polly-python 2.5.0__tar.gz → 3.0.0__tar.gz

{polly_python-2.5.0/polly_python.egg-info → polly_python-3.0.0}/PKG-INFO

@@ -1,14 +1,14 @@
 Metadata-Version: 2.1
 Name: polly_python
-Version: 2.5.0
+Version: 3.0.0
 Summary: Polly SDK
 Home-page: https://github.com/ElucidataInc/polly-python
 Project-URL: Documentation, https://docs.elucidata.io
 Project-URL: Tutorial Notebooks, https://github.com/ElucidataInc/polly-python
-Requires-Python: >3.8
+Requires-Python: <=3.11,>=3.9
 Description-Content-Type: text/markdown
 License-File: LICENSE.md
-Requires-Dist: elucidatacmapPy==3.3.4
+Requires-Dist: cmapPy<=4.0.1
 Requires-Dist: cloudpathlib>=0.15.0
 Requires-Dist: retrying==1.3.4
 Requires-Dist: rst2txt==1.1.0
@@ -17,22 +17,15 @@ Requires-Dist: mixpanel==4.10.0
 Requires-Dist: Deprecated>=1.2.12
 Requires-Dist: pytest>=6.2.5
 Requires-Dist: cryptography<=38.0.0,>=37.0.1
-Requires-Dist: plotly<5.0.0,>=4.8.1; python_version > "3.6" and python_version < "3.7"
 Requires-Dist: plotly>=5.0.0; python_version >= "3.7"
-Requires-Dist: pandas<1.2.0,>=1.1.0; python_version > "3.6" and python_version < "3.7"
-Requires-Dist: pandas>=1.3.5; python_version >= "3.7"
-Requires-Dist: pydantic<1.10.0a1,>=1.8.2; python_version > "3.6" and python_version < "3.7"
+Requires-Dist: pandas<=2.2.2,>=1.3.5; python_version >= "3.7"
+Requires-Dist: numpy<=1.26.4
 Requires-Dist: pydantic==1.10.12; python_version >= "3.7"
 Requires-Dist: requests==2.28.1
-Requires-Dist: numpy==1.26.4
-Requires-Dist: boto3<1.24.0,>=1.17.73; python_version > "3.6" and python_version < "3.7"
-Requires-Dist: boto3>=1.24.0; python_version >= "3.7"
-Requires-Dist: botocore<1.27.0,>=1.20.73; python_version > "3.6" and python_version < "3.7"
-Requires-Dist: botocore>=1.27.0; python_version >= "3.7"
-Requires-Dist: joblib<=1.1.0,>0.11.0; python_version > "3.6" and python_version < "3.7"
+Requires-Dist: boto3<2.0,>=1.24.0; python_version >= "3.7"
+Requires-Dist: botocore<2.0,>=1.27.0; python_version >= "3.7"
 Requires-Dist: joblib>=1.2.0; python_version >= "3.7"
 Requires-Dist: tabulate==0.9.0
-Requires-Dist: tqdm<4.65.0,>=4.61.0; python_version > "3.6" and python_version < "3.7"
 Requires-Dist: tqdm==4.65.0; python_version >= "3.7"
 Provides-Extra: testing
 Requires-Dist: black; extra == "testing"

polly_python-3.0.0/polly/__init__.py

@@ -0,0 +1 @@
+__version__ = "3.0.0"

{polly_python-2.5.0 → polly_python-3.0.0}/polly/constants.py

@@ -92,22 +92,22 @@ IO_CHUNKSIZE_LARGE_FILE_SIZE = 100 * MB
 EXPIRED_TOKEN = "ExpiredToken"
 
 # cohort constants
-COHORT_VERSION = "0.2"
-COHORT_CONSTANTS_URL = (
-    "https://elucidatainc.github.io/PublicAssets/cohort_constants.txt"
-)
+# COHORT_VERSION = "0.2"
+# COHORT_CONSTANTS_URL = (
+#     "https://elucidatainc.github.io/PublicAssets/cohort_constants.txt"
+# )
 
-OBSOLETE_METADATA_FIELDS = [
-    "package",
-    "region",
-    "bucket",
-    "key",
-    "file_type",
-    "file_location",
-    "src_uri",
-    "timestamp_",
-]
-dot = "."
+# OBSOLETE_METADATA_FIELDS = [
+#     "package",
+#     "region",
+#     "bucket",
+#     "key",
+#     "file_type",
+#     "file_location",
+#     "src_uri",
+#     "timestamp_",
+# ]
+# dot = "."
 
