit(1)
-        else:
-            self.log.info(f"The {self.workbench_bucket} bucket exists")
-
-        # Check if the sub-folders exists
-        sub_folders = ["incoming-data", "data-sources", "feature-sets", "athena-queries"]
-
-        # Get all object prefixes in the bucket
-        prefixes = set()
-        for obj in bucket.objects.all():
-            prefix = obj.key.split("/")[0]
-            prefixes.add(prefix)
-
-        # Check for the existence of the sub-folders
-        for folder in sub_folders:
-            if folder in prefixes:
-                self.log.info(f"The {folder} prefix exists")
-            else:
-                self.log.info(f"The {folder} prefix does not exist...which is fine...")
-
     def check(self):
         """Check if the AWS Account is set up Correctly"""
         self.log.info("*** Caller/Base AWS Identity Check ***")
@@ -85,13 +55,10 @@ class AWSAccountCheck:
         self.aws_clamp.check_assumed_role()
         print("\n")
 
-        self.log.info("*** AWS S3 Access Check ***")
-        self.aws_clamp.check_s3_access()
+        self.log.info("*** AWS Workbench Bucket Check ***")
+        self.aws_clamp.check_workbench_bucket()
         print("\n")
 
-        # Check that the Workbench S3 Bucket and Sub-folders are created
-        self.check_s3_bucket_subfolders()
-
         self.log.info("*** AWS Sagemaker Session/Client Check ***")
         sm_client = self.aws_clamp.sagemaker_client()
         for feature_group in sm_client.list_feature_groups()["FeatureGroupSummaries"]:

{workbench-0.8.156 → workbench-0.8.157}/aws_setup/workbench_core/workbench_core/workbench_core_stack.py

@@ -29,7 +29,6 @@ class WorkbenchCoreStack(Stack):
         super().__init__(scope, construct_id, description=desc, **kwargs)
 
         # Grab our properties
-        self.account_id = env.account
         self.workbench_bucket = props.workbench_bucket
         self.workbench_role_name = props.workbench_role_name
         self.sso_group = props.sso_group
@@ -60,25 +59,12 @@ class WorkbenchCoreStack(Stack):
         arns = []
         for bucket_name_template in bucket_list:
             # Dynamically construct the bucket name
-            bucket_name = bucket_name_template.format(region=self.region, account_id=self.account_id)
+            bucket_name = bucket_name_template.format(region=self.region, account_id=self.account)
             bucket_arn = f"arn:aws:s3:::{bucket_name}"
             arns.append(bucket_arn)
             arns.append(f"{bucket_arn}/*")
         return arns
 
-    @staticmethod
-    def s3_list_policy_statement() -> iam.PolicyStatement:
-        """Create policy statement for listing S3 buckets
-
-        Returns:
-           iam.PolicyStatement: A policy statements for listing S3 buckets
-        """
-        list_all_buckets_policy = iam.PolicyStatement(
-            actions=["s3:ListAllMyBuckets"],
-            resources=["*"],  # ListAllMyBuckets applies to all buckets
-        )
-        return list_all_buckets_policy
-
     def s3_policy_statement(self) -> iam.PolicyStatement:
         """Create a policy statement for S3 access.
 
@@ -118,7 +104,7 @@ class WorkbenchCoreStack(Stack):
         """Allow passing the specific Glue role to AWS Glue."""
         return iam.PolicyStatement(
             actions=["iam:PassRole"],
-            resources=[f"arn:aws:iam::{self.account_id}:role/Workbench-GlueRole"],
+            resources=[f"arn:aws:iam::{self.account}:role/Workbench-GlueRole"],
             conditions={"StringEquals": {"iam:PassedToService": "glue.amazonaws.com"}},
         )
 
@@ -143,13 +129,11 @@ class WorkbenchCoreStack(Stack):
                 "glue:CreateJob",
                 "glue:UpdateJob",
                 "glue:StartJobRun",
-                "glue:StopJobRun",
-                "glue:DeleteJob",
                 "glue:CreateTrigger",
             ],
             resources=[
-                f"arn:aws:glue:{self.region}:{self.account_id}:job/*",
-                f"arn:aws:glue:{self.region}:{self.account_id}:trigger/*",
+                f"arn:aws:glue:{self.region}:{self.account}:job/*",
+                f"arn:aws:glue:{self.region}:{self.account}:trigger/*",
             ],
         )
 
