rasa-pro 3.9.17__py3-none-any.whl → 3.10.3__py3-none-any.whl

rasa/shared/utils/llm.py

@@ -1,46 +1,60 @@
-import os
-import warnings
-from typing import Any, Dict, Optional, Text, Type, TYPE_CHECKING, Union
-
+from functools import wraps
+from typing import (
+    Any,
+    Callable,
+    Dict,
+    Optional,
+    Text,
+    Type,
+    TypeVar,
+    TYPE_CHECKING,
+    Union,
+    cast,
+)
+import json
 import structlog
 
 import rasa.shared.utils.io
 from rasa.shared.constants import (
     RASA_PATTERN_INTERNAL_ERROR_USER_INPUT_TOO_LONG,
     RASA_PATTERN_INTERNAL_ERROR_USER_INPUT_EMPTY,
-    OPENAI_API_TYPE_ENV_VAR,
-    OPENAI_API_VERSION_ENV_VAR,
-    OPENAI_API_BASE_ENV_VAR,
-    REQUESTS_CA_BUNDLE_ENV_VAR,
-    OPENAI_API_BASE_NO_PREFIX_CONFIG_KEY,
-    OPENAI_API_TYPE_NO_PREFIX_CONFIG_KEY,
-    OPENAI_API_VERSION_CONFIG_KEY,
-    OPENAI_API_VERSION_NO_PREFIX_CONFIG_KEY,
-    OPENAI_API_TYPE_CONFIG_KEY,
-    OPENAI_API_BASE_CONFIG_KEY,
-    OPENAI_DEPLOYMENT_NAME_CONFIG_KEY,
-    OPENAI_DEPLOYMENT_CONFIG_KEY,
-    OPENAI_ENGINE_CONFIG_KEY,
-    LANGCHAIN_TYPE_CONFIG_KEY,
-    RASA_TYPE_CONFIG_KEY,
+    PROVIDER_CONFIG_KEY,
 )
 from rasa.shared.core.events import BotUttered, UserUttered
 from rasa.shared.core.slots import Slot, BooleanSlot, CategoricalSlot
-from rasa.shared.engine.caching import get_local_cache_location
+from rasa.shared.engine.caching import (
+    get_local_cache_location,
+)
 from rasa.shared.exceptions import (
     FileIOException,
     FileNotFoundException,
+    ProviderClientValidationError,
+)
+from rasa.shared.providers._configs.azure_openai_client_config import (
+    is_azure_openai_config,
+)
+from rasa.shared.providers._configs.huggingface_local_embedding_client_config import (
+    is_huggingface_local_config,
 )
+from rasa.shared.providers._configs.openai_client_config import is_openai_config
+from rasa.shared.providers._configs.self_hosted_llm_client_config import (
+    is_self_hosted_config,
+)
+from rasa.shared.providers.embedding.embedding_client import EmbeddingClient
+from rasa.shared.providers.llm.llm_client import LLMClient
+from rasa.shared.providers.mappings import (
+    get_llm_client_from_provider,
+    AZURE_OPENAI_PROVIDER,
+    OPENAI_PROVIDER,
+    SELF_HOSTED_PROVIDER,
+    get_embedding_client_from_provider,
+    HUGGINGFACE_LOCAL_EMBEDDING_PROVIDER,
+    get_client_config_class_from_provider,
+)
+from rasa.shared.utils.cli import print_error_and_exit
 
 if TYPE_CHECKING:
-    from langchain.chat_models import AzureChatOpenAI
-    from langchain.schema.embeddings import Embeddings
-    from langchain.llms.base import BaseLLM
     from rasa.shared.core.trackers import DialogueStateTracker
-    from rasa.shared.providers.openai.clients import (
-        AioHTTPSessionAzureChatOpenAI,
-        AioHTTPSessionOpenAIChat,
-    )
 
 structlogger = structlog.get_logger()
 
@@ -70,6 +84,94 @@ ERROR_PLACEHOLDER = {
     "default": "[User input triggered an error]",
 }
 