 GETTING_UPLOAD_URLS_PAYLOAD = {"data": {"type": "files", "attributes": {"folder": ""}}}
 

{polly_python-2.5.0 → polly_python-3.0.0}/polly/curation.py

@@ -1,9 +1,11 @@
 import json
 from collections import namedtuple
-import os
-import shutil
+
+# import os
+# import shutil
 from typing import Dict, Optional, List
-import warnings
+
+# import warnings
 import pandas as pd
 from functools import lru_cache
 from polly.errors import (
@@ -13,15 +15,19 @@ from polly.errors import (
     RequestException,
     UnauthorizedException,
     extract_json_api_error,
-    paramException,
+    # paramException,
 )
 from polly.auth import Polly
-from polly.cohort import Cohort
+
+# from polly.cohort import Cohort
 from polly import helpers, constants as const, application_error_info as app_err_info
 from polly.help import example
 import polly.http_response_codes as http_codes
-from polly.constants import SUPPORTED_ENTITY_TYPES, CURATION_COHORT_CACHE
-from polly.helpers import get_cohort_constants
+
+# from polly.constants import SUPPORTED_ENTITY_TYPES, CURATION_COHORT_CACHE
+from polly.constants import SUPPORTED_ENTITY_TYPES
+
+# from polly.helpers import get_cohort_constants
 from polly.tracking import Track
 
 
@@ -60,162 +66,162 @@ class Curation:
             f"https://api.datalake.discover.{self.session.env}.elucidata.io/elastic/v2"
         )
         self.inference_url = f"https://api.discover.{self.session.env}.elucidata.io/curations/inferences/"
-        self.cohort = Cohort()
-        self.cohort_constants = get_cohort_constants()
+        # self.cohort = Cohort()
+        # self.cohort_constants = get_cohort_constants()
 
     def _handle_errors(self, response):
         detail = response.get("errors")[0].get("detail", [])
         title = response.get("errors")[0].get("title", [])
         return title, detail
 