@@ -235,7 +219,7 @@ class WorkbenchCoreStack(Stack):
                 "events:DescribeEventBus",
             ],
             resources=[
-                f"arn:aws:events:{self.region}:{self.account_id}:event-bus/workbench",
+                f"arn:aws:events:{self.region}:{self.account}:event-bus/workbench",
             ],
         )
 
@@ -691,26 +675,9 @@ class WorkbenchCoreStack(Stack):
             resources=["*"],  # Broad permission necessary for Parameter Store operations
         )
 
-    def secrets_read_policy_statement(self) -> iam.PolicyStatement:
-        """Create a policy statement for accessing AWS Secrets Manager.
-
-        Returns:
-            iam.PolicyStatement: The policy statement for accessing AWS Secrets Manager.
-        """
-        return iam.PolicyStatement(
-            actions=[
-                # Read only policy for Secrets Manager
-                "secretsmanager:ListSecrets",
-                "secretsmanager:DescribeSecret",
-                "secretsmanager:GetSecretValue",
-            ],
-            resources=["*"],  # We can/should narrow this down
-        )
-
     def workbench_datasource_policy(self) -> iam.ManagedPolicy:
         """Create a managed policy for the Workbench DataSources"""
         policy_statements = [
-            self.s3_list_policy_statement(),
             self.s3_policy_statement(),
             self.s3_public_policy_statement(),
             self.glue_job_read_policy(),
@@ -722,7 +689,6 @@ class WorkbenchCoreStack(Stack):
             self.athena_workgroup_policy_statement(),
             self.parameter_store_policy_statement(),
             self.eventbridge_policy(),
-            # self.secrets_read_policy_statement(),
         ]
 
         return iam.ManagedPolicy(
@@ -829,13 +795,15 @@ class WorkbenchCoreStack(Stack):
         # If sso_group is provided, add the condition to the trust relationship
         if self.sso_group:
             sso_group_arn_1 = (
-                f"arn:aws:iam::{self.account_id}:role/aws-reserved/sso.amazonaws.com/AWSReservedSSO_{self.sso_group}_*"
+                f"arn:aws:iam::{self.account}:role/aws-reserved/sso.amazonaws.com/AWSReservedSSO_{self.sso_group}_*"
+            )
+            sso_group_arn_2 = (
+                f"arn:aws:iam::{self.account}:role/aws-reserved/sso.amazonaws.com/*/AWSReservedSSO_{self.sso_group}_*"
             )
-            sso_group_arn_2 = f"arn:aws:iam::{self.account_id}:role/aws-reserved/sso.amazonaws.com/*/AWSReservedSSO_{self.sso_group}_*"
             condition = {"ArnLike": {"aws:PrincipalArn": [sso_group_arn_1, sso_group_arn_2]}}
-            assumed_by.add_principals(iam.AccountPrincipal(self.account_id).with_conditions(condition))
+            assumed_by.add_principals(iam.AccountPrincipal(self.account).with_conditions(condition))
         else:
-            assumed_by.add_principals(iam.AccountPrincipal(self.account_id))
+            assumed_by.add_principals(iam.AccountPrincipal(self.account))
 
         # Create the role with the trust relationships
         api_execution_role = iam.Role(

{workbench-0.8.156 → workbench-0.8.157}/aws_setup/workbench_dashboard_full/app.py

@@ -12,7 +12,7 @@ print(f"Account: {aws_account}")
 print(f"Region: {aws_region}")
 
 # When you want a different docker image change this line
-dashboard_image = "public.ecr.aws/m6i5k1r2/workbench_dashboard:v0_8_153_amd64"
+dashboard_image = "public.ecr.aws/m6i5k1r2/workbench_dashboard:v0_8_156_amd64"
 
 # Workbench Configuration
 try:

{workbench-0.8.156 → workbench-0.8.157}/aws_setup/workbench_dashboard_lite/app.py

@@ -13,7 +13,7 @@ print(f"Account: {aws_account}")
 print(f"Region: {aws_region}")
 
 # When you want a different docker image change this line
-dashboard_image = "public.ecr.aws/m6i5k1r2/workbench_dashboard:v0_8_153_amd64"
+dashboard_image = "public.ecr.aws/m6i5k1r2/workbench_dashboard:v0_8_156_amd64"
 
 # Workbench Configuration
 try:

{workbench-0.8.156 → workbench-0.8.157}/src/workbench/core/cloud_platform/aws/aws_account_clamp.py

@@ -105,12 +105,22 @@ class AWSAccountClamp:
             self.log.critical("AWS Identity Check Failure: Check AWS_PROFILE and/or Renew SSO Token...")
             return info
 
-    def check_s3_access(self) -> bool:
-        s3 = self.boto3_session.client("s3")
-        results = s3.list_buckets()
-        for bucket in results["Buckets"]:
-            self.log.info(f"\t{bucket['Name']}")
-        return True
+    def check_workbench_bucket(self) -> bool:
+        """Check if the Workbench S3 Bucket exists and is accessible"""
+        s3 = self.boto3_session.client("s3")  # Use client, not resource
+        try:
+            s3.head_bucket(Bucket=self.workbench_bucket_name)
+            self.log.info(f"The {self.workbench_bucket_name} bucket exists and is accessible")
+            return True
+        except ClientError as e:
+            error_code = e.response["Error"]["Code"]
+            if error_code == "404":
+                self.log.critical(f"The {self.workbench_bucket_name} bucket does not exist")
+            elif error_code == "403":
+                self.log.critical(f"Access denied to {self.workbench_bucket_name} bucket")
+            else:
+                self.log.error(f"Error checking S3 bucket: {e}")
+            return False
 
     def sagemaker_session(self) -> "SageSession":
         """Create a workbench SageMaker session (using our boto3 refreshable session)
@@ -145,9 +155,9 @@ if __name__ == "__main__":
     aws_account_clamp.check_assumed_role()
     print("Assumed Role Check Success...")
 
-    print("\n\n*** AWS S3 Access Check ***")
-    aws_account_clamp.check_s3_access()
-    print("S3 Access Check Success...")
+    print("\n\n*** AWS Workbench Bucket Check ***")
+    aws_account_clamp.check_workbench_bucket()
+    print("Workbench Bucket Check Success...")
 
     print("\n\n*** AWS Sagemaker Session/Client Check ***")
     sm_client = aws_account_clamp.sagemaker_client()

{workbench-0.8.156 → workbench-0.8.157}/src/workbench/core/cloud_platform/aws/aws_df_store.py

@@ -192,7 +192,7 @@ class AWSDFStore:
         # Update/Insert the DataFrame to S3
         s3_uri = self._generate_s3_uri(location)
         try:
-            wr.s3.to_parquet(df=data, path=s3_uri, dataset=True, mode="overwrite")
+            wr.s3.to_parquet(df=data, path=s3_uri, dataset=True, mode="overwrite", index=True)
             self.log.info(f"Dataframe cached {s3_uri}...")
         except Exception as e:
             self.log.error(f"Failed to cache dataframe '{s3_uri}': {e}")

{workbench-0.8.156 → workbench-0.8.157}/src/workbench/core/transforms/features_to_model/features_to_model.py

@@ -167,6 +167,9 @@ class FeaturesToModel(Transform):
         if self.custom_script:
             script_path = self.custom_script
             if self.custom_script.endswith(".template"):
+                # Model Type is an enumerated type, so we need to convert it to a string
+                template_params["model_type"] = template_params["model_type"].value
+
                 # Fill in the custom script template with specific parameters (include any custom args)
                 template_params.update(self.custom_args)
                 script_path = fill_template(self.custom_script, template_params, "generated_model_script.py")
@@ -316,10 +319,10 @@ if __name__ == "__main__":
 
     # Regression Model
     input_name = "abalone_features"
-    output_name = "abalone-regression"
+    output_name = "test-abalone-regression"
     to_model = FeaturesToModel(input_name, output_name, model_type=ModelType.REGRESSOR)
-    to_model.set_output_tags(["abalone", "public"])
-    to_model.transform(target_column="class_number_of_rings", description="Abalone Regression")
+    to_model.set_output_tags(["test"])
+    to_model.transform(target_column="class_number_of_rings", description="Test Abalone Regression")
 
     """
     # Classification Model

workbench-0.8.157/src/workbench/model_scripts/custom_models/nn_models/Readme.md

@@ -0,0 +1,9 @@
+# Neural Network Model Scripts
+Welcome to the set of neural network model scripts.
+Right now we just have:
+
+- PyTorch
+- ChemProp
+
+
+### References