run formatting (#186)

Author: strickvl

databricks-production-qa-demo/steps/deployment/deployment_deploy.py

@@ -31,10 +31,14 @@
 
 
 @step(enable_cache=False)
-def deployment_deploy() -> Annotated[
-    Optional[DatabricksDeploymentService],
-    ArtifactConfig(name="databricks_deployment", is_deployment_artifact=True),
-]:
+def deployment_deploy() -> (
+    Annotated[
+        Optional[DatabricksDeploymentService],
+        ArtifactConfig(
+            name="databricks_deployment", is_deployment_artifact=True
+        ),
+    ]
+):
     """Predictions step.
 
     This is an example of a predictions step that takes the data in and returns

end-to-end-computer-vision/steps/export_label_studio.py

@@ -28,7 +28,6 @@
 logger = get_logger(__name__)
 
 
-
 @step(
     output_materializers={
         LABELED_DATASET_NAME: LabelStudioAnnotationMaterializer

eurorate-predictor/pipelines/training.py

@@ -24,7 +24,9 @@
 
 
 @pipeline
-def ecb_predictor_model_training_pipeline(augmented_dataset_id, mode: str = "develop"):
+def ecb_predictor_model_training_pipeline(
+    augmented_dataset_id, mode: str = "develop"
+):
     """A pipeline to train an XGBoost model and promote it.
 
     Args:

eurorate-predictor/run.py

@@ -101,7 +101,9 @@ def main(
         pipeline_args["config_path"] = os.path.join(
             config_folder, f"etl_{mode}.yaml"
         )
-        ecb_predictor_etl_pipeline.with_options(**pipeline_args)(**run_args_etl)
+        ecb_predictor_etl_pipeline.with_options(**pipeline_args)(
+            **run_args_etl
+        )
         logger.info("ETL pipeline finished successfully!\n")
 
     # Execute Feature Engineering Pipeline
@@ -126,9 +128,9 @@ def main(
         pipeline_args["config_path"] = os.path.join(
             config_folder, f"feature_engineering_{mode}.yaml"
         )
-        ecb_predictor_feature_engineering_pipeline.with_options(**pipeline_args)(
-            **run_args_feature
-        )
+        ecb_predictor_feature_engineering_pipeline.with_options(
+            **pipeline_args
+        )(**run_args_feature)
         logger.info("Feature Engineering pipeline finished successfully!\n")
 
     # Execute Model Training Pipeline
@@ -153,7 +155,9 @@ def main(
         pipeline_args["config_path"] = os.path.join(
             config_folder, f"training_{mode}.yaml"
         )
-        ecb_predictor_model_training_pipeline.with_options(**pipeline_args)(**run_args_train)
+        ecb_predictor_model_training_pipeline.with_options(**pipeline_args)(
+            **run_args_train
+        )
         logger.info("Model Training pipeline finished successfully!\n")
 
 

gamesense/steps/finetune.py

@@ -28,12 +28,11 @@
 from utils.loaders import load_base_model
 from utils.tokenizer import load_tokenizer
 from zenml import ArtifactConfig, step
+from zenml.client import Client
 from zenml.enums import ArtifactType
 from zenml.logger import get_logger
 from zenml.materializers import BuiltInMaterializer
 from zenml.utils.cuda_utils import cleanup_gpu_memory
-from zenml.client import Client
-
 
 logger = get_logger(__name__)
 

huggingface-sagemaker/steps/deploying/huggingface_deployment.py

@@ -47,9 +47,7 @@ def deploy_to_huggingface(
     save_model_to_deploy.entrypoint()
 
     logger.info("Model saved locally. Pushing to HuggingFace...")
-    assert secret, (
-        "No secret found with name 'huggingface_creds'. Please create one with your `token`."
-    )
+    assert secret, "No secret found with name 'huggingface_creds'. Please create one with your `token`."
 
     token = secret.secret_values["token"]
     api = HfApi(token=token)

huggingface-sagemaker/steps/promotion/promote_get_metrics.py

@@ -27,10 +27,12 @@
 
 
 @step
-def promote_get_metrics() -> Tuple[
-    Annotated[Dict[str, Any], "latest_metrics"],
-    Annotated[Dict[str, Any], "current_metrics"],
-]:
+def promote_get_metrics() -> (
+    Tuple[
+        Annotated[Dict[str, Any], "latest_metrics"],
+        Annotated[Dict[str, Any], "current_metrics"],
+    ]
+):
     """Get metrics for comparison for promoting a model.
 
