validmind 2.8.10__tar.gz → 2.8.20__tar.gz

{validmind-2.8.10 → validmind-2.8.20}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: validmind
-Version: 2.8.10
+Version: 2.8.20
 Summary: ValidMind Library
 License: Commercial License
 Author: Andres Rodriguez
@@ -35,12 +35,12 @@ Requires-Dist: numba (<0.59.0)
 Requires-Dist: numpy
 Requires-Dist: openai (>=1)
 Requires-Dist: pandas (>=1.1,<=2.0.3)
-Requires-Dist: plotly
+Requires-Dist: plotly (<6.0.0)
 Requires-Dist: plotly-express
 Requires-Dist: polars
 Requires-Dist: pycocoevalcap (>=1.2,<2.0) ; extra == "all" or extra == "llm"
 Requires-Dist: python-dotenv
-Requires-Dist: ragas (>=0.2.3) ; extra == "all" or extra == "llm"
+Requires-Dist: ragas (>=0.2.3,<=0.2.7) ; extra == "all" or extra == "llm"
 Requires-Dist: rouge (>=1)
 Requires-Dist: scikit-learn (<1.6.0)
 Requires-Dist: scipy
@@ -57,7 +57,6 @@ Requires-Dist: torch (>=1.10.0) ; extra == "all" or extra == "llm" or extra == "
 Requires-Dist: tqdm
 Requires-Dist: transformers (>=4.32.0,<5.0.0) ; extra == "all" or extra == "huggingface" or extra == "llm"
 Requires-Dist: xgboost (>=1.5.2,<3)
-Requires-Dist: ydata-profiling
 Requires-Dist: yfinance (>=0.2.48,<0.3.0)
 Description-Content-Type: text/markdown
 
@@ -84,7 +83,7 @@ ValidMind helps developers, data scientists and risk and compliance stakeholders
 >
 > Signing up is FREE — **[Register with ValidMind](https://docs.validmind.ai/guide/configuration/register-with-validmind.html)**
 
-That's right — you can run tests and log documentation even if you don't have a model available, so go ahead and [**Get started with the ValidMind Library**](https://docs.validmind.ai/developer/get-started-validmind-library.html)!
+That's right — you can run tests and log documentation even if you don't have a model available, so go ahead and get started with the [**ValidMind Library**](https://docs.validmind.ai/developer/validmind-library.html)!
 
 ### How do I do more with the ValidMind Library?
 

{validmind-2.8.10 → validmind-2.8.20}/README.pypi.md

@@ -21,7 +21,7 @@ ValidMind helps developers, data scientists and risk and compliance stakeholders
 >
 > Signing up is FREE — **[Register with ValidMind](https://docs.validmind.ai/guide/configuration/register-with-validmind.html)**
 
-That's right — you can run tests and log documentation even if you don't have a model available, so go ahead and [**Get started with the ValidMind Library**](https://docs.validmind.ai/developer/get-started-validmind-library.html)!
+That's right — you can run tests and log documentation even if you don't have a model available, so go ahead and get started with the [**ValidMind Library**](https://docs.validmind.ai/developer/validmind-library.html)!
 
 ### How do I do more with the ValidMind Library?
 

{validmind-2.8.10 → validmind-2.8.20}/pyproject.toml

@@ -10,7 +10,7 @@ description = "ValidMind Library"
 license = "Commercial License"
 name = "validmind"
 readme = "README.pypi.md"
-version = "2.8.10"
+version = "2.8.20"
 
 [tool.poetry.dependencies]
 aiohttp = {extras = ["speedups"], version = "*"}
@@ -33,13 +33,13 @@ numba = "<0.59.0" # TODO: https://github.com/validmind/validmind-library/pull/28
 numpy = "*"
 openai = ">=1"
 pandas = ">=1.1,<=2.0.3"
-plotly = "*"
+plotly = "<6.0.0"
 plotly-express = "*"
 polars = "*"
 pycocoevalcap = {version = "^1.2", optional = true}
 python = ">=3.8.1,<3.12"
 python-dotenv = "*"
-ragas = {version = ">=0.2.3", optional = true}
+ragas = {version = ">=0.2.3,<=0.2.7", optional = true}
 rouge = ">=1"
 scikit-learn = "*,<1.6.0"
 scipy = "*"
@@ -56,17 +56,19 @@ torch = {version = ">=1.10.0", optional = true}
 tqdm = "*"
 transformers = {version = "^4.32.0", optional = true}
 xgboost = ">=1.5.2,<3"
-ydata-profiling = "*"
 yfinance = "^0.2.48"
 
