klas/t6_mem0
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[Improvements] Add support for creating app from YAML string config (#980)

Browse Source
This commit is contained in:
Deven Patel
2023-11-29 12:25:30 -08:00
committed by GitHub
parent e35eaf1bfc
commit 406c46e7f4
34 changed files with 351 additions and 179 deletions
Download Patch File Download Diff File Expand all files Collapse all files

141
docs/advanced/configuration.mdx
View File

@@ -6,15 +6,16 @@ Embedchain is made to work out of the box. However, for advanced users we're als
You can configure different components of your app (`llm`, `embedding model`, or `vector database`) through a simple yaml configuration that Embedchain offers. Here is a generic full-stack example of the yaml config:
```yaml
<Tip>
Embedchain applications are configurable using YAML file, JSON file or by directly passing the config dictionary.
</Tip>
<CodeGroup>
```yaml config.yaml
app:
config:
id: 'full-stack-app'
chunker:
chunk_size: 100
chunk_overlap: 20
length_function: 'len'
name: 'full-stack-app'
llm:
provider: openai
@@ -47,38 +48,138 @@ embedder:
provider: openai
config:
model: 'text-embedding-ada-002'
chunker:
chunk_size: 2000
chunk_overlap: 100
length_function: 'len'
```
```json config.json
{
"app": {
"config": {
"name": "full-stack-app"
}
},
"llm": {
"provider": "openai",
"config": {
"model": "gpt-3.5-turbo",
"temperature": 0.5,
"max_tokens": 1000,
"top_p": 1,
"stream": false,
"template": "Use the following pieces of context to answer the query at the end.\nIf you don't know the answer, just say that you don't know, don't try to make up an answer.\n$context\n\nQuery: $query\n\nHelpful Answer:",
"system_prompt": "Act as William Shakespeare. Answer the following questions in the style of William Shakespeare."
}
},
"vectordb": {
"provider": "chroma",
"config": {
"collection_name": "full-stack-app",
"dir": "db",
"allow_reset": true
}
},
"embedder": {
"provider": "openai",
"config": {
"model": "text-embedding-ada-002"
}
},
"chunker": {
"chunk_size": 2000,
"chunk_overlap": 100,
"length_function": "len"
}
}
```
```python config.py
config = {
'app': {
'config': {
'name': 'full-stack-app'
}
},
'llm': {
'provider': 'openai',
'config': {
'model': 'gpt-3.5-turbo',
'temperature': 0.5,
'max_tokens': 1000,
'top_p': 1,
'stream': False,
'template': (
"Use the following pieces of context to answer the query at the end.\n"
"If you don't know the answer, just say that you don't know, don't try to make up an answer.\n"
"$context\n\nQuery: $query\n\nHelpful Answer:"
),
'system_prompt': (
"Act as William Shakespeare. Answer the following questions in the style of William Shakespeare."
)
}
},
'vectordb': {
'provider': 'chroma',
'config': {
'collection_name': 'full-stack-app',
'dir': 'db',
'allow_reset': True
}
},
'embedder': {
'provider': 'openai',
'config': {
'model': 'text-embedding-ada-002'
}
},
'chunker': {
'chunk_size': 2000,
'chunk_overlap': 100,
'length_function': 'len'
}
}
```
</CodeGroup>
Alright, let's dive into what each key means in the yaml config above:
1. `app` Section:
- `config`:
- `id` (String): The ID or name of your full-stack application.
2. `chunker` Section:
- `chunk_size` (Integer): The size of each chunk of text that is sent to the language model.
- `chunk_overlap` (Integer): The amount of overlap between each chunk of text.
- `length_function` (String): The function used to calculate the length of each chunk of text. In this case, it's set to 'len'. You can also use any function import directly as a string here.
3. `llm` Section:
- `name` (String): The name of your full-stack application.
- `id` (String): The id of your full-stack application.
<Note>Only use this to reload already created apps. We recommend users to not create their own ids.</Note>
- `collect_metrics` (Boolean): Indicates whether metrics should be collected for the app, defaults to `True`
- `log_level` (String): The log level for the app, defaults to `WARNING`
2. `llm` Section:
- `provider` (String): The provider for the language model, which is set to 'openai'. You can find the full list of llm providers in [our docs](/components/llms).
- `model` (String): The specific model being used, 'gpt-3.5-turbo'.
- `config`:
- `model` (String): The specific model being used, 'gpt-3.5-turbo'.
- `temperature` (Float): Controls the randomness of the model's output. A higher value (closer to 1) makes the output more random.
- `max_tokens` (Integer): Controls how many tokens are used in the response.
- `top_p` (Float): Controls the diversity of word selection. A higher value (closer to 1) makes word selection more diverse.
- `stream` (Boolean): Controls if the response is streamed back to the user (set to false).