+_Factory_F = TypeVar(
+    "_Factory_F",
+    bound=Callable[[Dict[str, Any], Dict[str, Any]], Union[EmbeddingClient, LLMClient]],
+)
+_CombineConfigs_F = TypeVar(
+    "_CombineConfigs_F",
+    bound=Callable[[Dict[str, Any], Dict[str, Any]], Dict[str, Any]],
+)
+
+
+def _compute_hash_for_cache_from_configs(
+    config_x: Dict[str, Any], config_y: Dict[str, Any]
+) -> int:
+    """Get a unique hash of the default and custom configs."""
+    return hash(
+        json.dumps(config_x, sort_keys=True) + json.dumps(config_y, sort_keys=True)
+    )
+
+
+def _retrieve_from_cache(
+    cache: Dict[int, Any], unique_hash: int, function: Callable, function_kwargs: dict
+) -> Any:
+    """Retrieve the value from the cache if it exists. If it does not exist, cache it"""
+    if unique_hash in cache:
+        return cache[unique_hash]
+    else:
+        return_value = function(**function_kwargs)
+        cache[unique_hash] = return_value
+        return return_value
+
+
+def _cache_factory(function: _Factory_F) -> _Factory_F:
+    """Memoize the factory methods based on the arguments."""
+    cache: Dict[int, Union[EmbeddingClient, LLMClient]] = {}
+
+    @wraps(function)
+    def factory_method_wrapper(
+        config_x: Dict[str, Any], config_y: Dict[str, Any]
+    ) -> Union[EmbeddingClient, LLMClient]:
+        # Get a unique hash of the default and custom configs.
+        unique_hash = _compute_hash_for_cache_from_configs(config_x, config_y)
+        return _retrieve_from_cache(
+            cache=cache,
+            unique_hash=unique_hash,
+            function=function,
+            function_kwargs={"custom_config": config_x, "default_config": config_y},
+        )
+
+    def clear_cache() -> None:
+        cache.clear()
+        structlogger.debug(
+            "Cleared cache for factory method",
+            function_name=function.__name__,
+        )
+
+    setattr(factory_method_wrapper, "clear_cache", clear_cache)
+    return cast(_Factory_F, factory_method_wrapper)
+
+
+def _cache_combine_custom_and_default_configs(
+    function: _CombineConfigs_F,
+) -> _CombineConfigs_F:
+    """Memoize the combine_custom_and_default_config method based on the arguments."""
+    cache: Dict[int, dict] = {}
+
+    @wraps(function)
+    def combine_configs_wrapper(
+        config_x: Dict[str, Any], config_y: Dict[str, Any]
+    ) -> dict:
+        # Get a unique hash of the default and custom configs.
+        unique_hash = _compute_hash_for_cache_from_configs(config_x, config_y)
+        return _retrieve_from_cache(
+            cache=cache,
+            unique_hash=unique_hash,
+            function=function,
+            function_kwargs={"custom_config": config_x, "default_config": config_y},
+        )
+
+    def clear_cache() -> None:
+        cache.clear()
+        structlogger.debug(
+            "Cleared cache for combine_custom_and_default_config method",
+            function_name=function.__name__,
+        )
+
+    setattr(combine_configs_wrapper, "clear_cache", clear_cache)
+    return cast(_CombineConfigs_F, combine_configs_wrapper)
+
 
 def tracker_as_readable_transcript(
     tracker: "DialogueStateTracker",
@@ -138,11 +240,15 @@ def sanitize_message_for_prompt(text: Optional[str]) -> str:
     return text.replace("\n", " ") if text else ""
 
 
+@_cache_combine_custom_and_default_configs
 def combine_custom_and_default_config(
-    custom_config: Optional[Dict[Text, Any]], default_config: Dict[Text, Any]
+    custom_config: Optional[Dict[str, Any]], default_config: Dict[str, Any]
 ) -> Dict[Text, Any]:
     """Merges the given llm config with the default config.
 