 [tool.poetry.group.dev.dependencies]
 black = "^22.1.0"
 click = "*"
 cython = "^0.29.34"
+docstring_parser = "*"
 flake8 = "^4.0.1"
+griffe = "*"
 ipykernel = "^6.22.0"
 isort = "^5.12.0"
 jupyter = "^1.0.0"
+mdformat = "*"
 papermill = "^2.4.0"
 pdoc = "^14.4.0"
 pre-commit = "^3.3.3"

{validmind-2.8.10 → validmind-2.8.20}/validmind/__init__.py

@@ -43,7 +43,7 @@ warnings.simplefilter("ignore", category=NumbaDeprecationWarning)
 warnings.simplefilter("ignore", category=NumbaPendingDeprecationWarning)
 
 from .__version__ import __version__  # noqa: E402
-from .api_client import init, log_metric, reload
+from .api_client import init, log_metric, log_text, reload
 from .client import (  # noqa: E402
     get_test_suite,
     init_dataset,
@@ -99,19 +99,19 @@ __all__ = [  # noqa
     "__version__",
     # main library API
     "init",
-    "reload",
     "init_dataset",
     "init_model",
     "init_r_model",
+    "get_test_suite",
+    "log_metric",
     "preview_template",
+    "print_env",
+    "reload",
     "run_documentation_tests",
     # log metric function (for direct/bulk/retroactive logging of metrics)
-    "log_metric",
     # test suite functions (less common)
-    "get_test_suite",
     "run_test_suite",
     # helper functions (for troubleshooting)
-    "print_env",
     # decorators (for building tests
     "tags",
     "tasks",
@@ -125,4 +125,5 @@ __all__ = [  # noqa
     "tests",
     "unit_metrics",
     "test_suites",
+    "log_text",
 ]

validmind-2.8.20/validmind/__version__.py

@@ -0,0 +1 @@
+__version__ = "2.8.20"

{validmind-2.8.10 → validmind-2.8.20}/validmind/ai/test_descriptions.py

@@ -37,8 +37,10 @@ def _get_llm_global_context():
     return context if context_enabled and context else None
 
 
-def _truncate_summary(summary: str, test_id: str, max_tokens: int = 100_000):
-    if len(summary) < max_tokens:
+def _truncate_summary(
+    summary: Union[str, None], test_id: str, max_tokens: int = 100_000
+):
+    if summary is None or len(summary) < max_tokens:
         # since string itself is less than max_tokens, definitely small enough
         return summary
 
@@ -68,7 +70,7 @@ def generate_description(
     figures: List[Figure] = None,
     title: Optional[str] = None,
 ):
-    """Generate the description for the test results"""
+    """Generate the description for the test results."""
     from validmind.api_client import generate_test_result_description
 
     if not tables and not figures and not metric:
@@ -136,6 +138,10 @@ def background_generate_description(
                 logger.warning(
                     f"Test result {test_id} is too large to generate a description"
                 )
+            elif "Too many images" in str(e):
+                logger.warning(
+                    f"Test result {test_id} has too many figures to generate a description"
+                )
             else:
                 logger.warning(f"Failed to generate description for {test_id}: {e}")
             logger.warning(f"Using default description for {test_id}")
@@ -154,7 +160,7 @@ def get_result_description(
     should_generate: bool = True,
     title: Optional[str] = None,
 ):
-    """Get Metadata Dictionary for a Test or Metric Result
+    """Get the metadata dictionary for a test or metric result.
 
     Generates an LLM interpretation of the test results or uses the default
     description and returns a metadata object that can be logged with the test results.
@@ -168,15 +174,15 @@ def get_result_description(
     Note: Either the tables or figures must be provided to generate the description.
 
     Args:
-        test_id (str): The test ID
-        test_description (str): The default description for the test
-        tables (Any): The test tables or results to interpret
-        figures (List[Figure]): The figures to attach to the test suite result
-        metric (Union[int, float]): Unit metrics attached to the test result
-        should_generate (bool): Whether to generate the description or not (Default: True)
+        test_id (str): The test ID.
+        test_description (str): The default description for the test.
+        tables (Any): The test tables or results to interpret.
+        figures (List[Figure]): The figures to attach to the test suite result.
+        metric (Union[int, float]): Unit metrics attached to the test result.
+        should_generate (bool): Whether to generate the description or not. Defaults to True.
 
