aiverify-moonshot 0.6.1__py3-none-any.whl → 0.6.2__py3-none-any.whl

{aiverify_moonshot-0.6.1.dist-info → aiverify_moonshot-0.6.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: aiverify-moonshot
-Version: 0.6.1
+Version: 0.6.2
 Summary: AI Verify advances Gen AI testing with Project Moonshot.
 Project-URL: Repository, https://github.com/aiverify-foundation/moonshot
 Project-URL: Documentation, https://aiverify-foundation.github.io/moonshot/
@@ -23,6 +23,7 @@ Requires-Dist: pandas>=2.2.2
 Requires-Dist: pydantic==2.8.2
 Requires-Dist: pyparsing>=3.1.4
 Requires-Dist: python-dotenv>=1.0.1
+Requires-Dist: python-multipart>=0.0.9
 Requires-Dist: python-slugify>=8.0.4
 Requires-Dist: tenacity>=8.5.0
 Requires-Dist: xxhash>=3.5.0
@@ -47,7 +48,7 @@ Description-Content-Type: text/markdown
 
 ![Moonshot Logo](https://github.com/aiverify-foundation/moonshot/raw/main/misc/aiverify-moonshot-logo.png)
 
-**Version 0.6.1**
+**Version 0.6.2**
 
 A simple and modular tool to evaluate any LLM application.
 

