Fourth pass at type fixes

src/ontogpt/engines/enrichment.py

@@ -59,10 +59,10 @@ class EnrichmentPayload(BaseModel):
     summary: str = ""
     """The summary of the gene set generated by the LLM."""
 
-    term_strings: List[str] = []
+    term_strings: List[str] = [""]
     """The raw term labels parsed from the LLM completion payload"""
 
-    term_ids: List[str] = []
+    term_ids: List[str] = [""]
     """The normalized terms"""
 
     ontological_synopsis: bool = None
@@ -74,7 +74,7 @@ class EnrichmentPayload(BaseModel):
     annotations: bool = None
     """True if the gene descriptions used the annotations (vs latent KB)"""
 
-    response_token_length: int = None
+    response_token_length: int = 0
     """The number of tokens in the response text"""
 
     model: str = ""

src/ontogpt/engines/halo_engine.py

@@ -40,9 +40,9 @@
 
 
 class StructuredPrompt(pydantic.BaseModel):
-    header: str = None
-    body: str = None
-    main_prompt: str = None
+    header: str = ""
+    body: str = ""
+    main_prompt: str = ""
 
     @property
     def text(self) -> str:
@@ -58,8 +58,8 @@ class HALOEngine(KnowledgeEngine):
     traverse_slots: List[FIELD] = field(
         default_factory=lambda: ["subtypes", "parts", "subclass_of", "part_of"]
     )
-    fixed_slot_values: Dict[str, str] = None
-    adapter: OboGraphInterface = None
+    fixed_slot_values: Optional[Dict[str, str]] = None
+    adapter: OboGraphInterface
     visited: Set[ELEMENT_NAME] = field(default_factory=lambda: set())
     candidates: List[ELEMENT_NAME] = None
     always_extend: bool = False

src/ontogpt/engines/hfhub_engine.py

@@ -265,7 +265,7 @@ def _raw_extract(self, text, cls: ClassDefinition = None, object: OBJECT = None)
         return payload
 
     def get_completion_prompt(
-        self, cls: ClassDefinition = None, text: str = None, object: OBJECT = None
+        self, cls: ClassDefinition = None, text: str = "", object: OBJECT = None
     ) -> str:
         """Get the prompt for the given template."""
         if cls is None:

src/ontogpt/engines/pheno_engine.py

@@ -25,13 +25,13 @@
 
 class DiagnosisPrediction(BaseModel):
     case_id: str
-    validated_disease_ids: List[str] = None
-    validated_disease_labels: List[str] = None
-    validated_mondo_disease_ids: List[str] = None
-    validated_mondo_disease_labels: List[str] = None
-    predicted_disease_ids: List[str] = None
-    predicted_disease_labels: List[str] = None
-    matching_disease_ids: List[str] = None
+    validated_disease_ids: Optional[List[str]] = None
+    validated_disease_labels: Optional[List[str]] = None
+    validated_mondo_disease_ids: Optional[List[str]] = None
+    validated_mondo_disease_labels: Optional[List[str]] = None
+    predicted_disease_ids: Optional[List[str]] = None
+    predicted_disease_labels: Optional[List[str]] = None
+    matching_disease_ids: Optional[List[str]] = None
     rank: Optional[int] = None
     model: Optional[str] = None
     prompt: Optional[str] = None

src/ontogpt/evaluation/ctd/eval_ctd.py

@@ -22,7 +22,7 @@
 from dataclasses import dataclass
 from pathlib import Path
 from random import shuffle
-from typing import Any, Dict, Iterable, List, Set, Tuple
+from typing import Any, Dict, Iterable, List, Optional, Set, Tuple
 
 import yaml
 from bioc import biocxml
@@ -59,12 +59,12 @@ class PredictionRE(BaseModel):
     test_object: TextWithTriples = None
     """source of truth to evaluate against"""
 
-    true_positives: List[Tuple] = None
-    num_true_positives: int = None
-    false_positives: List[Tuple] = None
-    num_false_positives: int = None
-    false_negatives: List[Tuple] = None
-    num_false_negatives: int = None
+    true_positives: Optional[List[Tuple]] = None
+    num_true_positives: Optional[int] = None
+    false_positives: Optional[List[Tuple]] = None
+    num_false_positives: Optional[int] = None
+    false_negatives: Optional[List[Tuple]] = None
+    num_false_negatives: Optional[int] = None
     scores: Dict[str, SimilarityScore] = None
     predicted_object: TextWithTriples = None
     named_entities: List[Any] = None

src/ontogpt/evaluation/enrichment/eval_enrichment.py

@@ -58,13 +58,13 @@ class Overlap(BaseModel):
     jaccard: Optional[float] = None
     common: Optional[List[str]] = None
     overlap_score: Optional[int] = None
-    left_jaccard: float = None
-    right_jaccard: float = None
-    summary_jaccard: float = None
+    left_jaccard: Optional[float] = None
+    right_jaccard: Optional[float] = None
+    summary_jaccard: Optional[float] = None
 
 
 class GeneSetComparison(BaseModel):
-    name: str
+    name: str = ""
     gene_symbols: List[str]
     gene_ids: List[str] = None
     model: str = ""

src/ontogpt/evaluation/hpoa/eval_hpoa.py

@@ -56,14 +56,14 @@ class EvaluationObjectSetHPOA(BaseModel):
 
     test: List[MendelianDisease] = None
     training: List[MendelianDisease] = None
-    predictions: List = None
+    predictions: List = []
 
 
 class HPOAnnotation(BaseModel):
-    subject: DISEASE_ID = None
-    term: TERM = None
-    aspect: ASPECT = None
-    publication: PUBLICATION = None
+    subject: DISEASE_ID = ""
+    term: TERM = ""
+    aspect: ASPECT = ""
+    publication: PUBLICATION = ""
 
 
 @dataclass

src/ontogpt/io/html_exporter.py

@@ -19,7 +19,7 @@ class HTMLExporter(Exporter):
     TODO: rewrite to use bootstrap
     """
 
-    output: TextIO = None
+    output: TextIO
 
     def export(self, extraction_output: ExtractionResult, output: Union[str, Path, TextIO]):
         if isinstance(output, Path):

src/ontogpt/ontex/extractor.py

@@ -80,8 +80,8 @@ class Ontology(BaseModel):
     axioms: List[Axiom]
     """All axioms in the ontology"""
 
-    terms: List[CURIE] = None
-    predicates: List[PRED_CURIE] = None
+    terms: List[CURIE] = [""]
+    predicates: List[PRED_CURIE] = [""]
 
     comments: Optional[List[str]] = None
 
@@ -196,7 +196,6 @@ class Task(BaseModel):
     query: Query = None
     answers: Optional[List[Answer]] = None
     examples: Optional[List[Example]] = None
-    description: Optional[str] = None
     obfuscated: Optional[bool] = False
 
     method: Optional[GPTReasonMethodType] = None