- `template` (String): A custom template for the prompt that the model uses to generate responses.
- `system_prompt` (String): A system prompt for the model to follow when generating responses, in this case, it's set to the style of William Shakespeare.
4. `vectordb` Section:
- `stream` (Boolean): Controls if the response is streamed back to the user (set to false).
- `number_documents` (Integer): Number of documents to pull from the vectordb as context, defaults to 1
3. `vectordb` Section:
- `provider` (String): The provider for the vector database, set to 'chroma'. You can find the full list of vector database providers in [our docs](/components/vector-databases).
- `config`:
- `collection_name` (String): The initial collection name for the database, set to 'full-stack-app'.
- `dir` (String): The directory for the database, set to 'db'.
- `allow_reset` (Boolean): Indicates whether resetting the database is allowed, set to true.
5. `embedder` Section:
- `collection_name` (String): The initial collection name for the vectordb, set to 'full-stack-app'.
- `dir` (String): The directory for the local database, set to 'db'.
- `allow_reset` (Boolean): Indicates whether resetting the vectordb is allowed, set to true.
<Note>We recommend you to checkout vectordb specific config [here](https://docs.embedchain.ai/components/vector-databases)</Note>
4. `embedder` Section:
- `provider` (String): The provider for the embedder, set to 'openai'. You can find the full list of embedding model providers in [our docs](/components/embedding-models).
- `config`:
- `model` (String): The specific model used for text embedding, 'text-embedding-ada-002'.
5. `chunker` Section:
- `chunk_size` (Integer): The size of each chunk of text that is sent to the language model.
- `chunk_overlap` (Integer): The amount of overlap between each chunk of text.
- `length_function` (String): The function used to calculate the length of each chunk of text. In this case, it's set to 'len'. You can also use any function import directly as a string here.
If you have questions about the configuration above, please feel free to reach out to us using one of the following methods:
<Snippet file="get-help.mdx" />

10
docs/components/embedding-models.mdx
View File

@@ -29,7 +29,7 @@ from embedchain import Pipeline as App
os.environ['OPENAI_API_KEY'] = 'xxx'
# load embedding model configuration from config.yaml file
app = App.from_config(yaml_path="config.yaml")
app = App.from_config(config_path="config.yaml")
app.add("https://en.wikipedia.org/wiki/OpenAI")
app.query("What is OpenAI?")
@@ -59,7 +59,7 @@ os.environ["AZURE_OPENAI_ENDPOINT"] = "https://xxx.openai.azure.com/"
os.environ["AZURE_OPENAI_API_KEY"] = "xxx"
os.environ["OPENAI_API_VERSION"] = "xxx"
app = App.from_config(yaml_path="config.yaml")
app = App.from_config(config_path="config.yaml")
```
```yaml config.yaml
@@ -93,7 +93,7 @@ GPT4All supports generating high quality embeddings of arbitrary length document
from embedchain import Pipeline as App
# load embedding model configuration from config.yaml file
app = App.from_config(yaml_path="config.yaml")
app = App.from_config(config_path="config.yaml")
```
```yaml config.yaml
@@ -122,7 +122,7 @@ Hugging Face supports generating embeddings of arbitrary length documents of tex
from embedchain import Pipeline as App
# load embedding model configuration from config.yaml file
app = App.from_config(yaml_path="config.yaml")
app = App.from_config(config_path="config.yaml")
```
```yaml config.yaml
@@ -153,7 +153,7 @@ Embedchain supports Google's VertexAI embeddings model through a simple interfac
from embedchain import Pipeline as App
# load embedding model configuration from config.yaml file
app = App.from_config(yaml_path="config.yaml")
app = App.from_config(config_path="config.yaml")
```
```yaml config.yaml

18
docs/components/llms.mdx
View File

@@ -46,7 +46,7 @@ from embedchain import Pipeline as App
os.environ['OPENAI_API_KEY'] = 'xxx'
# load llm configuration from config.yaml file
app = App.from_config(yaml_path="config.yaml")
app = App.from_config(config_path="config.yaml")
```
```yaml config.yaml
@@ -78,7 +78,7 @@ os.environ["OPENAI_API_BASE"] = "https://xxx.openai.azure.com/"
os.environ["OPENAI_API_KEY"] = "xxx"
os.environ["OPENAI_API_VERSION"] = "xxx"
app = App.from_config(yaml_path="config.yaml")
app = App.from_config(config_path="config.yaml")
```
```yaml config.yaml
@@ -115,7 +115,7 @@ from embedchain import Pipeline as App
os.environ["ANTHROPIC_API_KEY"] = "xxx"
# load llm configuration from config.yaml file
app = App.from_config(yaml_path="config.yaml")
app = App.from_config(config_path="config.yaml")
```
```yaml config.yaml
@@ -152,7 +152,7 @@ from embedchain import Pipeline as App