+    This method guarantees that the provider is set and all the deprecated keys are
+    resolved. Hence, produces only a valid client config.
+
     Only uses the default configuration arguments, if the type set in the
     custom config matches the type in the default config. Otherwise, only
     the custom config is used.
@@ -155,155 +261,96 @@ def combine_custom_and_default_config(
         The merged config.
     """
     if custom_config is None:
-        return default_config
-
-    if RASA_TYPE_CONFIG_KEY in custom_config:
-        # rename type to _type as "type" is the convention we use
-        # across the different components in config files.
-        # langchain expects "_type" as the key though
-        custom_config[LANGCHAIN_TYPE_CONFIG_KEY] = custom_config.pop(
-            RASA_TYPE_CONFIG_KEY
+        return default_config.copy()
+
+    # Get the provider from the custom config.
+    custom_config_provider = get_provider_from_config(custom_config)
+    # We expect the provider to be set in the default configs of all Rasa components.
+    default_config_provider = default_config[PROVIDER_CONFIG_KEY]
+
+    if (
+        custom_config_provider is not None
+        and custom_config_provider != default_config_provider
+    ):
+        # Get the provider-specific config class
+        client_config_clazz = get_client_config_class_from_provider(
+            custom_config_provider
         )
+        # Checks for deprecated keys, resolves aliases and returns a valid config.
+        # This is done to ensure that the custom config is valid.
+        return client_config_clazz.from_dict(custom_config).to_dict()
+
+    # If the provider is the same in both configs
+    # OR provider is not specified in the custom config
+    # perform MERGE by overriding the default config keys and values
+    # with custom config keys and values.
+    merged_config = {**default_config.copy(), **custom_config.copy()}
+    # Check for deprecated keys, resolve aliases and return a valid config.
+    # This is done to ensure that the merged config is valid.
+    default_config_clazz = get_client_config_class_from_provider(
+        default_config_provider
+    )
+    return default_config_clazz.from_dict(merged_config).to_dict()
 
-    if LANGCHAIN_TYPE_CONFIG_KEY in custom_config and custom_config[
-        LANGCHAIN_TYPE_CONFIG_KEY
-    ] != default_config.get(LANGCHAIN_TYPE_CONFIG_KEY):
-        return custom_config
-    return {**default_config, **custom_config}
+
+def get_provider_from_config(config: dict) -> Optional[str]:
+    """Try to get the provider from the passed llm/embeddings configuration.
+    If no provider can be found, return None.
+    """
+    if not config:
+        return None
+    if is_self_hosted_config(config):
+        return SELF_HOSTED_PROVIDER
+    elif is_azure_openai_config(config):
+        return AZURE_OPENAI_PROVIDER
+    elif is_openai_config(config):
+        return OPENAI_PROVIDER
+    elif is_huggingface_local_config(config):
+        return HUGGINGFACE_LOCAL_EMBEDDING_PROVIDER
+    else:
+        return config.get(PROVIDER_CONFIG_KEY)
 
 
 def ensure_cache() -> None:
     """Ensures that the cache is initialized."""
-    import langchain
-    from langchain.cache import SQLiteCache
+    import litellm
 
-    # ensure the cache directory exists
-    cache_location = get_local_cache_location()
+    # Ensure the cache directory exists
+    cache_location = get_local_cache_location() / "rasa-llm-cache"
     cache_location.mkdir(parents=True, exist_ok=True)
 
