fix ci, formatting, flake8, and types

.flake8

@@ -14,6 +14,8 @@ extend-ignore =
 		E501,
 		# Stop finding commented out code because it's mistaking shape annotations for code
 		E800,
+		# Don't complain about asserts
+		S101,
 		# Stop complaining about subprocess, we need it for this project
 		S404,S602,S603,S607,
 		# Stop complaining about using functions from random

.pre-commit-config.yaml

@@ -5,7 +5,7 @@ default_language_version:
 repos:
   # -------------------------- Version control checks -------------------------- #
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.4.0
+    rev: v4.5.0
     hooks:
       - id: check-merge-conflict
         name: Check for merge conflicts
@@ -21,15 +21,15 @@ repos:
         name: Check for destroyed symlinks
 
   - repo: https://github.com/sirosen/check-jsonschema
-    rev: 0.22.0
+    rev: 0.27.2
     hooks:
       - id: check-github-workflows
         name: Validate GitHub workflows
         types: [yaml]
 
   # ----------------------------- Check file issues ---------------------------- #
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.4.0
+    rev: v4.5.0
     hooks:
       - id: check-toml
         name: Check TOML
@@ -53,7 +53,7 @@ repos:
 
   # ------------------------------ Python checking ----------------------------- #
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.4.0
+    rev: v4.5.0
     hooks:
       - id: debug-statements
         name: Check for debugger statements
@@ -77,19 +77,19 @@ repos:
 
   # ----------------------------- Automatic linters ---------------------------- #
   - repo: https://github.com/asottile/pyupgrade
-    rev: v3.3.1
+    rev: v3.15.0
     hooks:
       - id: pyupgrade
         name: Update syntax for newer Python
         types: [python]
         args: ["--py39-plus"]
   - repo: https://github.com/sirosen/texthooks
-    rev: 0.5.0
+    rev: 0.6.3
     hooks:
       - id: fix-smartquotes
         name: Fix Smart Quotes
   - repo: https://github.com/asottile/yesqa
-    rev: v1.4.0
+    rev: v1.5.0
     hooks:
       - id: yesqa
         name: Remove unnecessary `# noqa` comments
@@ -98,7 +98,7 @@ repos:
 
   # ------------------------------ Python imports ------------------------------ #
   - repo: https://github.com/hakancelik96/unimport
-    rev: 0.16.0
+    rev: 1.1.0
     hooks:
       - id: unimport
         name: Remove any unused imports
@@ -128,20 +128,20 @@ repos:
 
   # -------------------------------- Formatting -------------------------------- #
   - repo: https://github.com/pre-commit/mirrors-prettier
-    rev: v3.0.0-alpha.6
+    rev: v3.1.0
     hooks:
       - id: prettier
         name: Prettier
         exclude: ^.*/?CHANGELOG\.md$
   - repo: https://github.com/myint/docformatter
-    rev: v1.6.1
+    rev: v1.7.5
     hooks:
       - id: docformatter
         name: Format docstrings
         types: [python]
         args: [--in-place, --wrap-summaries=99, --wrap-descriptions=99]
   - repo: https://github.com/psf/black
-    rev: 23.3.0
+    rev: 23.11.0
     hooks:
       - id: black-jupyter
         types: [python]

src/emma_experience_hub/api/clients/simbot/session_db.py

@@ -122,4 +122,4 @@ def _get_all_session_turns(self, session_id: str) -> list[dict[str, Any]]:
 
             all_response_items.extend(response["Items"])
 
-        return all_response_items
+        return all_response_items  # type: ignore[unreachable]

src/emma_experience_hub/commands/build_emma.py

@@ -16,10 +16,10 @@ def torch_cuda_version_callback(torch_version: Optional[str]) -> str:
 
     try:
         int(torch_version.split("+cu")[-1])
-    except ValueError:
+    except ValueError as err:
         raise typer.BadParameter(
             "The cuda version must be castable to an integer. For example, `113` and not `11.3`"
-        )
+        ) from err
 
     return torch_version
 

src/emma_experience_hub/commands/teach/dataset.py

@@ -90,7 +90,7 @@ def filter_edh_instances(
         edh_instances = [
             instance
             for instance in edh_instances
-            if any([action.obj_interaction_action for action in instance.driver_actions_future])
+            if any(action.obj_interaction_action for action in instance.driver_actions_future)
         ]
         console.log(
             f"{num_instances_before_filter - len(edh_instances)} EDH instances [cyan]do not have an interaction action[/] in their future. {len(edh_instances)} EDH instances remaining..."

src/emma_experience_hub/datamodels/simbot/feedback.py

@@ -177,7 +177,7 @@ def calculate_rule_score(cls, score: int, values: dict[str, Any]) -> int:  # noq
     def is_query_suitable(self, query: dict[str, Any]) -> bool:
         """Evaluate the rule given the query and ensure it is suitable."""
         try:
-            return self.rule.matches(query) and all([name in query for name in self.slot_names])
+            return self.rule.matches(query) and all(name in query for name in self.slot_names)
         except Exception:
             return False
 