-    def _fetch_metadata_from_cohort(self, repo_name: str, dataset_ids: List[str]):
-        """
-        Utility function for fetching metadata using cohorts.
-
-        Arguments:
-            repo_name (str) : name of the repository for fetching datasets.
-            dataset_ids (List[str]): dataset ids to be used for inference
-
-        Returns:
-            Returns sample metadata, dataset and sample ids.
-        """
-        sample_metadata = {}
-        dataset_to_sample_id = {"dataset_id": [], "sample_id": []}
-
-        if not (os.path.isdir(CURATION_COHORT_CACHE)):
-            os.mkdir(CURATION_COHORT_CACHE)
-        else:
-            shutil.rmtree(CURATION_COHORT_CACHE)
-            os.mkdir(CURATION_COHORT_CACHE)
-
-        self.cohort.create_cohort(
-            CURATION_COHORT_CACHE, "sample_metadata_query", "desc"
-        )
-
-        # Fetch metadata using cohorts
-        for dataset_id in dataset_ids:
-            datasets_sample_metadata = []
-
-            if not (
-                repo_name in self.cohort_constants
-                and self.cohort_constants[repo_name]["file_structure"] != "multiple"
-            ):
-                # multiple mapped repo such as GEO
-                self.cohort.add_to_cohort(repo_name, dataset_id=dataset_id)
-            else:
-                # for single mapped repos such as TCGA
-                self.cohort.add_to_cohort(repo_name, dataset_id=[dataset_id])
-
-            col_metadata = self.cohort.merge_data("sample")
-            all_sample_ids = col_metadata.index.tolist()
-
-            col_metadata.loc[:, "dataset_id"] = dataset_id
-            dataset_to_sample_id["dataset_id"] += [dataset_id] * len(all_sample_ids)
-
-            col_metadata.loc[:, "sample_id"] = all_sample_ids
-            dataset_to_sample_id["sample_id"] += all_sample_ids
-
-            datasets_sample_metadata += list(col_metadata.T.to_dict().values())
-
-            if not (
-                repo_name in self.cohort_constants
-                and self.cohort_constants[repo_name]["file_structure"] != "multiple"
-            ):
-                self.cohort.remove_from_cohort(dataset_id)
-            else:
-                self.cohort.remove_from_cohort([dataset_id])
-
-            sample_metadata[dataset_id] = datasets_sample_metadata
-
-        dataset_to_sample_id = pd.DataFrame.from_dict(dataset_to_sample_id)
-
-        return sample_metadata, dataset_to_sample_id
-
-    def _clinical_model_param_checks(
-        self,
-        repo_name: str,
-        dataset_ids: List[str],
-        sample_ids: Optional[List[str]] = None,
-    ):
-        """
-        Checking the parameter passed to the clinical label assigning model.
-
-        Arguments:
-            repo_name (str): repo name
-            dataset_ids (list[str]): list of dataset ids
-
-        Keyword Arguments:
-            sample_ids (list[str], optional): Optional Parameter. List of sample ids.
-            Default is 'None'.
-
-        Raises:
-            paramException
-        """
-        if dataset_ids is None or type(dataset_ids) is not list:
-            raise paramException(
-                title="Param Exception",
-                detail="Dataset IDs should be given as a valid list of strings",
-            )
-
-        if sample_ids is not None and type(sample_ids) is not list:
-            raise paramException(
-                title="Param Exception",
-                detail="Sample IDs should be given as a valid list of strings",
-            )
-
-        if repo_name != "geo" and not any(
-            ["GSE" in dataset_id for dataset_id in dataset_ids]
-        ):
-            warnings.warn(
-                "The model is tested with GEO metadata and the labels may be wrong for other repos"
-            )
-
-    def _post_process_clinical_tags(
-        self,
-        clinical_tags: pd.DataFrame,
-        is_sample_tag: bool,
-        sample_ids: Optional[List[str]] = None,
-    ) -> pd.DataFrame:
-        """