-    db_location = cache_location / "rasa-llm-cache.db"
-    langchain.llm_cache = SQLiteCache(database_path=str(db_location))
-
-
-def preprocess_config_for_azure(config: Dict[str, Any]) -> Dict[str, Any]:
-    """Preprocesses the config for Azure deployments.
-
-    This function is used to preprocess the config for Azure deployments.
-    AzureChatOpenAI does not expect the _type key, as it is not a defined parameter
-    in the class. So we need to remove it before passing the config to the class.
-    AzureChatOpenAI expects the openai_api_type key to be set instead.
-
-    Args:
-        config: The config to preprocess.
-
-    Returns:
-        The preprocessed config.
-    """
-    config["deployment_name"] = (
-        config.get(OPENAI_DEPLOYMENT_NAME_CONFIG_KEY)
-        or config.get(OPENAI_DEPLOYMENT_CONFIG_KEY)
-        or config.get(OPENAI_ENGINE_CONFIG_KEY)
-    )
-    config["openai_api_base"] = (
-        config.get(OPENAI_API_BASE_CONFIG_KEY)
-        or config.get(OPENAI_API_BASE_NO_PREFIX_CONFIG_KEY)
-        or os.environ.get(OPENAI_API_BASE_ENV_VAR)
-    )
-    config["openai_api_type"] = (
-        config.get(OPENAI_API_TYPE_CONFIG_KEY)
-        or config.get(OPENAI_API_TYPE_NO_PREFIX_CONFIG_KEY)
-        or os.environ.get(OPENAI_API_TYPE_ENV_VAR)
-    )
-    config["openai_api_version"] = (
-        config.get(OPENAI_API_VERSION_CONFIG_KEY)
-        or config.get(OPENAI_API_VERSION_NO_PREFIX_CONFIG_KEY)
-        or os.environ.get(OPENAI_API_VERSION_ENV_VAR)
-    )
-    for keys in [
-        OPENAI_API_BASE_NO_PREFIX_CONFIG_KEY,
-        OPENAI_API_TYPE_NO_PREFIX_CONFIG_KEY,
-        OPENAI_API_VERSION_NO_PREFIX_CONFIG_KEY,
-        OPENAI_DEPLOYMENT_CONFIG_KEY,
-        OPENAI_ENGINE_CONFIG_KEY,
-        LANGCHAIN_TYPE_CONFIG_KEY,
-    ]:
-        config.pop(keys, None)
-
-    return config
-
-
-def process_config_for_aiohttp_chat_openai(config: Dict[str, Any]) -> Dict[str, Any]:
-    config = config.copy()
-    config.pop(LANGCHAIN_TYPE_CONFIG_KEY)
-    return config
+    # Set diskcache as a caching option
+    litellm.cache = litellm.Cache(type="disk", disk_cache_dir=cache_location)
 
 
+@_cache_factory
 def llm_factory(
     custom_config: Optional[Dict[str, Any]], default_config: Dict[str, Any]
-) -> Union[
-    "BaseLLM",
-    "AzureChatOpenAI",
-    "AioHTTPSessionAzureChatOpenAI",
-    "AioHTTPSessionOpenAIChat",
-]:
+) -> LLMClient:
     """Creates an LLM from the given config.
 
     Args:
         custom_config: The custom config  containing values to overwrite defaults
         default_config: The default config.
 
-
     Returns:
-    Instantiated LLM based on the configuration.
+        Instantiated LLM based on the configuration.
     """
-    from langchain.llms.loading import load_llm_from_config
-
-    ensure_cache()
-
     config = combine_custom_and_default_config(custom_config, default_config)
 
-    # need to create a copy as the langchain function modifies the
-    # config in place...
-    structlogger.debug("llmfactory.create.llm", config=config)
-    # langchain issues a user warning when using chat models. at the same time
-    # it doesn't provide a way to instantiate a chat model directly using the
-    # config. so for now, we need to suppress the warning here. Original
-    # warning:
-    #   packages/langchain/llms/openai.py:189: UserWarning: You are trying to
-    #   use a chat model. This way of initializing it is no longer supported.
-    #   Instead, please use: `from langchain.chat_models import ChatOpenAI
-    with warnings.catch_warnings():
-        warnings.simplefilter("ignore", category=UserWarning)
-        if is_azure_config(config):
-            # Azure deployments are treated differently. This is done as the
-            # GPT-3.5 Turbo newer versions 0613 and 1106 only support the
-            # Chat Completions API.
-            from langchain.chat_models import AzureChatOpenAI
-            from rasa.shared.providers.openai.clients import (
-                AioHTTPSessionAzureChatOpenAI,
-            )
-
-            transformed_config = preprocess_config_for_azure(config.copy())
-            if os.environ.get(REQUESTS_CA_BUNDLE_ENV_VAR) is None:
-                return AzureChatOpenAI(**transformed_config)
-            else:
-                return AioHTTPSessionAzureChatOpenAI(**transformed_config)
-
-        if (
-            os.environ.get(REQUESTS_CA_BUNDLE_ENV_VAR) is not None
-            and config.get(LANGCHAIN_TYPE_CONFIG_KEY) == "openai"
-        ):
-            from rasa.shared.providers.openai.clients import AioHTTPSessionOpenAIChat
-
-            config = process_config_for_aiohttp_chat_openai(config)
-            return AioHTTPSessionOpenAIChat(**config.copy())
+    ensure_cache()
 