{aiverify_moonshot-0.6.1.dist-info → aiverify_moonshot-0.6.2.dist-info}/RECORD

@@ -1,5 +1,5 @@
 moonshot/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-moonshot/__main__.py,sha256=1lpqD3azEnA0wMwIJ0K6cJd5sQFl-v3M8M0ehr-wrAU,11801
+moonshot/__main__.py,sha256=5lD240TY2YfG_p0sQtoHsw9V4DWWI3otpzIYP1hUbC8,12056
 moonshot/api.py,sha256=wvad-BcKDKEu25c6-YrsBx_uPiLKIBRsbwgThT50Uh0,4877
 moonshot/integrations/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 moonshot/integrations/cli/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -34,7 +34,7 @@ moonshot/integrations/cli/utils/process_data.py,sha256=QVL5vp2_8ZgGicmCAdeYEHkeb
 moonshot/integrations/web_api/.env.dev,sha256=0z5_Ut8rF-UqFZtgjkH2qoqORhD5_nSs2w_OeX2SteI,182
 moonshot/integrations/web_api/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 moonshot/integrations/web_api/__main__.py,sha256=MdnLi_ZF-olAAEJwTPU1iGYFYwo-fNWNT2qfchkH3y4,2050
-moonshot/integrations/web_api/app.py,sha256=Jr6mYvfjiPKMUWU58QxvYS-bpvkUotd728t6up3ZS-w,3651
+moonshot/integrations/web_api/app.py,sha256=6EkH5LAwEE5iQhGGDl5TIXDv_mmgWlpXCgUM9fH0Uow,3651
 moonshot/integrations/web_api/container.py,sha256=DVkJG_qm7ItcG6tgMYOqIj07wpKhPWOOfy6-bEv72y4,5915
 moonshot/integrations/web_api/logging_conf.py,sha256=t3EGRV6tZhV732KXe8_Tiy0fiwVAWxZX5Tt8VTgrrfg,3388
 moonshot/integrations/web_api/log/.gitkeep,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
@@ -45,7 +45,7 @@ moonshot/integrations/web_api/routes/benchmark_result.py,sha256=WZ_dI8qT4dli9hKP
 moonshot/integrations/web_api/routes/bookmark.py,sha256=aHUT86Llbzqo1CT3Dy7ciIhxVEzu1YgZk_VkxVeOZ3s,6304
 moonshot/integrations/web_api/routes/context_strategy.py,sha256=kJTpjrwxfYGyBLY_hAgpHOMZMtjV5Z6vpu7RIdHDylg,4828
 moonshot/integrations/web_api/routes/cookbook.py,sha256=oddmcdfhgH3qZb4_ThfUk8SBKmHOt51dFlAHubQh2fQ,8648
-moonshot/integrations/web_api/routes/dataset.py,sha256=9o8K81xn-fKappgPU4pcRSnLUl2J79MyR5PZijY5ueI,7312
+moonshot/integrations/web_api/routes/dataset.py,sha256=YhX4xNEgI5KwbmQ-SsLc5SpD3-9EZ6-e2AxBh1QJnUc,8238
 moonshot/integrations/web_api/routes/endpoint.py,sha256=ZFx0WUe3-GGdmBz_hzYiOmjvJHN4PQy_8lCKJMBjxcE,10715
 moonshot/integrations/web_api/routes/metric.py,sha256=f_HHexxKUfqFE5FkeCwRh8n36H2mREtLnK2pDrw3A-w,2856
 moonshot/integrations/web_api/routes/prompt_template.py,sha256=M3adeNeWvLQJJlFQ0uZqSXEuNpTcagApnuqWvLiL1mg,4890
@@ -57,7 +57,7 @@ moonshot/integrations/web_api/schemas/benchmark_runner_dto.py,sha256=IIn6KeMcwxT
 moonshot/integrations/web_api/schemas/bookmark_create_dto.py,sha256=C78vG8UG02N7Cmt6RSuS8e4sX_G-MLCiAWT-cF5BE8s,374
 moonshot/integrations/web_api/schemas/cookbook_create_dto.py,sha256=wXC0tu1Q8SpSI3Qk0xKPj1vKsOJEYmfPgU4rl6QopUY,826
 moonshot/integrations/web_api/schemas/cookbook_response_model.py,sha256=COLvaE4Hrz_w-C_HQkB7feztweIr0wkY9h8N6NKNIr8,332
-moonshot/integrations/web_api/schemas/dataset_create_dto.py,sha256=GRqIIlQZEpzzEXwAFcbDlxOuKg0JZ399axBjg34LMp8,915
+moonshot/integrations/web_api/schemas/dataset_create_dto.py,sha256=923Etaq2J1-S9-Xvh4tcrMIHpRrJ381tA8tj0M2kj1A,907
 moonshot/integrations/web_api/schemas/dataset_response_dto.py,sha256=s5x4-UXEWccWhK42E0FPXiHG6VqjuFuph-2t5atEkg4,171
 moonshot/integrations/web_api/schemas/endpoint_create_dto.py,sha256=WS8AfRybrweoOgZx6K6jiNy1Z6J3IZS1PUNnrRxGKyM,678
 moonshot/integrations/web_api/schemas/endpoint_response_model.py,sha256=OmmM2uaPSgB2aqPFfkhseKkI5OKCKilXR19gDmwFlLc,321
@@ -80,7 +80,7 @@ moonshot/integrations/web_api/services/benchmarking_service.py,sha256=lJZeNTqxEP
 moonshot/integrations/web_api/services/bookmark_service.py,sha256=jI9nXs1hjzO0CLG2LKaXSzDApLThkfCvPUkaNNV9A5A,3546
 moonshot/integrations/web_api/services/context_strategy_service.py,sha256=6YKnnG8JlE_1nlnr4Hq7rgz-sxI6oQglK0STaWPFQxQ,710