os.environ["COHERE_API_KEY"] = "xxx"
# load llm configuration from config.yaml file
app = App.from_config(yaml_path="config.yaml")
app = App.from_config(config_path="config.yaml")
```
```yaml config.yaml
@@ -183,7 +183,7 @@ GPT4all is a free-to-use, locally running, privacy-aware chatbot. No GPU or inte
from embedchain import Pipeline as App
# load llm configuration from config.yaml file
app = App.from_config(yaml_path="config.yaml")
app = App.from_config(config_path="config.yaml")
```
```yaml config.yaml
@@ -216,7 +216,7 @@ from embedchain import Pipeline as App
os.environ["JINACHAT_API_KEY"] = "xxx"
# load llm configuration from config.yaml file
app = App.from_config(yaml_path="config.yaml")
app = App.from_config(config_path="config.yaml")
```
```yaml config.yaml
@@ -253,7 +253,7 @@ from embedchain import Pipeline as App
os.environ["HUGGINGFACE_ACCESS_TOKEN"] = "xxx"
# load llm configuration from config.yaml file
app = App.from_config(yaml_path="config.yaml")
app = App.from_config(config_path="config.yaml")
```
```yaml config.yaml
@@ -283,7 +283,7 @@ from embedchain import Pipeline as App
os.environ["REPLICATE_API_TOKEN"] = "xxx"
# load llm configuration from config.yaml file
app = App.from_config(yaml_path="config.yaml")
app = App.from_config(config_path="config.yaml")
```
```yaml config.yaml
@@ -308,7 +308,7 @@ Setup Google Cloud Platform application credentials by following the instruction
from embedchain import Pipeline as App
# load llm configuration from config.yaml file
app = App.from_config(yaml_path="config.yaml")
app = App.from_config(config_path="config.yaml")
```
```yaml config.yaml

21
docs/components/vector-databases.mdx
View File

@@ -25,7 +25,7 @@ Utilizing a vector database alongside Embedchain is a seamless process. All you
from embedchain import Pipeline as App
# load chroma configuration from yaml file
app = App.from_config(yaml_path="config1.yaml")
app = App.from_config(config_path="config1.yaml")
```
```yaml config1.yaml
@@ -64,7 +64,7 @@ pip install --upgrade 'embedchain[elasticsearch]'
from embedchain import Pipeline as App
# load elasticsearch configuration from yaml file
app = App.from_config(yaml_path="config.yaml")
app = App.from_config(config_path="config.yaml")
```
```yaml config.yaml
@@ -73,8 +73,11 @@ vectordb:
config:
collection_name: 'es-index'
es_url: http://localhost:9200
allow_reset: true
http_auth:
- admin
- admin
api_key: xxx
verify_certs: false
```
</CodeGroup>
@@ -92,19 +95,19 @@ pip install --upgrade 'embedchain[opensearch]'
from embedchain import Pipeline as App
# load opensearch configuration from yaml file
app = App.from_config(yaml_path="config.yaml")
app = App.from_config(config_path="config.yaml")
```
```yaml config.yaml
vectordb:
provider: opensearch
config:
collection_name: 'my-app'
opensearch_url: 'https://localhost:9200'
http_auth:
- admin
- admin
vector_dimension: 1536
collection_name: 'my-app'
use_ssl: false
verify_certs: false
```
@@ -131,7 +134,7 @@ os.environ['ZILLIZ_CLOUD_URI'] = 'https://xxx.zillizcloud.com'
os.environ['ZILLIZ_CLOUD_TOKEN'] = 'xxx'
# load zilliz configuration from yaml file
app = App.from_config(yaml_path="config.yaml")
app = App.from_config(config_path="config.yaml")
```
```yaml config.yaml
@@ -167,7 +170,7 @@ In order to use Pinecone as vector database, set the environment variables `PINE
from embedchain import Pipeline as App
# load pinecone configuration from yaml file
app = App.from_config(yaml_path="config.yaml")
app = App.from_config(config_path="config.yaml")
```
```yaml config.yaml
@@ -190,7 +193,7 @@ In order to use Qdrant as a vector database, set the environment variables `QDRA
from embedchain import Pipeline as App
# load qdrant configuration from yaml file
app = App.from_config(yaml_path="config.yaml")
app = App.from_config(config_path="config.yaml")
```
```yaml config.yaml
@@ -210,7 +213,7 @@ In order to use Weaviate as a vector database, set the environment variables `WE
from embedchain import Pipeline as App
# load weaviate configuration from yaml file
app = App.from_config(yaml_path="config.yaml")
app = App.from_config(config_path="config.yaml")
```
```yaml config.yaml

12
docs/data-sources/data-type-handling.mdx
View File

@@ -50,3 +50,15 @@ from embedchain import Pipeline as App
naval_chat_bot = App()
print(naval_chat_bot.query("What unique capacity does Naval argue humans possess when it comes to understanding explanations or concepts?"))
```
## Resetting an app and vector database
You can reset the app by simply calling the `reset` method. This will delete the vector database and all other app related files.
```python
from embedchain import Pipeline as App
app = App()
app.add("https://www.youtube.com/watch?v=3qHkcs3kG44")
app.reset()
```

2
docs/data-sources/docs-site.mdx
View File

@@ -1,5 +1,5 @@
---
title: '📚🌐 Code documentation'
title: '📚 Code documentation'
---
To add any code documentation website as a loader, use the data_type as `docs_site`. Eg:

12
docs/data-sources/overview.mdx
View File

@@ -5,20 +5,20 @@ title: Overview
Embedchain comes with built-in support for various data sources. We handle the complexity of loading unstructured data from these data sources, allowing you to easily customize your app through a user-friendly interface.
<CardGroup cols={4}>
<Card title="📊 csv" href="/data-sources/csv"></Card>
<Card title="📊 CSV" href="/data-sources/csv"></Card>
<Card title="📃 JSON" href="/data-sources/json"></Card>
<Card title="📚🌐 docs site" href="/data-sources/docs-site"></Card>
<Card title="📚 docs site" href="/data-sources/docs-site"></Card>
<Card title="📄 docx" href="/data-sources/docx"></Card>
<Card title="📝 mdx" href="/data-sources/mdx"></Card>
<Card title="📓 notion" href="/data-sources/notion"></Card>
<Card title="📰 pdf" href="/data-sources/pdf-file"></Card>
<Card title="📓 Notion" href="/data-sources/notion"></Card>
<Card title="📰 PDF" href="/data-sources/pdf-file"></Card>
<Card title="❓💬 q&a pair" href="/data-sources/qna"></Card>
<Card title="🗺️ sitemap" href="/data-sources/sitemap"></Card>
<Card title="📝 text" href="/data-sources/text"></Card>
<Card title="🌐📄 web page" href="/data-sources/web-page"></Card>
<Card title="🌐 web page" href="/data-sources/web-page"></Card>
<Card title="🧾 xml" href="/data-sources/xml"></Card>
<Card title="🙌 OpenAPI" href="/data-sources/openapi"></Card>
<Card title="📺 youtube video" href="/data-sources/youtube-video"></Card>
<Card title="📺 Youtube" href="/data-sources/youtube-video"></Card>
<Card title="📬 Gmail" href="/data-sources/gmail"></Card>
<Card title="🐘 Postgres" href="/data-sources/postgres"></Card>
<Card title="🐬 MySQL" href="/data-sources/mysql"></Card>