-        return load_llm_from_config(config.copy())
+    client_clazz: Type[LLMClient] = get_llm_client_from_provider(
+        config[PROVIDER_CONFIG_KEY]
+    )
+    client = client_clazz.from_config(config)
+    return client
 
 
+@_cache_factory
 def embedder_factory(
     custom_config: Optional[Dict[str, Any]], default_config: Dict[str, Any]
-) -> "Embeddings":
+) -> EmbeddingClient:
     """Creates an Embedder from the given config.
 
     Args:
@@ -312,55 +359,17 @@ def embedder_factory(
 
 
     Returns:
-    Instantiated Embedder based on the configuration.
+        Instantiated Embedder based on the configuration.
     """
-    from langchain.schema.embeddings import Embeddings
-    from langchain.embeddings import (
-        CohereEmbeddings,
-        HuggingFaceHubEmbeddings,
-        HuggingFaceInstructEmbeddings,
-        HuggingFaceEmbeddings,
-        HuggingFaceBgeEmbeddings,
-        LlamaCppEmbeddings,
-        OpenAIEmbeddings,
-        SpacyEmbeddings,
-        VertexAIEmbeddings,
-    )
-    from rasa.shared.providers.openai.clients import AioHTTPSessionOpenAIEmbeddings
-
-    type_to_embedding_cls_dict: Dict[str, Type[Embeddings]] = {
-        "azure": OpenAIEmbeddings,
-        "openai": OpenAIEmbeddings,
-        "openai-aiohttp-session": AioHTTPSessionOpenAIEmbeddings,
-        "cohere": CohereEmbeddings,
-        "spacy": SpacyEmbeddings,
-        "vertexai": VertexAIEmbeddings,
-        "huggingface_instruct": HuggingFaceInstructEmbeddings,
-        "huggingface_hub": HuggingFaceHubEmbeddings,
-        "huggingface_bge": HuggingFaceBgeEmbeddings,
-        "huggingface": HuggingFaceEmbeddings,
-        "llamacpp": LlamaCppEmbeddings,
-    }
-
     config = combine_custom_and_default_config(custom_config, default_config)
-    embedding_type = config.get(LANGCHAIN_TYPE_CONFIG_KEY)
-
-    if (
-        os.environ.get(REQUESTS_CA_BUNDLE_ENV_VAR) is not None
-        and embedding_type is not None
-    ):
-        embedding_type = f"{embedding_type}-aiohttp-session"
 
-    structlogger.debug("llmfactory.create.embedder", config=config)
+    ensure_cache()
 
-    if not embedding_type:
-        return OpenAIEmbeddings()
-    elif embeddings_cls := type_to_embedding_cls_dict.get(embedding_type):
-        parameters = config.copy()
-        parameters.pop(LANGCHAIN_TYPE_CONFIG_KEY)
-        return embeddings_cls(**parameters)
-    else:
-        raise ValueError(f"Unsupported embeddings type '{embedding_type}'")
+    client_clazz: Type[EmbeddingClient] = get_embedding_client_from_provider(
+        config[PROVIDER_CONFIG_KEY]
+    )
+    client = client_clazz.from_config(config)
+    return client
 
 
 def get_prompt_template(
@@ -396,9 +405,23 @@ def allowed_values_for_slot(slot: Slot) -> Union[str, None]:
         return None
 