 moonshot/integrations/web_api/services/cookbook_service.py,sha256=37iJZn4ybe9tugBWB99g1SAN1YUtkmaq2mLQWj_HBQo,8736
-moonshot/integrations/web_api/services/dataset_service.py,sha256=ZWb3FqyDkA0C9qhlQ3X_zR0ohAlwlLsJi-mgKLvXpnI,2407
+moonshot/integrations/web_api/services/dataset_service.py,sha256=FUXLgU32nghoLWWXBA_4GzeQb8eK31tjbvLu4OJBxoc,2441
 moonshot/integrations/web_api/services/endpoint_service.py,sha256=N5SXNAh44UNeBpMhA9baL0VZoTx4sHzpy4y7-Ch8O4E,2395
 moonshot/integrations/web_api/services/metric_service.py,sha256=xWC5Dk8aiU7tuHsxYedTTrEkbA3Ug1pV2nbaBas6cAg,456
 moonshot/integrations/web_api/services/prompt_template_service.py,sha256=5ds7pKDB2R0_0slVDwsCRIpIVdsgpqhI-3wQqSYcpuE,1226
@@ -103,7 +103,7 @@ moonshot/src/api/api_connector.py,sha256=Q_of-aHPuWkbefMJq4uXctJl89G2Tt6J_HfSuf1
 moonshot/src/api/api_connector_endpoint.py,sha256=lwfhlWNBJ6QotqffmURtjRmxfzbBlSIAZupeSpMt9VU,5584
 moonshot/src/api/api_context_strategy.py,sha256=uRIfNjKJ_Wk9nSrvbPRfrdQLpG0K6kH9rl5tmmHui40,2151
 moonshot/src/api/api_cookbook.py,sha256=V05abHvzElrO7LkSyhOMcAHEfCfIgopd6L0cSSO3Dro,6722
-moonshot/src/api/api_dataset.py,sha256=i2KwnZ-6fTm_tyn8cRw8iesrGi7_Nh0-1bFuN7m0TVo,4066
+moonshot/src/api/api_dataset.py,sha256=POpkrmo_vLeUPOkuEShS8eadXdRluzbvOlUQ1DhsNqM,5521
 moonshot/src/api/api_environment_variables.py,sha256=wRx6rm95ItyL_uKUAYfSjcPZNbRxKl1GGS4PpWcTE1s,712
 moonshot/src/api/api_metrics.py,sha256=x5DiysTYQsMmcAS2y2XpgvrPobZk7GT2rhO-MaIRun4,1603
 moonshot/src/api/api_prompt_template.py,sha256=HQUl7-HGcxA620cY0vDqqo7CoY9uONkXIOlolduIgbE,1959
@@ -128,7 +128,7 @@ moonshot/src/cookbooks/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3
 moonshot/src/cookbooks/cookbook.py,sha256=DdZwRGx5-xTDIKcXtZRpp7Qb9Mm9dNGwXWLQXoQrBBo,10412
 moonshot/src/cookbooks/cookbook_arguments.py,sha256=SmNG8D5qN2K2dcImDaSBPHsna0Gy60ZR49_eTKEsvVU,1445
 moonshot/src/datasets/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-moonshot/src/datasets/dataset.py,sha256=-_uhjR7zi50nkLu1WWlPCCWr14VwFUDfhTeeBHOhb70,14236
+moonshot/src/datasets/dataset.py,sha256=sAPPdUJTB_aAFw3lmp2CY9gqbF2Mu8qQ9Soc1syMNGg,14940
 moonshot/src/datasets/dataset_arguments.py,sha256=rUcxxo2WTcHhLLV-WoixjOfT_Ju7hFCq811_ctjegt8,1751
 moonshot/src/metrics/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 moonshot/src/metrics/metric.py,sha256=llqJYnwtllJRMfNhRRbKWjhzKymY961yR3Jw24COR-Y,7512
@@ -172,9 +172,9 @@ moonshot/src/utils/import_modules.py,sha256=T9zTN59PFnvY2rjyWhSV9KSIAHxWV1pyBemF
 moonshot/src/utils/log.py,sha256=YNgD7Eh2OT36XlmVBKCGUTAh9TRp4Akfe4kDdvHASgs,2502
 moonshot/src/utils/pagination.py,sha256=5seymyRoqyENIhKllAatr1T91kMCGFslcvRnJHyMSvc,814
 moonshot/src/utils/timeit.py,sha256=TvuF0w8KWhp0oZFY0cUU3UY0xlGKjchb0OkfYfgVTlc,866
-aiverify_moonshot-0.6.1.dist-info/METADATA,sha256=Um1dy4p7R1ZqYm9X_wnmzsVi2qclr6trbA11ijKYiRs,12419
-aiverify_moonshot-0.6.1.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-aiverify_moonshot-0.6.1.dist-info/licenses/AUTHORS.md,sha256=mmAbe3i3sT8JZHJMBhxp3i1xRehV0g7WB4T_eyIBuBs,59
-aiverify_moonshot-0.6.1.dist-info/licenses/LICENSE.md,sha256=53izDRmJZZCjpYGfyLqlxnGQN-aNWBxasuzuMXC5Ias,11347
-aiverify_moonshot-0.6.1.dist-info/licenses/NOTICES.md,sha256=vS1zZYAnGjCJdwQ13xv3b2zc30wOS98ZnCKluT-AhHs,123266
-aiverify_moonshot-0.6.1.dist-info/RECORD,,
+aiverify_moonshot-0.6.2.dist-info/METADATA,sha256=nijW6md0PES6huZ6YbdfwQkiD9BN6gH80TWXabbamFs,12458
+aiverify_moonshot-0.6.2.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+aiverify_moonshot-0.6.2.dist-info/licenses/AUTHORS.md,sha256=mmAbe3i3sT8JZHJMBhxp3i1xRehV0g7WB4T_eyIBuBs,59
+aiverify_moonshot-0.6.2.dist-info/licenses/LICENSE.md,sha256=53izDRmJZZCjpYGfyLqlxnGQN-aNWBxasuzuMXC5Ias,11347
+aiverify_moonshot-0.6.2.dist-info/licenses/NOTICES.md,sha256=vS1zZYAnGjCJdwQ13xv3b2zc30wOS98ZnCKluT-AhHs,123266
+aiverify_moonshot-0.6.2.dist-info/RECORD,,

moonshot/__main__.py

@@ -116,6 +116,13 @@ def download_nltk_resources() -> None:
             raise
 
 
+def download_spacy_model() -> None:
+    """
+    Downloads the en_core_web_lg model using the spacy module (for entity processor module).
+    """
+    subprocess.run(["python", "-m", "spacy", "download", "en_core_web_lg"])
+
+
 def moonshot_data_installation(unattended: bool, overwrite: bool) -> None:
     """
     Install Moonshot Data from GitHub.
@@ -175,10 +182,12 @@ def moonshot_data_installation(unattended: bool, overwrite: bool) -> None:
     if os.path.exists("requirements.txt"):
         run_subprocess(["pip", "install", "-r", "requirements.txt"], check=True)
         download_nltk_resources()
+        download_spacy_model()
 
     # Change back to the base directory
     os.chdir("..")
 
+
 def check_node() -> bool:
     """
     Check if Node.js is installed on the user's machine.

moonshot/integrations/web_api/app.py

@@ -71,7 +71,7 @@ def create_app(cfg: providers.Configuration) -> CustomFastAPI:
     }
 
     app: CustomFastAPI = CustomFastAPI(
-        title="Project Moonshot", version="0.6.1", **app_kwargs
+        title="Project Moonshot", version="0.6.2", **app_kwargs
     )
 
     if cfg.cors.enabled():

moonshot/integrations/web_api/routes/dataset.py

@@ -1,5 +1,8 @@
+import os
+import tempfile
+
 from dependency_injector.wiring import Provide, inject
-from fastapi import APIRouter, Depends, HTTPException
+from fastapi import APIRouter, Depends, File, Form, HTTPException, UploadFile
 
 from ..container import Container
 from ..schemas.dataset_create_dto import CSV_Dataset_DTO, HF_Dataset_DTO
@@ -10,10 +13,14 @@ from ..services.utils.exceptions_handler import ServiceException
 router = APIRouter(tags=["Datasets"])
 
