Refinery reduction (#267)

* Removes overview charts but label distribution

* Remove record IDE

* Removes ide reference from start

* Submodule merge & remove unused

Author: JWittmeyer

app.py

@@ -29,7 +29,6 @@
 from fast_api.routes.heuristic import router as heuristic_router
 from fast_api.routes.data_browser import router as data_browser_router
 from fast_api.routes.labeling import router as labeling_router
-from fast_api.routes.record_ide import router as record_ide_router
 from fast_api.routes.record import router as record_router
 from fast_api.routes.weak_supervision import router as weak_supervision_router
 from fast_api.routes.labeling_tasks import router as labeling_tasks_router
@@ -54,7 +53,6 @@
     PREFIX_HEURISTIC,
     PREFIX_DATA_BROWSER,
     PREFIX_LABELING,
-    PREFIX_RECORD_IDE,
     PREFIX_RECORD,
     PREFIX_WEAK_SUPERVISION,
     PREFIX_LABELING_TASKS,
@@ -100,9 +98,6 @@
     data_browser_router, prefix=PREFIX_DATA_BROWSER, tags=["data-browser"]
 )
 fastapi_app.include_router(labeling_router, prefix=PREFIX_LABELING, tags=["labeling"])
-fastapi_app.include_router(
-    record_ide_router, prefix=PREFIX_RECORD_IDE, tags=["record-ide"]
-),
 fastapi_app.include_router(record_router, prefix=PREFIX_RECORD, tags=["record"]),
 fastapi_app.include_router(
     weak_supervision_router, prefix=PREFIX_WEAK_SUPERVISION, tags=["weak-supervision"]

controller/project/manager.py

@@ -8,7 +8,6 @@
 from controller.labeling_access_link import manager as link_manager
 from submodules.model import Project, enums
 from submodules.model.business_objects import (
-    labeling_task,
     organization,
     project,
     record,
@@ -42,10 +41,6 @@ def get_project(project_id: str) -> Project:
     return project.get(project_id)
 
 
-def get_project_with_labeling_tasks(project_id: str) -> Project:
-    return project.get_with_labling_tasks(project_id)
-
-
 def get_project_with_labeling_tasks_info_attributes(project_id: str) -> Project:
     return project.get_with_labling_tasks_info_attributes(project_id)
 
@@ -109,10 +104,6 @@ def get_max_running_id(project_id: str) -> int:
     return project.get_max_running_id(project_id)
 
 
-def is_rats_tokenization_still_running(project_id: str) -> bool:
-    return project.is_rats_tokenization_still_running(project_id)
-
-
 def create_project(
     organization_id: str, name: str, description: str, user_id: str
 ) -> Project:
@@ -202,31 +193,6 @@ def get_label_distribution(
     return project.get_label_distribution(project_id, labeling_task_id, slice_id)
 
 
-def get_confidence_distribution(
-    project_id: str,
-    labeling_task_id: str,
-    slice_id: Optional[str] = None,
-    num_samples: Optional[int] = None,
-) -> str:
-    return project.get_confidence_distribution(
-        project_id, labeling_task_id, slice_id, num_samples
-    )
-
-
-def get_confusion_matrix(
-    project_id: str,
-    labeling_task_id: str,
-    slice_id: Optional[str] = None,
-) -> str:
-    for_classification = (
-        labeling_task.get(project_id, labeling_task_id).task_type
-        == enums.LabelingTaskType.CLASSIFICATION.value
-    )
-    return project.get_confusion_matrix(
-        project_id, labeling_task_id, for_classification, slice_id
-    )
-
-
 def resolve_request_huddle_data(
     project_id: str, user_id: str, data_id: str, huddle_type: str
 ) -> HuddleData:

controller/record_ide/__init__.py

@@ -107,10 +107,6 @@ class UploadCredentialsAndIdBody(BaseModel):
     key: Optional[StrictStr] = None
 
 
-class RecordIdeBody(BaseModel):
-    code: StrictStr
-
-
 class NotificationsBody(BaseModel):
     project_filter: List[StrictStr]
     level_filter: List[StrictStr]

controller/record_ide/manager.py

@@ -11,15 +11,13 @@
 )
 from fastapi import APIRouter, Body, Depends, Request
 from fast_api.routes.client_response import pack_json_result
-from typing import Dict, List
+from typing import Dict
 from controller.auth import manager as auth_manager
 from controller.attribute import manager as attr_manager
-from controller.labeling_task import manager as task_manager
 from controller.upload_task import manager as upload_task_manager
 from submodules.model.business_objects import information_source, labeling_task
-from submodules.model import enums, events
+from submodules.model import enums
 from submodules.model.business_objects.embedding import get_all_embeddings_by_project_id
-from submodules.model.enums import LabelingTaskType
 from submodules.model.business_objects.project import get_project_by_project_id_sql
 from submodules.model.business_objects.labeling_task import (
     get_labeling_tasks_by_project_id_full,
@@ -33,10 +31,6 @@
     to_frontend_obj_raw,
 )
 from util import notification
-from util.inter_annotator.functions import (
-    resolve_inter_annotator_matrix_classification,
-    resolve_inter_annotator_matrix_extraction,
-)
 from controller.misc import manager as misc
 from exceptions.exceptions import NotAllowedInOpenSourceError
 from submodules.model.business_objects import notification as notification_model
@@ -137,87 +131,6 @@ def general_project_stats(
     )
 
 
-@router.get(
-    "/{project_id}/inter-annotator-matrix",
-    dependencies=[Depends(auth_manager.check_project_access_dep)],
-)
-def inter_annotator_matrix(
-    project_id: str,
-    labeling_task_id: str,
-    include_gold_star: Optional[bool] = True,
-    include_all_org_user: Optional[bool] = False,
-    only_on_static_slice: Optional[str] = None,
-) -> Dict:
-
-    labeling_task = task_manager.get_labeling_task(project_id, labeling_task_id)
-    if not labeling_task:
-        raise ValueError("Can't match labeling task to given Ids")
-    fp = None
-    if labeling_task.task_type == LabelingTaskType.CLASSIFICATION.value:
-        fp = resolve_inter_annotator_matrix_classification
-    elif labeling_task.task_type == LabelingTaskType.INFORMATION_EXTRACTION.value:
-        fp = resolve_inter_annotator_matrix_extraction
-    else:
-        raise ValueError(f"Can't match task type {labeling_task.task_type}")
-
-    return pack_json_result(
-        {
-            "data": {
-                "interAnnotatorMatrix": fp(
-                    labeling_task,
-                    include_gold_star,
-                    include_all_org_user,
-                    only_on_static_slice,
-                )
-            }
-        },
-        wrap_for_frontend=False,  # not wrapped as the prepared results in snake_case are still the expected form the frontend
-    )
-
-
-@router.get(
-    "/{project_id}/confusion-matrix",
-    dependencies=[Depends(auth_manager.check_project_access_dep)],
-)
-def confusion_matrix(
-    project_id: str,
-    labeling_task_id: str,
-    slice_id: Optional[str] = None,
-) -> Dict:
-    return pack_json_result(
-        {
-            "data": {
-                "confusionMatrix": manager.get_confusion_matrix(
-                    project_id, labeling_task_id, slice_id
-                )
-            }
-        },
-        wrap_for_frontend=False,  # not wrapped as the prepared results in snake_case are still the expected form the frontend
-    )
-
-
-@router.get(
-    "/{project_id}/confidence-distribution",
-    dependencies=[Depends(auth_manager.check_project_access_dep)],
-)
-def confidence_distribution(
-    project_id: str,
-    labeling_task_id: Optional[str] = None,
-    slice_id: Optional[str] = None,
-    num_samples: int = 100,
-) -> List:
-    return pack_json_result(
-        {
-            "data": {
-                "confidenceDistribution": manager.get_confidence_distribution(
-                    project_id, labeling_task_id, slice_id, num_samples
-                )
-            }
-        },
-        wrap_for_frontend=False,  # not wrapped as the prepared results in snake_case are still the expected form the frontend
-    )
-
-
 @router.get(
     "/{project_id}/label-distribution",
     dependencies=[Depends(auth_manager.check_project_access_dep)],
@@ -386,19 +299,6 @@ def get_model_provider_info(request: Request) -> Dict:
     return pack_json_result({"data": {"modelProviderInfo": data}})
 
 
-@router.get(
-    "/{project_id}/rats-running",
-    dependencies=[Depends(auth_manager.check_project_access_dep)],
-)
-def is_rats_running(
-    request: Request,
-    project_id: str,
-) -> Dict:
-
-    data = manager.is_rats_tokenization_still_running(project_id)
-    return pack_json_result({"data": {"isRatsTokenizationStillRunning": data}})
-
-
 @router.get(
     "/{project_id}/last-export-credentials",
     dependencies=[Depends(auth_manager.check_project_access_dep)],

fast_api/models.py

@@ -12,7 +12,6 @@
 PREFIX_HEURISTIC = PREFIX + "/heuristic"
 PREFIX_DATA_BROWSER = PREFIX + "/data-browser"
 PREFIX_LABELING = PREFIX + "/labeling"
-PREFIX_RECORD_IDE = PREFIX + "/record-ide"
 PREFIX_RECORD = PREFIX + "/record"
 PREFIX_WEAK_SUPERVISION = PREFIX + "/weak-supervision"
 PREFIX_LABELING_TASKS = PREFIX + "/labeling-tasks"

fast_api/routes/project.py

@@ -54,7 +54,6 @@ docker run -d --rm \
 -e AC_EXEC_ENV_IMAGE=registry.dev.kern.ai/code-kern-ai/refinery-ac-exec-env:dev$IS_ARM64 \
 -e LF_EXEC_ENV_IMAGE=registry.dev.kern.ai/code-kern-ai/refinery-lf-exec-env:dev$IS_ARM64 \
 -e ML_EXEC_ENV_IMAGE=registry.dev.kern.ai/code-kern-ai/refinery-ml-exec-env:dev$IS_ARM64 \
--e RECORD_IDE_IMAGE=registry.dev.kern.ai/code-kern-ai/refinery-record-ide-env:dev$IS_ARM64 \
 -e LF_NETWORK=dev-setup_default \
 -e S3_ENDPOINT="http://$HOST_IP:7053" \
 -e S3_ENDPOINT_LOCAL=object-storage:9000 \