     This is an example of a metric retrieval step. It is used to retrieve

llm-complete-guide/pipelines/llm_eval.py

@@ -17,7 +17,7 @@
 from typing import Optional
 
 import click
-from steps.create_prompt import PROMPT, create_prompt
+from steps.create_prompt import PROMPT
 from steps.eval_e2e import e2e_evaluation, e2e_evaluation_llm_judged
 from steps.eval_retrieval import (
     retrieval_evaluation_full,

llm-complete-guide/steps/create_prompt.py

@@ -24,6 +24,7 @@
 answers. \
 """
 
+
 @step
 def create_prompt() -> str:
     """Create a prompt for the RAG pipeline."""

llm-complete-guide/steps/eval_retrieval.py

@@ -275,9 +275,9 @@ def perform_small_retrieval_evaluation(use_reranking: bool) -> float:
 
 
 @step
-def retrieval_evaluation_small() -> Annotated[
-    float, "small_failure_rate_retrieval"
-]:
+def retrieval_evaluation_small() -> (
+    Annotated[float, "small_failure_rate_retrieval"]
+):
     """Executes the retrieval evaluation step without reranking.
 
     Returns:
@@ -287,9 +287,9 @@ def retrieval_evaluation_small() -> Annotated[
 
 
 @step
-def retrieval_evaluation_small_with_reranking() -> Annotated[
-    float, "small_failure_rate_retrieval_reranking"
-]:
+def retrieval_evaluation_small_with_reranking() -> (
+    Annotated[float, "small_failure_rate_retrieval_reranking"]
+):
     """Executes the retrieval evaluation step with reranking.
 
     Returns:

llm-complete-guide/steps/eval_visualisation.py

@@ -12,13 +12,13 @@
 #  or implied. See the License for the specific language governing
 #  permissions and limitations under the License.
 
-from typing import Annotated, Dict, List, Tuple
+from typing import Annotated, Dict, List
 
 import plotly.graph_objects as go
-from plotly.subplots import make_subplots
-from zenml import ArtifactConfig, get_step_context, log_metadata, step
+from zenml import get_step_context, log_metadata, step
 from zenml.types import HTMLString
 
+
 def create_plotly_bar_chart(
     labels: List[str],
     scores: List[float],
@@ -43,54 +43,66 @@ def create_plotly_bar_chart(
     """
     # Generate colors for bars
     if alternate_colors:
-        colors = ["rgba(66, 133, 244, 0.8)" if i % 2 == 0 else "rgba(219, 68, 55, 0.8)" for i in range(len(labels))]
+        colors = [
+            "rgba(66, 133, 244, 0.8)"
+            if i % 2 == 0
+            else "rgba(219, 68, 55, 0.8)"
+            for i in range(len(labels))
+        ]
     else:
         colors = ["rgba(66, 133, 244, 0.8)" for _ in range(len(labels))]
 
     # Prepare hover text
     if descriptions:
-        hover_text = [f"<b>{label}</b><br>Value: {score:.2f}<br>{descriptions.get(label, '')}" 
-                    for label, score in zip(labels, scores)]
+        hover_text = [
+            f"<b>{label}</b><br>Value: {score:.2f}<br>{descriptions.get(label, '')}"
+            for label, score in zip(labels, scores)
+        ]
     else:
-        hover_text = [f"<b>{label}</b><br>Value: {score:.2f}" for label, score in zip(labels, scores)]
+        hover_text = [
+            f"<b>{label}</b><br>Value: {score:.2f}"
+            for label, score in zip(labels, scores)
+        ]
 