 
-@router.post("/api/v1/datasets/csv")
+@router.post("/api/v1/datasets/file")
 @inject
-def convert_dataset(
-    dataset_data: CSV_Dataset_DTO,
+async def upload_dataset(
+    file: UploadFile = File(...),
+    name: str = Form(..., min_length=1),
+    description: str = Form(default="", min_length=1),
+    license: str = Form(default=""),
+    reference: str = Form(default=""),
     dataset_service: DatasetService = Depends(Provide[Container.dataset_service]),
 ) -> str:
     """
@@ -32,7 +39,24 @@ def convert_dataset(
                        An error with status code 400 if there is a validation error.
                        An error with status code 500 for any other server-side error.
     """
+
+    # Create a temporary file with a secure random name
+    with tempfile.NamedTemporaryFile(
+        delete=False, suffix=os.path.splitext(file.filename)[1]
+    ) as tmp_file:
+        content = await file.read()
+        tmp_file.write(content)
+        temp_file_path = tmp_file.name
+
     try:
+        # Create the DTO with the form data including optional fields
+        dataset_data = CSV_Dataset_DTO(
+            name=name,
+            description=description,
+            license=license,
+            reference=reference,
+            file_path=temp_file_path,
+        )
         return dataset_service.convert_dataset(dataset_data)
     except ServiceException as e:
         if e.error_code == "FileNotFound":
@@ -47,6 +71,10 @@ def convert_dataset(
             raise HTTPException(
                 status_code=500, detail=f"Failed to convert dataset: {e.msg}"
             )
+    finally:
+        # Clean up the temporary file
+        if os.path.exists(temp_file_path):
+            os.unlink(temp_file_path)
 
 
 @router.post("/api/v1/datasets/hf")

moonshot/integrations/web_api/schemas/dataset_create_dto.py

@@ -8,13 +8,13 @@ from moonshot.src.datasets.dataset_arguments import (
 
 
 class CSV_Dataset_DTO(DatasetPydanticModel):
-    id: Optional[str] = None  # Not a required from user
-    examples: Optional[Any] = None  # Not a required from user
+    id: Optional[str] = None  # Not required from user
+    examples: Optional[Any] = None  # Not required from user
     name: str = Field(..., min_length=1)
     description: str = Field(default="", min_length=1)
     license: Optional[str] = ""
     reference: Optional[str] = ""
-    csv_file_path: str = Field(..., min_length=1)
+    file_path: str = Field(..., min_length=1)
 
 
 class HF_Dataset_DTO(DatasetPydanticModel):

moonshot/integrations/web_api/services/dataset_service.py

@@ -4,6 +4,7 @@ from ..schemas.dataset_response_dto import DatasetResponseDTO
 from ..services.base_service import BaseService
 from ..services.utils.exceptions_handler import exception_handler
 from .utils.file_manager import copy_file
+import os
 
 
 class DatasetService(BaseService):
@@ -16,7 +17,7 @@ class DatasetService(BaseService):
             dataset_data (CSV_Dataset_DTO): The data required to convert the dataset.
 
         Returns:
-            str: The path to the newly created dataset.
+            str: The filename of the newly created dataset.
 
         Raises:
             Exception: If an error occurs during dataset conversion.
@@ -27,9 +28,9 @@ class DatasetService(BaseService):
             description=dataset_data.description,
             reference=dataset_data.reference,
             license=dataset_data.license,
-            csv_file_path=dataset_data.csv_file_path,
+            file_path=dataset_data.file_path,
         )
-        return copy_file(new_ds_path)
+        return os.path.splitext(os.path.basename(new_ds_path))[0]
 
     @exception_handler
     def download_dataset(self, dataset_data: HF_Dataset_DTO) -> str:

moonshot/src/api/api_dataset.py

@@ -1,3 +1,6 @@
+import json
+import os
+
 from pydantic import validate_call
 
 from moonshot.src.datasets.dataset import Dataset
@@ -81,10 +84,10 @@ def api_download_dataset(
 
 
 def api_convert_dataset(
-    name: str, description: str, reference: str, license: str, csv_file_path: str
+    name: str, description: str, reference: str, license: str, file_path: str
 ) -> str:
     """
-    Converts a CSV file to a dataset and creates a new dataset with the provided details.
+    Converts a CSV or JSON file to a dataset and creates a new dataset with the provided details.
 