@@ -323,7 +323,7 @@ def from_all_information(
             interaction_action.payload, SimBotObjectInteractionPayload
         )
         if interaction_action_has_bbox:
-            object_area = get_area_from_compressed_mask(interaction_action.payload.object.mask)
+            object_area = get_area_from_compressed_mask(interaction_action.payload.object.mask)  # type: ignore[union-attr]
 
         return cls(
             # Require a lightweight dialog action when the model does not decode a <stop token

src/emma_experience_hub/datamodels/simbot/session.py

@@ -613,7 +613,7 @@ def try_to_update_agent_inventory(self) -> None:
                 viewpoint=self.previous_valid_turn.environment.get_closest_viewpoint_name(),
                 action=self.previous_valid_turn.actions.interaction,
                 inventory_entity=self.previous_valid_turn.state.inventory.entity,
-                action_history=[turn.actions.interaction for turn in past_turns],  # type: ignore[union-attr]
+                action_history=[turn.actions.interaction for turn in past_turns],  # type: ignore[misc]
                 inventory_history=[turn.state.inventory.entity for turn in past_turns],
             )
         self.current_state.memory.update_interaction_turn_index(

src/emma_experience_hub/parsers/simbot/action_predictor_output.py

@@ -64,9 +64,9 @@ def __call__(
 
         try:
             decoded_action = self._separate_decoded_trajectory(decoded_trajectory)[0]
-        except IndexError:
+        except IndexError as err:
             # If there is a problem when decoding the action
-            raise IndexError("Could not decode any actions from the trajectory")
+            raise IndexError("Could not decode any actions from the trajectory") from err
 
         # Just use the first action, because if that is wrong, any future ones after it are likely
         # hallucinated

src/emma_experience_hub/parsers/simbot/qa_output.py

@@ -84,14 +84,14 @@ def _process_intent(self, utterance: str, intent: str, entities: list[Any]) -> s
 
         return intent
 
-    def _is_regex_match_for_incomplete_utterance(self, utterance) -> bool:
+    def _is_regex_match_for_incomplete_utterance(self, utterance: str) -> bool:
         if re.search(self._incomplete_utterance_regex_pattern, utterance):
             logger.debug(f"found incomplete regex match for the utterance: {utterance}")
             return True
         logger.debug(f"No incomplete regex match for the utterance: {utterance}")
         return False
 
-    def _regex_patterns_map(self) -> bool:
+    def _regex_patterns_map(self) -> str:
         verbs = [
             "find",
             "search",

src/emma_experience_hub/pipelines/simbot/agent_intent_selection.py

@@ -95,7 +95,7 @@ def run(self, session: SimBotSession) -> SimBotAgentIntents:  # noqa: WPS212
 
             # check for object-qa and return
             if session.current_turn.intent.user.is_user_qa_about_object:
-                return self.handle_object_qa_intent(session)
+                return self.handle_object_qa_intent(session)  # type: ignore[return-value]
 
             # If we have received an invalid utterance, the agent does not act
             should_skip_action_selection = self._should_skip_action_selection(

src/emma_experience_hub/pipelines/simbot/find_object.py

@@ -206,6 +206,7 @@ def _should_scan_found_object(
             class_label=found_object_label,
             extracted_features=extracted_features,
         )
+        assert scene_object_tokens.object_index is not None
 
         object_idx = scene_object_tokens.object_index - 1
         area = extracted_features[frame_idx - 1].bbox_areas[object_idx].item()

src/emma_experience_hub/pipelines/simbot/user_utterance_verification.py

@@ -103,7 +103,7 @@ def _utterance_only_contains_wake_word(
         self, speech_recognition_payload: SimBotSpeechRecognitionPayload
     ) -> bool:
         """Detect whether the utterance only contains the wake word or not."""
-        return all([token.is_wake_word for token in speech_recognition_payload.tokens])
+        return all(token.is_wake_word for token in speech_recognition_payload.tokens)
 
     @tracer.start_as_current_span("Check for empty utterance")
     def _utterance_is_empty(

tests/pipelines/simbot/test_language_generator.py

@@ -59,7 +59,7 @@ def test_response_slots_in_all_rules(rule_parser: SimBotFeedbackFromSessionState
     ]
     for rule_text, slots in rules_and_slots:
         rule_words = rule_text.split()
-        assert all([slot_name in rule_words for slot_name in slots])
+        assert all(slot_name in rule_words for slot_name in slots)
 
 
 def test_all_response_slots_are_validated_by_rules(
@@ -71,10 +71,10 @@ def test_all_response_slots_are_validated_by_rules(
     ]
     for rule_text, slots in rules_and_slots:
         # Each slot name must be used in the rule so that it exists in some way
-        assert all([slot_name in rule_text for slot_name in slots])
+        assert all(slot_name in rule_text for slot_name in slots)
 
         # Any slot name in the rule must not be checking for it to be equal to None
-        assert not any([f"{slot_name} == null" in rule_text for slot_name in slots])
+        assert not any(f"{slot_name} == null" in rule_text for slot_name in slots)
 
 
 def test_all_rule_symbols_in_state(rule_parser: SimBotFeedbackFromSessionStateParser) -> None: