validmind 2.0.7__tar.gz → 2.1.1__tar.gz

{validmind-2.0.7 → validmind-2.1.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: validmind
-Version: 2.0.7
+Version: 2.1.1
 Summary: ValidMind Developer Framework
 License: Commercial License
 Author: Andres Rodriguez
@@ -27,23 +27,27 @@ Requires-Dist: ipywidgets (>=8.0.6,<9.0.0)
 Requires-Dist: kaleido (>=0.2.1,<0.3.0,!=0.2.1.post1)
 Requires-Dist: langdetect (>=1.0.9,<2.0.0)
 Requires-Dist: levenshtein (>=0.21.1,<0.22.0) ; extra == "all" or extra == "llm"
-Requires-Dist: markdown (>=3.4.3,<4.0.0)
+Requires-Dist: llvmlite (>=0.42.0) ; python_version >= "3.12"
+Requires-Dist: llvmlite ; python_version >= "3.8" and python_full_version <= "3.11.0"
 Requires-Dist: matplotlib (<3.8.3)
+Requires-Dist: mistune (>=3.0.2,<4.0.0)
 Requires-Dist: nltk (>=3.8.1,<4.0.0)
 Requires-Dist: numba (<0.59.0)
-Requires-Dist: numpy (>=1.23.3,<2.0.0)
+Requires-Dist: numpy (>=1.23.3,<2.0.0) ; python_version >= "3.8" and python_full_version <= "3.11.0"
+Requires-Dist: numpy (>=1.26.0) ; python_version >= "3.12"
 Requires-Dist: openai (>=1.3.7,<2.0.0) ; extra == "all" or extra == "llm"
-Requires-Dist: pandas (==1.5.3)
+Requires-Dist: pandas (>=1.1,<2)
 Requires-Dist: plotly (>=5.14.1,<6.0.0)
 Requires-Dist: plotly-express (>=0.4.1,<0.5.0)
 Requires-Dist: polars (>=0.20.15,<0.21.0)
 Requires-Dist: pycocoevalcap (>=1.2,<2.0) ; extra == "all" or extra == "llm"
 Requires-Dist: pypmml (>=0.9.17,<0.10.0)
 Requires-Dist: python-dotenv (>=0.20.0,<0.21.0)
-Requires-Dist: requests (>=2.27.1,<3.0.0)
 Requires-Dist: rouge (>=1.0.1,<2.0.0)
 Requires-Dist: rpy2 (>=3.5.10,<4.0.0) ; extra == "all" or extra == "r-support"
 Requires-Dist: scikit-learn (>=1.0.2,<2.0.0)
+Requires-Dist: scipy (>=1.12) ; python_version >= "3.12"
+Requires-Dist: scipy ; python_version >= "3.8" and python_full_version <= "3.11.0"
 Requires-Dist: scorecardpy (>=0.1.9.6,<0.2.0.0)
 Requires-Dist: seaborn (>=0.11.2,<0.12.0)
 Requires-Dist: selfcheckgpt (>=0.1.7,<0.2.0)
@@ -56,7 +60,7 @@ Requires-Dist: torch (>=1.10.0) ; extra == "all" or extra == "llm" or extra == "
 Requires-Dist: torchmetrics (>=1.1.1,<2.0.0) ; extra == "all" or extra == "llm"
 Requires-Dist: tqdm (>=4.64.0,<5.0.0)
 Requires-Dist: transformers (>=4.32.0,<5.0.0) ; extra == "all" or extra == "llm" or extra == "transformers"
-Requires-Dist: xgboost (>=1.5.2,<2.0.0)
+Requires-Dist: xgboost (>=1.5.2,<3)
 Requires-Dist: ydata-profiling (>=4.7.0,<5.0.0)
 Description-Content-Type: text/markdown
 

{validmind-2.0.7 → validmind-2.1.1}/pyproject.toml

@@ -10,7 +10,7 @@ description = "ValidMind Developer Framework"
 license = "Commercial License"
 name = "validmind"
 readme = "README.pypi.md"
-version = "2.0.7"
+version = "2.1.1"
 
 [tool.poetry.dependencies]
 aiohttp = {extras = ["speedups"], version = "^3.8.4"}
@@ -23,13 +23,20 @@ ipywidgets = "^8.0.6"
 kaleido = "^0.2.1,!=0.2.1.post1"
 langdetect = "^1.0.9"
 levenshtein = {version = "^0.21.1", optional = true}
-markdown = "^3.4.3"
+llvmlite = [
+  {version = "*", python = ">=3.8,<=3.11"},
+  {version = ">=0.42.0", python = ">=3.12"},
+]
 matplotlib = "<3.8.3"
+mistune = "^3.0.2"
 nltk = "^3.8.1"
 numba = "<0.59.0" # TODO: https://github.com/validmind/developer-framework/pull/28
-numpy = "^1.23.3"
+numpy = [
+  {version = "^1.23.3", python = ">=3.8,<=3.11"},
+  {version = ">=1.26.0", python = ">=3.12"},
+]
 openai = {version = "^1.3.7", optional = true}
-pandas = "1.5.3"
+pandas = ">=1.1,<2" # TODO small issues blocking 2.0 but it would speed things up to use arrow backend
 plotly = "^5.14.1"
 plotly-express = "^0.4.1"
 polars = "^0.20.15"
@@ -37,10 +44,13 @@ pycocoevalcap = {version = "^1.2", optional = true}
 pypmml = "^0.9.17"
 python = ">=3.8,<3.12"
 python-dotenv = "^0.20.0"
-requests = "^2.27.1"
 rouge = "^1.0.1"
 rpy2 = {version = "^3.5.10", optional = true}
 scikit-learn = "^1.0.2"
+scipy = [
+  {version = "*", python = ">=3.8,<=3.11"},
+  {version = ">=1.12", python = ">=3.12"},
+]
 scorecardpy = "^0.1.9.6"
 seaborn = "^0.11.2"
 selfcheckgpt = "^0.1.7"
@@ -53,7 +63,7 @@ torch = {version = ">=1.10.0", optional = true}
 torchmetrics = {version = "^1.1.1", optional = true}
 tqdm = "^4.64.0"
 transformers = {version = "^4.32.0", optional = true}
-xgboost = "^1.5.2"
+xgboost = ">=1.5.2,<3"
 ydata-profiling = "^4.7.0"
 
 [tool.poetry.group.dev.dependencies]

{validmind-2.0.7 → validmind-2.1.1}/validmind/__init__.py

@@ -60,8 +60,7 @@ from .client import (  # noqa: E402
     run_documentation_tests,
     run_test_suite,
 )
-from .tests.decorator import metric
-from .unit_metrics import run_metric
+from .tests.decorator import metric, tags, tasks
 from .utils import run_async  # noqa: E402
 
 
@@ -111,6 +110,8 @@ __all__ = [  # noqa
     "reload",
     "run_documentation_tests",
     "run_test_suite",
+    "tags",
+    "tasks",
     "tests",
     "test_suites",
     "vm_models",
@@ -119,5 +120,4 @@ __all__ = [  # noqa
     "log_figure",
     "log_metrics",
     "log_test_results",
-    "run_metric",
 ]

validmind-2.1.1/validmind/__version__.py

@@ -0,0 +1 @@
+__version__ = "2.1.1"

{validmind-2.0.7 → validmind-2.1.1}/validmind/ai.py

@@ -7,8 +7,6 @@ import os
 
 from openai import AzureOpenAI, OpenAI
 
-from .utils import clean_docstring
-
 SYSTEM_PROMPT = """
 You are an expert data scientist and MRM specialist tasked with providing concise and'
 objective insights based on the results of quantitative model or dataset analysis.
@@ -23,22 +21,20 @@ Your analysis will act as the description of the result in the model documentati
 
 Avoid long sentences and complex vocabulary.
 Structure the response clearly and logically.
-Use Markdown syntax to format the response.
+Use valid Markdown syntax to format the response (tables are supported).
 Use the Test ID that is provided to form the Test Name e.g. "ClassImbalance" -> "Class Imbalance".
-Use the following format for the response:
+Use the following format for the response (feel free to modify slightly if necessary):
 ```
 **<Test Name>** <continue to explain what it does in detail>...
 
 The results of this test <detailed explanation of the results>...
 
-In summary the following key insights can be gained from this <Test Type>
+In summary the following key insights can be gained:
 
 - **<key insight 1 - title>**: <explanation of key insight 1>
 - ...<continue with any other key insights using the same format>
 ```
 It is very important that the text is nicely formatted and contains enough information to be useful to the user as documentation.
-
-- use valid markdown syntax: make sure to have two newlines between paragraphs and before bullet points etc.
 """.strip()
 USER_PROMPT = """
 Test ID: {test_name}
@@ -71,7 +67,7 @@ def __get_client_and_model():
 
     if "OPENAI_API_KEY" in os.environ:
         __client = OpenAI(api_key=os.environ.get("OPENAI_API_KEY"))
-        __model = os.environ.get("VM_OPENAI_MODEL", "gpt-4-turbo-preview")
+        __model = os.environ.get("VM_OPENAI_MODEL", "gpt-4-turbo")
 
     elif "AZURE_OPENAI_KEY" in os.environ:
         if "AZURE_OPENAI_ENDPOINT" not in os.environ:
@@ -111,7 +107,7 @@ class DescriptionFuture:
 
     def get_description(self):
         # This will block until the future is completed
-        return clean_docstring(self._future.result())
+        return self._future.result()
 
 
 def generate_description_async(
@@ -132,7 +128,7 @@ def generate_description_async(
             raise ValueError("No results, summary or figures provided")
 
         response = client.chat.completions.create(
-            model="gpt-4-1106-vision-preview",
+            model="gpt-4-turbo",
             messages=[
                 {"role": "system", "content": SYSTEM_PROMPT},
                 {
@@ -160,7 +156,7 @@ def generate_description_async(
         )
     else:
         response = client.chat.completions.create(
-            model="gpt-4-turbo-preview",
+            model="gpt-4-turbo",
             messages=[
                 {"role": "system", "content": SYSTEM_PROMPT},
                 {