 
-def is_azure_config(config: Dict) -> bool:
-    return (
-        config.get(OPENAI_API_TYPE_CONFIG_KEY) == "azure"
-        or config.get(OPENAI_API_TYPE_NO_PREFIX_CONFIG_KEY) == "azure"
-        or os.environ.get(OPENAI_API_TYPE_ENV_VAR) == "azure"
-    )
+def try_instantiate_llm_client(
+    custom_llm_config: Optional[Dict],
+    default_llm_config: Optional[Dict],
+    log_source_function: str,
+    log_source_component: str,
+) -> None:
+    """Validate llm configuration."""
+    try:
+        llm_factory(custom_llm_config, default_llm_config)
+    except (ProviderClientValidationError, ValueError) as e:
+        structlogger.error(
+            f"{log_source_function}.llm_instantiation_failed",
+            message="Unable to instantiate LLM client.",
+            error=e,
+        )
+        print_error_and_exit(
+            f"Unable to create the LLM client for component - {log_source_component}. "
+            f"Please make sure you specified the required environment variables. "
+            f"Error: {e}"
+        )

rasa/shared/utils/yaml.py

@@ -1,3 +1,4 @@
+import datetime
 import logging
 import os
 import re
@@ -12,15 +13,17 @@ from typing import Dict, List, Optional, Any, Callable, Tuple, Union
 import jsonschema
 from importlib_resources import files
 from packaging import version
-from packaging.version import LegacyVersion
 from pykwalify.core import Core
 from pykwalify.errors import SchemaError
 from ruamel import yaml as yaml
 from ruamel.yaml import RoundTripRepresenter, YAMLError
 from ruamel.yaml.constructor import DuplicateKeyError, BaseConstructor, ScalarNode
 from ruamel.yaml.comments import CommentedSeq, CommentedMap
+from ruamel.yaml.loader import SafeLoader
 
 from rasa.shared.constants import (
+    ASSERTIONS_SCHEMA_EXTENSIONS_FILE,
+    ASSERTIONS_SCHEMA_FILE,
     MODEL_CONFIG_SCHEMA_FILE,
     CONFIG_SCHEMA_FILE,
     DOCS_URL_TRAINING_DATA,
@@ -413,12 +416,17 @@ def validate_raw_yaml_using_schema_file_with_responses(
     )
 
 
-def read_yaml(content: str, reader_type: Union[str, List[str]] = "safe") -> Any:
+def read_yaml(
+    content: str,
+    reader_type: Union[str, List[str]] = "safe",
+    **kwargs: Any,
+) -> Any:
     """Parses yaml from a text.
 
     Args:
         content: A text containing yaml content.
         reader_type: Reader type to use. By default, "safe" will be used.
+        **kwargs: Any
 
     Raises:
         ruamel.yaml.parser.ParserError: If there was an error when parsing the YAML.
@@ -432,11 +440,93 @@ def read_yaml(content: str, reader_type: Union[str, List[str]] = "safe") -> Any:
             .decode("utf-16")
         )
 
+    custom_constructor = kwargs.get("custom_constructor", None)
+
+    # Create YAML parser with custom constructor
+    yaml_parser, reset_constructors = create_yaml_parser(
+        reader_type, custom_constructor
+    )
+    yaml_content = yaml_parser.load(content) or {}
+
+    # Reset to default constructors
+    reset_constructors()
+
+    return yaml_content
+
+
+def create_yaml_parser(
+    reader_type: str,
+    custom_constructor: Optional[Callable] = None,
+) -> Tuple[yaml.YAML, Callable[[], None]]:
+    """Create a YAML parser with an optional custom constructor.
+
+    Args:
+        reader_type (str): The type of the reader
+        (e.g., 'safe', 'rt', 'unsafe').
+        custom_constructor (Optional[Callable]):
+        A custom constructor function for YAML parsing.
+
+    Returns:
+        Tuple[yaml.YAML, Callable[[], None]]: A tuple containing
+        the YAML parser and a function to reset constructors to
+        their original state.
+    """
     yaml_parser = yaml.YAML(typ=reader_type)
     yaml_parser.version = YAML_VERSION  # type: ignore[assignment]
     yaml_parser.preserve_quotes = True  # type: ignore[assignment]
 
