re-integrate docker settings and cleanup config.yaml

Author: marwan37

omni-reader/configs/config.yaml

@@ -15,18 +15,6 @@ parameters:
   input_image_paths: []
   selected_models: ["gpt4o", "granite", "llava34b", "pixtral", "gemma3", "llava-phi3"]
 
-# Pipeline settings
-settings:
-  docker:
-    requirements:
-      - "litellm>=1.10.0"
-      - "openai>=1.10.0"
-      - "pillow>=10.0.0"
-      - "ollama>=0.1.0"
-      - "jiwer>=3.0.0"
-      - "textdistance>=4.6.0"
-      - "polars>=0.19.19"
-
 # Component-specific configurations
 steps:
   run_ocr:

omni-reader/pipelines/batch_pipeline.py

@@ -15,10 +15,12 @@
 # limitations under the License.
 """OCR Batch Pipeline implementation for processing images with multiple models."""
 
+import os
 from typing import Any, Dict, List, Optional
 
 from dotenv import load_dotenv
 from zenml import pipeline
+from zenml.config import DockerSettings
 from zenml.logger import get_logger
 
 from steps import (
@@ -31,8 +33,16 @@
 
 logger = get_logger(__name__)
 
+docker_settings = DockerSettings(
+    requirements_file="requirements.txt",
+    environment={
+        "OPENAI_API_KEY": os.getenv("OPENAI_API_KEY"),
+        "MISTRAL_API_KEY": os.getenv("MISTRAL_API_KEY"),
+    },
+)
+
 
-@pipeline
+@pipeline(settings={"docker": docker_settings})
 def ocr_batch_pipeline(
     image_paths: Optional[List[str]] = None,
     image_folder: Optional[str] = None,

omni-reader/pipelines/evaluation_pipeline.py

@@ -15,11 +15,13 @@
 # limitations under the License.
 """OCR Evaluation Pipeline implementation for comparing models using existing results."""
 
+import os
 from typing import Any, Dict, List, Optional
 
 import polars as pl
 from dotenv import load_dotenv
 from zenml import pipeline, step
+from zenml.config import DockerSettings
 from zenml.logger import get_logger
 
 from steps import (
@@ -33,8 +35,16 @@
 
 logger = get_logger(__name__)
 
+docker_settings = DockerSettings(
+    requirements_file="requirements.txt",
+    environment={
+        "OPENAI_API_KEY": os.getenv("OPENAI_API_KEY"),
+        "MISTRAL_API_KEY": os.getenv("MISTRAL_API_KEY"),
+    },
+)
+
 
-@pipeline
+@pipeline(settings={"docker": docker_settings})
 def ocr_evaluation_pipeline(
     model_names: List[str] = None,
     results_dir: str = "ocr_results",