The Memory Layer for AI Agents - Drop-in memory infrastructure for AI agents and apps. Context that persists. Built for production.
Python agentic-memoryagentic-memory-systemagentsaiai-agentschatgptgenaillmlong-term-memorymemorymemory-managementpythonragstate-management
[Doc update] update mistral example (#1012)
Deven Patel0efbc80ac972
diff --git a/docs/get-started/faq.mdx b/docs/get-started/faq.mdxindex d5899f91a..096a7330b 100644--- a/docs/get-started/faq.mdx+++ b/docs/get-started/faq.mdx@@ -13,7 +13,6 @@ Use the model provided on huggingface: `mistralai/Mistral-7B-v0.1`import osfrom embedchain import Pipeline as App-os.environ["OPENAI_API_KEY"] = "sk-xxx"os.environ["HUGGINGFACE_ACCESS_TOKEN"] = "hf_your_token"app = App.from_config("huggingface.yaml")@@ -27,6 +26,11 @@ llm:max_tokens: 1000top_p: 0.5stream: false++embedder:+ provider: huggingface+ config:+ model: 'sentence-transformers/all-mpnet-base-v2'```</CodeGroup></Accordion>diff --git a/embedchain/loaders/directory_loader.py b/embedchain/loaders/directory_loader.pyindex bc670bc6e..d4941939e 100644--- a/embedchain/loaders/directory_loader.py+++ b/embedchain/loaders/directory_loader.py@@ -1,14 +1,14 @@-from pathlib import Pathimport hashlibimport logging-from typing import Optional, Dict, Any+from pathlib import Path+from typing import Any, Dict, Optional-from embedchain.utils import detect_datatype+from embedchain.config import AddConfig+from embedchain.data_formatter.data_formatter import DataFormatterfrom embedchain.helpers.json_serializable import register_deserializablefrom embedchain.loaders.base_loader import BaseLoaderfrom embedchain.loaders.local_text import LocalTextLoader-from embedchain.data_formatter.data_formatter import DataFormatter-from embedchain.config import AddConfig+from embedchain.utils import detect_datatype@register_deserializablediff --git a/embedchain/utils.py b/embedchain/utils.pyindex 7396fb5c3..4db82d609 100644--- a/embedchain/utils.py+++ b/embedchain/utils.py@@ -196,7 +196,8 @@ def detect_datatype(source: Any) -> DataType:formatted_source = format_source(str(source), 30)if url:- from langchain.document_loaders.youtube import ALLOWED_NETLOCK as YOUTUBE_ALLOWED_NETLOCS+ from langchain.document_loaders.youtube import \+ ALLOWED_NETLOCK as YOUTUBE_ALLOWED_NETLOCSif url.netloc in YOUTUBE_ALLOWED_NETLOCS:logging.debug(f"Source of `{formatted_source}` detected as `youtube_video`.")