-    return yaml_parser.load(content) or {}
+    # Save the original constructors
+    original_mapping_constructor = yaml_parser.constructor.yaml_constructors.get(
+        yaml.resolver.BaseResolver.DEFAULT_MAPPING_TAG
+    )
+    original_sequence_constructor = yaml_parser.constructor.yaml_constructors.get(
+        yaml.resolver.BaseResolver.DEFAULT_SEQUENCE_TAG
+    )
+
+    if custom_constructor is not None:
+        # Attach the custom constructor to the loader
+        yaml_parser.constructor.add_constructor(
+            yaml.resolver.BaseResolver.DEFAULT_MAPPING_TAG, custom_constructor
+        )
+        yaml_parser.constructor.add_constructor(
+            yaml.resolver.BaseResolver.DEFAULT_SEQUENCE_TAG, custom_constructor
+        )
+
+    def reset_constructors() -> None:
+        """Reset the constructors back to their original state."""
+        yaml_parser.constructor.add_constructor(
+            yaml.resolver.BaseResolver.DEFAULT_MAPPING_TAG, original_mapping_constructor
+        )
+        yaml_parser.constructor.add_constructor(
+            yaml.resolver.BaseResolver.DEFAULT_SEQUENCE_TAG,
+            original_sequence_constructor,
+        )
+
+    def custom_date_constructor(loader: SafeLoader, node: ScalarNode) -> str:
+        """Custom constructor for parsing dates in the format '%Y-%m-%d'.
+
+        This constructor parses dates in the '%Y-%m-%d' format and returns them as
+        strings instead of datetime objects. This change was introduced because the
+        default timestamp constructor in ruamel.yaml returns datetime objects, which
+        caused issues in our use case where the `api_version` in the LLM config must
+        be a string, but was being interpreted as a datetime object.
+        """
+        value = loader.construct_scalar(node)
+        try:
+            # Attempt to parse the date
+            date_obj = datetime.datetime.strptime(value, "%Y-%m-%d").date()
+            # Return the date as a string instead of a datetime object
+            return date_obj.strftime("%Y-%m-%d")
+        except ValueError:
+            # If the date is not in the correct format, return the original value
+            return value
+
+    # Add the custom date constructor
+    yaml_parser.constructor.add_constructor(
+        "tag:yaml.org,2002:timestamp", custom_date_constructor
+    )
+
+    return yaml_parser, reset_constructors
 
 
 def _is_ascii(text: str) -> bool:
@@ -684,9 +774,6 @@ def validate_training_data_format_version(
         parsed_version = version.parse(version_value)
         latest_version = version.parse(LATEST_TRAINING_DATA_FORMAT_VERSION)
 
-        if isinstance(parsed_version, LegacyVersion):
-            raise TypeError
-
         if parsed_version < latest_version:
             raise_warning(
                 f"Training data file {filename} has a lower "
@@ -702,7 +789,7 @@ def validate_training_data_format_version(
         if latest_version >= parsed_version:
             return True
 
-    except TypeError:
+    except (TypeError, version.InvalidVersion):
         raise_warning(
             f"Training data file {filename} must specify "
             f"'{KEY_TRAINING_DATA_FORMAT_VERSION}' as string, for example:\n"
@@ -784,3 +871,31 @@ def validate_yaml_with_jsonschema(
             errors,
             content=source_data,
         )
+
+
+def validate_yaml_data_using_schema_with_assertions(
+    yaml_data: Any,
+    schema_content: Union[List[Any], Dict[str, Any]],
+    package_name: str = PACKAGE_NAME,
+) -> None:
+    """Validate raw yaml content using a schema with assertions sub-schema.
+
+    Args:
+        yaml_data: the parsed yaml data to be validated
+        schema_content: the content of the YAML schema
+        package_name: the name of the package the schema is located in. defaults
+        to `rasa`.
+    """
+    # test case assertions are part of the schema extension
+    # it will be included if the schema explicitly references it with
+    # include: assertions
+    e2e_test_cases_schema_content = read_schema_file(
+        ASSERTIONS_SCHEMA_FILE, package_name
+    )
+
+    schema_content = dict(schema_content, **e2e_test_cases_schema_content)
+    schema_extensions = [
+        str(files(package_name).joinpath(ASSERTIONS_SCHEMA_EXTENSIONS_FILE))
+    ]
+
+    validate_yaml_content_using_schema(yaml_data, schema_content, schema_extensions)