     This function takes the name, description, reference, and license for a new dataset as input, along with the file
     path to a CSV file. It then creates a new DatasetArguments object with these details and an empty id. The id is left
@@ -96,18 +99,55 @@ def api_convert_dataset(
         description (str): A brief description of the new dataset.
         reference (str): A reference link for the new dataset.
         license (str): The license of the new dataset.
-        csv_file_path (str): The file path to the CSV file.
+        file_path (str): The file path to the CSV or JSONfile.
 
     Returns:
         str: The ID of the newly created dataset.
     """
-    examples = Dataset.convert_data(csv_file_path)
-    ds_args = DatasetArguments(
-        id="",
-        name=name,
-        description=description,
-        reference=reference,
-        license=license,
-        examples=examples,
-    )
+    ds_args = None
+
+    # Check if file is in a supported format
+    if not (file_path.endswith(".json") or file_path.endswith(".csv")):
+        raise ValueError("Unsupported file format. Please provide a JSON or CSV file.")
+
+    # Check that file is not empty
+    if os.path.getsize(file_path) == 0:
+        raise ValueError("The uploaded file is empty.")
+
+    # if file is already in json format
+    if file_path.endswith(".json"):
+        json_data = json.load(open(file_path))
+
+        try:
+            if "examples" in json_data and json_data["examples"]:
+                ds_args = DatasetArguments(
+                    id="",
+                    name=json_data.get("name", name),
+                    description=json_data.get("description", description),
+                    reference=json_data.get("reference", reference),
+                    license=json_data.get("license", license),
+                    examples=iter(json_data["examples"]),
+                )
+            else:
+                raise KeyError(
+                    "examples is either empty or this key is not in the JSON file. "
+                    "Please ensure that this field is present."
+                )
+        except Exception as e:
+            raise e
+
+    # if file is in csv format, convert data
+    else:
+        try:
+            examples = Dataset.convert_data(file_path)
+            ds_args = DatasetArguments(
+                id="",
+                name=name,
+                description=description,
+                reference=reference,
+                license=license,
+                examples=examples,
+            )
+        except Exception as e:
+            raise e
     return Dataset.create(ds_args)

moonshot/src/datasets/dataset.py

@@ -60,7 +60,6 @@ class Dataset:
             }
 
             examples = ds_args.examples
-
             # Write as JSON output
             file_path = Storage.create_object_with_iterator(
                 EnvVariables.DATASETS.name,
@@ -91,9 +90,26 @@ class Dataset:
         Returns:
             Iterator[dict]: An iterator of dictionaries representing the CSV data.
         """
+        # validate headers
+        df_header = pd.read_csv(csv_file_path, nrows=1)
+        headers = df_header.columns.tolist()
+        required_headers = ["input", "target"]
+        if not all(header in headers for header in required_headers):
+            raise KeyError(
+                f"Required headers not found in the dataset. Required headers are {required_headers}."
+            )
+
         df = pd.read_csv(csv_file_path, chunksize=1)
-        for chunk in df:
-            yield chunk.to_dict("records")[0]
+        # validate dataset
+        first_chunk = next(df, None)
+        if first_chunk is None or first_chunk.empty:
+            raise ValueError("The uploaded file does not contain any data.")
+
+        # Reset df after performing next(df)
+        df = pd.read_csv(csv_file_path, chunksize=1)
+
+        result = [chunk.to_dict("records")[0] for chunk in df]
+        return iter(result)
 
     @staticmethod
     @validate_call