     # Create figure
     fig = go.Figure()
-    
+
     fig.add_trace(
         go.Bar(
             y=labels,
             x=scores,
-            orientation='h',
+            orientation="h",
             marker_color=colors,
             text=[f"{score:.2f}" for score in scores],
-            textposition='auto',
+            textposition="auto",
             hovertext=hover_text,
-            hoverinfo='text',
+            hoverinfo="text",
         )
     )
 
     # Set layout
     max_value = max(scores) if scores else 5
-    xaxis_range = [0, 100] if percentage_scale else [0, max(5, max_value * 1.1)]
+    xaxis_range = (
+        [0, 100] if percentage_scale else [0, max(5, max_value * 1.1)]
+    )
     xaxis_title = "Percentage (%)" if percentage_scale else "Score"
-    
+
     fig.update_layout(
         title=title,
         xaxis=dict(
             title=xaxis_title,
             range=xaxis_range,
             showgrid=True,
-            gridcolor='rgba(230, 230, 230, 0.8)',
+            gridcolor="rgba(230, 230, 230, 0.8)",
         ),
         yaxis=dict(
             autorange="reversed",  # Make labels read top-to-bottom
         ),
         margin=dict(l=20, r=20, t=60, b=20),
         height=max(300, 70 * len(labels)),
-        plot_bgcolor='rgba(255, 255, 255, 1)',
+        plot_bgcolor="rgba(255, 255, 255, 1)",
     )
-    
+
     return fig
 
 
@@ -122,58 +134,49 @@ def generate_evaluation_html(
     """
     # Metric descriptions for hovering
     metric_descriptions = {
-        "Small Retrieval Eval Failure Rate": 
-            "Percentage of small test cases where retrieval failed to find relevant documents.",
-        "Small Retrieval Eval Failure Rate Reranking": 
-            "Percentage of small test cases where retrieval with reranking failed to find relevant documents.",
-        "Full Retrieval Eval Failure Rate": 
-            "Percentage of all test cases where retrieval failed to find relevant documents.",
-        "Full Retrieval Eval Failure Rate Reranking": 
-            "Percentage of all test cases where retrieval with reranking failed to find relevant documents.",
-        "Failure Rate Bad Answers": 
-            "Percentage of responses that were factually incorrect or misleading.",
-        "Failure Rate Bad Immediate Responses": 
-            "Percentage of immediate responses that did not adequately address the query.",
-        "Failure Rate Good Responses": 
-            "Percentage of responses rated as good by evaluators.",
-        "Average Toxicity Score": 
-            "Average score measuring harmful, offensive, or inappropriate content (lower is better).",
-        "Average Faithfulness Score": 
-            "Average score measuring how accurately the response represents the source material (higher is better).",
-        "Average Helpfulness Score": 
-            "Average score measuring the practical utility of responses to users (higher is better).",
-        "Average Relevance Score": 
-            "Average score measuring how well responses address the specific query intent (higher is better).",
+        "Small Retrieval Eval Failure Rate": "Percentage of small test cases where retrieval failed to find relevant documents.",
+        "Small Retrieval Eval Failure Rate Reranking": "Percentage of small test cases where retrieval with reranking failed to find relevant documents.",
+        "Full Retrieval Eval Failure Rate": "Percentage of all test cases where retrieval failed to find relevant documents.",
+        "Full Retrieval Eval Failure Rate Reranking": "Percentage of all test cases where retrieval with reranking failed to find relevant documents.",
+        "Failure Rate Bad Answers": "Percentage of responses that were factually incorrect or misleading.",
+        "Failure Rate Bad Immediate Responses": "Percentage of immediate responses that did not adequately address the query.",
+        "Failure Rate Good Responses": "Percentage of responses rated as good by evaluators.",
+        "Average Toxicity Score": "Average score measuring harmful, offensive, or inappropriate content (lower is better).",
+        "Average Faithfulness Score": "Average score measuring how accurately the response represents the source material (higher is better).",
+        "Average Helpfulness Score": "Average score measuring the practical utility of responses to users (higher is better).",
+        "Average Relevance Score": "Average score measuring how well responses address the specific query intent (higher is better).",
     }
 
     # Create individual charts
     retrieval_fig = create_plotly_bar_chart(
-        retrieval_labels, 
-        retrieval_scores, 
-        f"Retrieval Evaluation Metrics", 
+        retrieval_labels,
+        retrieval_scores,
+        f"Retrieval Evaluation Metrics",
         alternate_colors=True,
-        descriptions=metric_descriptions
+        descriptions=metric_descriptions,
     )
-    
+
     generation_basic_fig = create_plotly_bar_chart(
-        generation_basic_labels, 
-        generation_basic_scores, 
-        f"Basic Generation Metrics", 
+        generation_basic_labels,
+        generation_basic_scores,
+        f"Basic Generation Metrics",
         percentage_scale=True,
-        descriptions=metric_descriptions
+        descriptions=metric_descriptions,
     )
-    
+
     generation_quality_fig = create_plotly_bar_chart(
-        generation_quality_labels, 
-        generation_quality_scores, 
+        generation_quality_labels,
+        generation_quality_scores,
         f"Generation Quality Metrics",
-        descriptions=metric_descriptions
+        descriptions=metric_descriptions,
     )
 
     # Create summary metrics cards
     composite_quality = metrics_metadata.get("composite.overall_quality", 0)
-    retrieval_effectiveness = metrics_metadata.get("composite.retrieval_effectiveness", 0)
-    
+    retrieval_effectiveness = metrics_metadata.get(
+        "composite.retrieval_effectiveness", 0
+    )
+
     # Combine into complete HTML report
     html = f"""
     <!DOCTYPE html>
@@ -388,7 +391,7 @@ def generate_evaluation_html(
     </body>
     </html>
     """
-    
+
     return HTMLString(html)
 
 
@@ -434,10 +437,10 @@ def visualize_evaluation_results(
         + average_helpfulness_score
         + average_relevance_score
     ) / 3
-    
+
     composite_retrieval_effectiveness = (
-        (1 - small_retrieval_eval_failure_rate/100)
-        + (1 - full_retrieval_eval_failure_rate/100)
+        (1 - small_retrieval_eval_failure_rate / 100)
+        + (1 - full_retrieval_eval_failure_rate / 100)
     ) / 2
 
     # Collect all metrics for dashboard and logging

llm-complete-guide/steps/hf_dataset_loader.py

@@ -23,9 +23,9 @@
 
 
 @step(output_materializers=HFDatasetMaterializer)
-def load_hf_dataset() -> Tuple[
-    Annotated[Dataset, "train"], Annotated[Dataset, "test"]
-]:
+def load_hf_dataset() -> (
+    Tuple[Annotated[Dataset, "train"], Annotated[Dataset, "test"]]
+):
     train_dataset = load_dataset(DATASET_NAME_DEFAULT, split="train")
     test_dataset = load_dataset(DATASET_NAME_DEFAULT, split="test")
     return train_dataset, test_dataset

llm-complete-guide/steps/url_scraper.py

@@ -50,7 +50,7 @@ def url_scraper(
             "https://docs.zenml.io/stack-components/step-operators/azureml",
             # "https://docs.zenml.io/how-to/interact-with-secrets",
             # "https://docs.zenml.io/how-to/infrastructure-deployment/auth-management/service-connectors-guide",
-            # "https://docs.zenml.io/how-to/infrastructure-deployment/auth-management/hyperai-service-connector", 
+            # "https://docs.zenml.io/how-to/infrastructure-deployment/auth-management/hyperai-service-connector",
             # "https://docs.zenml.io/stack-components/data-validators/evidently",
             # "https://docs.zenml.io/stack-components/data-validators",
             # "https://docs.zenml.io/stack-components/step-operators/sagemaker",