     Returns:
-        str: The description to be logged with the test results
+        str: The description to be logged with the test results.
     """
     # Check the feature flag first, then the environment variable
     llm_descriptions_enabled = (

{validmind-2.8.10 → validmind-2.8.20}/validmind/ai/utils.py

@@ -24,7 +24,7 @@ class DescriptionFuture:
     the tests can continue to be run in parallel while the description is
     retrieved asynchronously.
 
-    The value will be retrieved later and if its not ready yet, it should
+    The value will be retrieved later and, if it is not ready yet, it should
     block until it is.
     """
 
@@ -42,7 +42,7 @@ class DescriptionFuture:
 
 
 def get_client_and_model():
-    """Get model and client to use for generating interpretations
+    """Get model and client to use for generating interpretations.
 
     On first call, it will look in the environment for the API key endpoint, model etc.
     and store them in a global variable to avoid loading them up again.

{validmind-2.8.10 → validmind-2.8.20}/validmind/api_client.py

@@ -18,11 +18,12 @@ from urllib.parse import urlencode, urljoin
 import aiohttp
 import requests
 from aiohttp import FormData
+from ipywidgets import HTML, Accordion
 
 from .client_config import client_config
 from .errors import MissingAPICredentialsError, MissingModelIdError, raise_api_error
 from .logging import get_logger, init_sentry, send_single_error
-from .utils import NumpyEncoder, run_async
+from .utils import NumpyEncoder, is_html, md_to_html, run_async
 from .vm_models import Figure
 
 logger = get_logger(__name__)
@@ -38,7 +39,7 @@ __api_session: Optional[aiohttp.ClientSession] = None
 
 @atexit.register
 def _close_session():
-    """Closes the async client session at exit"""
+    """Closes the async client session at exit."""
     global __api_session
 
     if __api_session and not __api_session.closed:
@@ -78,7 +79,7 @@ def _get_api_headers() -> Dict[str, str]:
 
 
 def _get_session() -> aiohttp.ClientSession:
-    """Initializes the async client session"""
+    """Initializes the async client session."""
     global __api_session
 
     if not __api_session or __api_session.closed:
@@ -156,7 +157,7 @@ async def _post(
 
 
 def _ping() -> Dict[str, Any]:
-    """Validates that we can connect to the ValidMind API (does not use the async session)"""
+    """Validates that we can connect to the ValidMind API (does not use the async session)."""
     r = requests.get(
         url=_get_url("ping"),
         headers=_get_api_headers(),
@@ -243,7 +244,7 @@ def init(
 
 
 def reload():
-    """Reconnect to the ValidMind API and reload the project configuration"""
+    """Reconnect to the ValidMind API and reload the project configuration."""
 
     try:
         _ping()
@@ -258,13 +259,13 @@ async def aget_metadata(content_id: str) -> Dict[str, Any]:
     """Gets a metadata object from ValidMind API.
 
     Args:
-        content_id (str): Unique content identifier for the metadata
+        content_id (str): Unique content identifier for the metadata.
 
     Raises:
-        Exception: If the API call fails
+        Exception: If the API call fails.
 
     Returns:
-        dict: Metadata object
+        dict: Metadata object.
     """
     return await _get(f"get_metadata/{content_id}")
 
@@ -277,15 +278,15 @@ async def alog_metadata(
     """Logs free-form metadata to ValidMind API.
 
     Args:
-        content_id (str): Unique content identifier for the metadata
+        content_id (str): Unique content identifier for the metadata.
         text (str, optional): Free-form text to assign to the metadata. Defaults to None.
         _json (dict, optional): Free-form key-value pairs to assign to the metadata. Defaults to None.
 
     Raises:
-        Exception: If the API call fails
+        Exception: If the API call fails.
 
     Returns:
-        dict: The response from the API
+        dict: The response from the API.
     """
     metadata_dict = {"content_id": content_id}
     if text is not None:
@@ -304,16 +305,16 @@ async def alog_metadata(
 
 
 async def alog_figure(figure: Figure) -> Dict[str, Any]:
-    """Logs a figure
+    """Logs a figure.
 
     Args:
-        figure (Figure): The Figure object wrapper
+        figure (Figure): The Figure object wrapper.
 
     Raises:
-        Exception: If the API call fails
+        Exception: If the API call fails.
 
     Returns:
-        dict: The response from the API
+        dict: The response from the API.
     """
     try:
         return await _post(
@@ -330,22 +331,24 @@ async def alog_test_result(
     result: Dict[str, Any],
     section_id: str = None,
     position: int = None,
+    unsafe: bool = False,
+    config: Dict[str, bool] = None,
 ) -> Dict[str, Any]:
-    """Logs test results information
+    """Logs test results information.
 