-        process the response of the model (dataframe with clinical tags and samples)
-        and return relevant feilds.
-        incase no sample_ids are provided by the user, we return the dataset_ids and the clinical tags
-        incase sample_ids are also provided, then we return the dataset_ids, the sample_ids and the clincal tags.
-
-        Arguments:
-            clinical_tags (pd.DataFrame): dataframe of the sample_ids and assigned clinical tags
-            is_sample_tag (bool): if samples passed
-
-        Keyword Arguments:
-            sample_ids (list[str]): list of sample ids (default: {None})
-
-        Returns:
-            a dataframe with the the dataset_ids, sample_ids and the assigned clinical tags
-        """
-        if is_sample_tag:
-            # if the user has provided list of samples, then we filter in just those sample ids
-            # for the dataset ids.
-            # taking only those clinical tags and samples where the sample_ids are in the sample_id list
-            # provided by the user.
-            clinical_tags = clinical_tags[
-                clinical_tags["sample_id"].isin(sample_ids)
-            ].reset_index(drop=True)
-
-            # in case the sample_ids provided by the user are not present in the dataset_ids provided.
-            if clinical_tags.empty or clinical_tags.shape[0] < len(sample_ids):
-                warnings.warn(
-                    "The output is empty or has missing sample ids because they are not present in given datasets."
-                )
-
-            # return sample level tags here
-            return clinical_tags
-        # if no sample_ids were passed by the user, then
-        # returning dataset level tags by removing sample id and removing duplicate columns
-        return (
-            clinical_tags.drop(columns=["sample_id"])
-            .drop_duplicates()
-            .reset_index(drop=True)
-        )
+    # def _fetch_metadata_from_cohort(self, repo_name: str, dataset_ids: List[str]):
+    #     """
+    #     Utility function for fetching metadata using cohorts.
+
+    #     Arguments:
+    #         repo_name (str) : name of the repository for fetching datasets.
+    #         dataset_ids (List[str]): dataset ids to be used for inference
+
+    #     Returns:
+    #         Returns sample metadata, dataset and sample ids.
+    #     """
+    #     sample_metadata = {}
+    #     dataset_to_sample_id = {"dataset_id": [], "sample_id": []}
+
+    #     if not (os.path.isdir(CURATION_COHORT_CACHE)):
+    #         os.mkdir(CURATION_COHORT_CACHE)
+    #     else:
+    #         shutil.rmtree(CURATION_COHORT_CACHE)
+    #         os.mkdir(CURATION_COHORT_CACHE)
+
+    #     self.cohort.create_cohort(
+    #         CURATION_COHORT_CACHE, "sample_metadata_query", "desc"
+    #     )
+
+    #     # Fetch metadata using cohorts
+    #     for dataset_id in dataset_ids:
+    #         datasets_sample_metadata = []
+
+    #         if not (
+    #             repo_name in self.cohort_constants
+    #             and self.cohort_constants[repo_name]["file_structure"] != "multiple"
+    #         ):
+    #             # multiple mapped repo such as GEO
+    #             self.cohort.add_to_cohort(repo_name, dataset_id=dataset_id)
+    #         else:
+    #             # for single mapped repos such as TCGA
+    #             self.cohort.add_to_cohort(repo_name, dataset_id=[dataset_id])
+
+    #         col_metadata = self.cohort.merge_data("sample")
+    #         all_sample_ids = col_metadata.index.tolist()
+
+    #         col_metadata.loc[:, "dataset_id"] = dataset_id
+    #         dataset_to_sample_id["dataset_id"] += [dataset_id] * len(all_sample_ids)
+
+    #         col_metadata.loc[:, "sample_id"] = all_sample_ids
+    #         dataset_to_sample_id["sample_id"] += all_sample_ids
+
+    #         datasets_sample_metadata += list(col_metadata.T.to_dict().values())
+
+    #         if not (
+    #             repo_name in self.cohort_constants
+    #             and self.cohort_constants[repo_name]["file_structure"] != "multiple"