{validmind-2.0.7 → validmind-2.1.1}/validmind/api_client.py

@@ -16,6 +16,7 @@ from io import BytesIO
 from typing import Any, Callable, Dict, List, Optional, Tuple, Union
 
 import aiohttp
+import mistune
 import requests
 from aiohttp import FormData
 
@@ -294,32 +295,33 @@ async def log_figures(figures: List[Figure]) -> Dict[str, Any]:
     Returns:
         dict: The response from the API
     """
-    if client_config.can_log_figures():  # check if the backend supports batch logging
-        try:
-            data = {}
-            files = {}
-            for figure in figures:
-                data.update(
-                    {f"{k}-{figure.key}": v for k, v in figure.serialize().items()}
-                )
-                files.update(
-                    {
-                        f"{k}-{figure.key}": v
-                        for k, v in figure.serialize_files().items()
-                    }
-                )
-
-            return await _post(
-                "log_figures",
-                data=data,
-                files=files,
-            )
-        except Exception as e:
-            logger.error("Error logging figures to ValidMind API")
-            raise e
-
-    else:
-        return await asyncio.gather(*[log_figure(figure) for figure in figures])
+    # this actually slows things down - better to log them in parallel
+    # if client_config.can_log_figures():  # check if the backend supports batch logging
+    #     try:
+    #         data = {}
+    #         files = {}
+    #         for figure in figures:
+    #             data.update(
+    #                 {f"{k}-{figure.key}": v for k, v in figure.serialize().items()}
+    #             )
+    #             files.update(
+    #                 {
+    #                     f"{k}-{figure.key}": v
+    #                     for k, v in figure.serialize_files().items()
+    #                 }
+    #             )
+
+    #         return await _post(
+    #             "log_figures",
+    #             data=data,
+    #             files=files,
+    #         )
+    #     except Exception as e:
+    #         logger.error("Error logging figures to ValidMind API")
+    #         raise e
+
+    # else:
+    return await asyncio.gather(*[log_figure(figure) for figure in figures])
 
 
 async def log_metadata(
@@ -342,7 +344,7 @@ async def log_metadata(
     """
     metadata_dict = {"content_id": content_id}
     if text is not None:
-        metadata_dict["text"] = text
+        metadata_dict["text"] = mistune.html(text)
     if _json is not None:
         metadata_dict["json"] = _json
 

{validmind-2.0.7 → validmind-2.1.1}/validmind/client.py

@@ -61,8 +61,13 @@ def init_dataset(
     """
     Initializes a VM Dataset, which can then be passed to other functions
     that can perform additional analysis and tests on the data. This function
-    also ensures we are reading a valid dataset type. We only support Pandas
-    DataFrames at the moment.
+    also ensures we are reading a valid dataset type.
+
+    The following dataset types are supported:
+    - Pandas DataFrame
+    - Polars DataFrame
+    - Numpy ndarray
+    - Torch TensorDataset
 
     Args:
         dataset : dataset from various python libraries
@@ -380,7 +385,7 @@ def preview_template():
 
 
 def run_documentation_tests(
-    section=None, send=True, fail_fast=False, inputs=None, **kwargs
+    section=None, send=True, fail_fast=False, inputs=None, config=None, **kwargs
 ):
     """Collect and run all the tests associated with a template
 
@@ -393,6 +398,7 @@ def run_documentation_tests(
         send (bool, optional): Whether to send the results to the ValidMind API. Defaults to True.
         fail_fast (bool, optional): Whether to stop running tests after the first failure. Defaults to False.
         inputs (dict, optional): A dictionary of test inputs to pass to the TestSuite
+        config: A dictionary of test parameters to override the defaults
         **kwargs: backwards compatibility for passing in test inputs using keyword arguments
 
     Returns:
@@ -421,6 +427,7 @@ def run_documentation_tests(
             send=send,
             fail_fast=fail_fast,
             inputs=inputs,
+            config=config,
             **kwargs,
         )
         test_suites[_section] = test_suite

validmind-2.1.1/validmind/datasets/credit_risk/__init__.py

@@ -0,0 +1,11 @@
+# Copyright © 2023-2024 ValidMind Inc. All rights reserved.
+# See the LICENSE file in the root of this repository for details.
+# SPDX-License-Identifier: AGPL-3.0 AND ValidMind Commercial
+
+"""
+Entrypoint for credit risk datasets.
+"""
+
+__all__ = [
+    "lending_club",
+]