     This method will be called automatically from any function running tests but
     can also be called directly if the user wants to run tests on their own.
 
     Args:
-        result (dict): A dictionary representing the test result
-        section_id (str, optional): The section ID add a test driven block to the documentation
-        position (int): The position in the section to add the test driven block
+        result (dict): A dictionary representing the test result.
+        section_id (str, optional): The section ID add a test driven block to the documentation.
+        position (int): The position in the section to add the test driven block.
 
     Raises:
-        Exception: If the API call fails
+        Exception: If the API call fails.
 
     Returns:
-        dict: The response from the API
+        dict: The response from the API.
     """
     request_params = {}
     if section_id:
@@ -357,7 +360,7 @@ async def alog_test_result(
             "log_test_results",
             params=request_params,
             data=json.dumps(
-                result,
+                {**result, "config": config},
                 cls=NumpyEncoder,
                 allow_nan=False,
             ),
@@ -405,6 +408,39 @@ def log_input(input_id: str, type: str, metadata: Dict[str, Any]) -> Dict[str, A
     return run_async(alog_input, input_id, type, metadata)
 
 
+def log_text(
+    content_id: str, text: str, _json: Optional[Dict[str, Any]] = None
+) -> Dict[str, Any]:
+    """Logs free-form text to ValidMind API.
+
+    Args:
+        content_id (str): Unique content identifier for the text.
+        text (str): The text to log. Will be converted to HTML with MathML support.
+        _json (dict, optional): Additional metadata to associate with the text. Defaults to None.
+
+    Raises:
+        ValueError: If content_id or text are empty or not strings.
+        Exception: If the API call fails.
+
+    Returns:
+        ipywidgets.Accordion: An accordion widget containing the logged text as HTML.
+    """
+    if not content_id or not isinstance(content_id, str):
+        raise ValueError("`content_id` must be a non-empty string")
+    if not text or not isinstance(text, str):
+        raise ValueError("`text` must be a non-empty string")
+
+    if not is_html(text):
+        text = md_to_html(text, mathml=True)
+
+    log_text = run_async(alog_metadata, content_id, text, _json)
+
+    return Accordion(
+        children=[HTML(log_text["text"])],
+        titles=[f"Text Block: '{log_text['content_id']}'"],
+    )
+
+
 async def alog_metric(
     key: str,
     value: Union[int, float],
@@ -413,7 +449,7 @@ async def alog_metric(
     recorded_at: Optional[str] = None,
     thresholds: Optional[Dict[str, Any]] = None,
 ):
-    """See log_metric for details"""
+    """See log_metric for details."""
     if not key or not isinstance(key, str):
         raise ValueError("`key` must be a non-empty string")
 
@@ -458,7 +494,7 @@ def log_metric(
     recorded_at: Optional[str] = None,
     thresholds: Optional[Dict[str, Any]] = None,
 ):
-    """Logs a unit metric
+    """Logs a unit metric.
 
     Unit metrics are key-value pairs where the key is the metric name and the value is
     a scalar (int or float). These key-value pairs are associated with the currently
@@ -468,18 +504,25 @@ def log_metric(
 
     Args:
         key (str): The metric key
-        value (float): The metric value
-        inputs (list, optional): A list of input IDs that were used to compute the metric.
-        params (dict, optional): Dictionary of parameters used to compute the metric.
-        recorded_at (str, optional): The timestamp of the metric. Server will use
-            current time if not provided.
-        thresholds (dict, optional): Dictionary of thresholds for the metric.
+        value (Union[int, float]): The metric value
+        inputs (List[str], optional): List of input IDs
+        params (Dict[str, Any], optional): Parameters used to generate the metric
+        recorded_at (str, optional): Timestamp when the metric was recorded
+        thresholds (Dict[str, Any], optional): Thresholds for the metric
     """
-    run_async(alog_metric, key, value, inputs, params, recorded_at, thresholds)
+    return run_async(
+        alog_metric,
+        key=key,
+        value=value,
+        inputs=inputs,
+        params=params,
+        recorded_at=recorded_at,
+        thresholds=thresholds,
+    )
 
 
 def get_ai_key() -> Dict[str, Any]:
-    """Calls the api to get an api key for our LLM proxy"""
+    """Calls the API to get an API key for our LLM proxy."""
     r = requests.get(
         url=_get_url("ai/key"),
         headers=_get_api_headers(),