+    #         ):
+    #             self.cohort.remove_from_cohort(dataset_id)
+    #         else:
+    #             self.cohort.remove_from_cohort([dataset_id])
+
+    #         sample_metadata[dataset_id] = datasets_sample_metadata
+
+    #     dataset_to_sample_id = pd.DataFrame.from_dict(dataset_to_sample_id)
+
+    #     return sample_metadata, dataset_to_sample_id
+
+    # def _clinical_model_param_checks(
+    #     self,
+    #     repo_name: str,
+    #     dataset_ids: List[str],
+    #     sample_ids: Optional[List[str]] = None,
+    # ):
+    #     """
+    #     Checking the parameter passed to the clinical label assigning model.
+
+    #     Arguments:
+    #         repo_name (str): repo name
+    #         dataset_ids (list[str]): list of dataset ids
+
+    #     Keyword Arguments:
+    #         sample_ids (list[str], optional): Optional Parameter. List of sample ids.
+    #         Default is 'None'.
+
+    #     Raises:
+    #         paramException
+    #     """
+    #     if dataset_ids is None or type(dataset_ids) is not list:
+    #         raise paramException(
+    #             title="Param Exception",
+    #             detail="Dataset IDs should be given as a valid list of strings",
+    #         )
+
+    #     if sample_ids is not None and type(sample_ids) is not list:
+    #         raise paramException(
+    #             title="Param Exception",
+    #             detail="Sample IDs should be given as a valid list of strings",
+    #         )
+
+    #     if repo_name != "geo" and not any(
+    #         ["GSE" in dataset_id for dataset_id in dataset_ids]
+    #     ):
+    #         warnings.warn(
+    #             "The model is tested with GEO metadata and the labels may be wrong for other repos"
+    #         )
+
+    # def _post_process_clinical_tags(
+    #     self,
+    #     clinical_tags: pd.DataFrame,
+    #     is_sample_tag: bool,
+    #     sample_ids: Optional[List[str]] = None,
+    # ) -> pd.DataFrame:
+    #     """
+    #     process the response of the model (dataframe with clinical tags and samples)
+    #     and return relevant feilds.
+    #     incase no sample_ids are provided by the user, we return the dataset_ids and the clinical tags
+    #     incase sample_ids are also provided, then we return the dataset_ids, the sample_ids and the clincal tags.
+
+    #     Arguments:
+    #         clinical_tags (pd.DataFrame): dataframe of the sample_ids and assigned clinical tags
+    #         is_sample_tag (bool): if samples passed
+
+    #     Keyword Arguments:
+    #         sample_ids (list[str]): list of sample ids (default: {None})
+
+    #     Returns:
+    #         a dataframe with the the dataset_ids, sample_ids and the assigned clinical tags
+    #     """
+    #     if is_sample_tag:
+    #         # if the user has provided list of samples, then we filter in just those sample ids
+    #         # for the dataset ids.
+    #         # taking only those clinical tags and samples where the sample_ids are in the sample_id list
+    #         # provided by the user.
+    #         clinical_tags = clinical_tags[
+    #             clinical_tags["sample_id"].isin(sample_ids)
+    #         ].reset_index(drop=True)
+
+    #         # in case the sample_ids provided by the user are not present in the dataset_ids provided.
+    #         if clinical_tags.empty or clinical_tags.shape[0] < len(sample_ids):
+    #             warnings.warn(
+    #                 "The output is empty or has missing sample ids because they are not present in given datasets."
+    #             )
+
+    #         # return sample level tags here
+    #         return clinical_tags
+    #     # if no sample_ids were passed by the user, then
+    #     # returning dataset level tags by removing sample id and removing duplicate columns
+    #     return (
+    #         clinical_tags.drop(columns=["sample_id"])
+    #         .drop_duplicates()
+    #         .reset_index(drop=True)
+    #     )
 
     def _handle_perform_inference_api_error(self, response):
         if response.status_code == http_codes.UNAUTHORIZED:
@@ -483,83 +489,83 @@ class Curation:
         sample_metadata["control_prob"] = output["control_prob"].values
         return sample_metadata
 
-    @Track.track_decorator
-    def assign_clinical_labels(
-        self,
-        repo_name: str,
-        dataset_ids: List[str],
-        sample_ids: Optional[List[str]] = None,
-    ) -> pd.DataFrame:
-        """
-        Returns a list of clinical or non clinical labels for the given datasets or samples.
-
-        Arguments:
-            repo_name (str): name of the repository for fetching datasets.
-            dataset_ids (List[str]): dataset ids to be used for inference
-
-        Keyword Arguments:
-            sample_ids (List[str], optional): Optional Parameter. Sample ids if that is needed.
-
-        Raises:
-            RequestException: API response exception
-            ParamException: Invalid parameters
-            err
-
-        Returns:
-            dataframe which is a list of clinical tags for given ids
-        """
-        warnings.formatwarning = lambda msg, *args, **kwargs: f"WARNING: {msg}\n"
-
-        try:
-            self._clinical_model_param_checks(repo_name, dataset_ids, sample_ids)
-            # evaluating the inference level based on if the user has provided sample_ids
-            is_sample_tag = sample_ids is not None
-            inference_level = "sample_id" if (is_sample_tag) else "dataset_id"
-
-            sample_metadata, dataset_to_sample_id = self._fetch_metadata_from_cohort(
-                repo_name=repo_name, dataset_ids=dataset_ids
-            )
-
-            clinical_model_predictions = []
-
-            for dataset_id in sample_metadata:
-                # Get output from model endpoint and structure output
-                payload = {
-                    "sample_metadata": sample_metadata[dataset_id],
-                    "sample_id_column": "sample_id",
-                    "dataset_id_column": "dataset_id",
-                    "is_sample_tag": is_sample_tag,
-                }
-
-                output = self._perform_inference("clinical-classifier", payload)
-                if "errors" in output:
-                    title, detail = self._handle_errors(output)
-                    raise RequestException(title, detail)
-
-                output = output["clinical_predictions"]
-
-                clinical_model_predictions += output
-
-            # creating dataframe with inference_level and clinical_tags with values from the clinical_model_predictions
-            clinical_tags = pd.DataFrame(
-                {
-                    inference_level: [
-                        tag["tag_id"] for tag in clinical_model_predictions
-                    ],
-                    "clinical_tag": [
-                        tag["clinical_tag"] for tag in clinical_model_predictions
-                    ],
-                }
-            )
-
-            clinical_tags = pd.merge(
-                dataset_to_sample_id, clinical_tags, on=inference_level
-            )
-
-            clinical_tags = self._post_process_clinical_tags(
-                clinical_tags, is_sample_tag, sample_ids
-            )
-        except Exception as err:
-            raise err
-
-        return clinical_tags
+    # @Track.track_decorator
+    # def assign_clinical_labels(
+    #     self,
+    #     repo_name: str,
+    #     dataset_ids: List[str],
+    #     sample_ids: Optional[List[str]] = None,
+    # ) -> pd.DataFrame:
+    #     """
+    #     Returns a list of clinical or non clinical labels for the given datasets or samples.
+
+    #     Arguments:
+    #         repo_name (str): name of the repository for fetching datasets.
+    #         dataset_ids (List[str]): dataset ids to be used for inference
+
+    #     Keyword Arguments:
+    #         sample_ids (List[str], optional): Optional Parameter. Sample ids if that is needed.
+
+    #     Raises:
+    #         RequestException: API response exception
+    #         ParamException: Invalid parameters
+    #         err
+
+    #     Returns:
+    #         dataframe which is a list of clinical tags for given ids
+    #     """
+    #     warnings.formatwarning = lambda msg, *args, **kwargs: f"WARNING: {msg}\n"
+
+    #     try:
+    #         self._clinical_model_param_checks(repo_name, dataset_ids, sample_ids)
+    #         # evaluating the inference level based on if the user has provided sample_ids
+    #         is_sample_tag = sample_ids is not None
+    #         inference_level = "sample_id" if (is_sample_tag) else "dataset_id"
+
+    #         sample_metadata, dataset_to_sample_id = self._fetch_metadata_from_cohort(
+    #             repo_name=repo_name, dataset_ids=dataset_ids
+    #         )
+
+    #         clinical_model_predictions = []
+
+    #         for dataset_id in sample_metadata:
+    #             # Get output from model endpoint and structure output
+    #             payload = {
+    #                 "sample_metadata": sample_metadata[dataset_id],
+    #                 "sample_id_column": "sample_id",
+    #                 "dataset_id_column": "dataset_id",
+    #                 "is_sample_tag": is_sample_tag,
+    #             }
+
+    #             output = self._perform_inference("clinical-classifier", payload)
+    #             if "errors" in output:
+    #                 title, detail = self._handle_errors(output)
+    #                 raise RequestException(title, detail)
+
+    #             output = output["clinical_predictions"]
+
+    #             clinical_model_predictions += output
+
+    #         # creating dataframe with inference_level and clinical_tags with values from the clinical_model_predictions
+    #         clinical_tags = pd.DataFrame(
+    #             {
+    #                 inference_level: [
+    #                     tag["tag_id"] for tag in clinical_model_predictions
+    #                 ],
+    #                 "clinical_tag": [
+    #                     tag["clinical_tag"] for tag in clinical_model_predictions
+    #                 ],
+    #             }
+    #         )
+
+    #         clinical_tags = pd.merge(
+    #             dataset_to_sample_id, clinical_tags, on=inference_level
+    #         )
+
+    #         clinical_tags = self._post_process_clinical_tags(
+    #             clinical_tags, is_sample_tag, sample_ids
+    #         )
+    #     except Exception as err:
+    #         raise err
+
+    #     return clinical_tags