2
docs/data-sources/web-page.mdx
View File

@@ -1,5 +1,5 @@
---
title: '🌐📄 Web page'
title: '🌐 Web page'
---
To add any web page, use the data_type as `web_page`. Eg:

2
docs/data-sources/youtube-video.mdx
View File

@@ -1,5 +1,5 @@
---
title: '📺 Youtube video'
title: '📺 Youtube'
---

9
docs/get-started/examples.mdx
View File

@@ -1,8 +1,15 @@
---
title: 🔎 Examples
description: 'Collection of Google colab notebook and Replit links for users'
---
# Explore awesome apps
Check out the remarkable work accomplished using [Embedchain](https://app.embedchain.ai/custom-gpts/).
## Collection of Google colab notebook and Replit links for users
Get started with Embedchain by trying out the examples below. You can run the examples in your browser using Google Colab or Replit.
<table>
<thead>
<tr>

53
docs/get-started/faq.mdx
View File

@@ -2,13 +2,36 @@
title: ❓ FAQs
description: 'Collections of all the frequently asked questions'
---
#### Does Embedchain support OpenAI's Assistant APIs?
<AccordionGroup>
<Accordion title="Does Embedchain support OpenAI's Assistant APIs?">
Yes, it does. Please refer to the [OpenAI Assistant docs page](/get-started/openai-assistant).
</Accordion>
<Accordion title="How to use MistralAI language model?">
Use the model provided on huggingface: `mistralai/Mistral-7B-v0.1`
<CodeGroup>
```python main.py
import os
from embedchain import Pipeline as App
#### How to use `gpt-4-turbo` model released on OpenAI DevDay?
os.environ["OPENAI_API_KEY"] = "sk-xxx"
os.environ["HUGGINGFACE_ACCESS_TOKEN"] = "hf_your_token"
app = App.from_config("huggingface.yaml")
```
```yaml huggingface.yaml
llm:
provider: huggingface
config:
model: 'mistralai/Mistral-7B-v0.1'
temperature: 0.5
max_tokens: 1000
top_p: 0.5
stream: false
```
</CodeGroup>
</Accordion>
<Accordion title="How to use ChatGPT 4 turbo model released on OpenAI DevDay?">
Use the model `gpt-4-turbo` provided my openai.
<CodeGroup>
```python main.py
@@ -18,7 +41,7 @@ from embedchain import Pipeline as App
os.environ['OPENAI_API_KEY'] = 'xxx'
# load llm configuration from gpt4_turbo.yaml file
app = App.from_config(yaml_path="gpt4_turbo.yaml")
app = App.from_config(config_path="gpt4_turbo.yaml")
```
```yaml gpt4_turbo.yaml
@@ -31,12 +54,9 @@ llm:
top_p: 1
stream: false
```
</CodeGroup>
#### How to use GPT-4 as the LLM model?
</Accordion>
<Accordion title="How to use GPT-4 as the LLM model?">
<CodeGroup>
```python main.py
@@ -46,7 +66,7 @@ from embedchain import Pipeline as App
os.environ['OPENAI_API_KEY'] = 'xxx'
# load llm configuration from gpt4.yaml file
app = App.from_config(yaml_path="gpt4.yaml")
app = App.from_config(config_path="gpt4.yaml")
```
```yaml gpt4.yaml
@@ -61,9 +81,8 @@ llm:
```
</CodeGroup>
#### I don't have OpenAI credits. How can I use some open source model?
</Accordion>
<Accordion title="I don't have OpenAI credits. How can I use some open source model?">
<CodeGroup>
```python main.py
@@ -73,7 +92,7 @@ from embedchain import Pipeline as App
os.environ['OPENAI_API_KEY'] = 'xxx'
# load llm configuration from opensource.yaml file
app = App.from_config(yaml_path="opensource.yaml")
app = App.from_config(config_path="opensource.yaml")
```
```yaml opensource.yaml
@@ -93,8 +112,10 @@ embedder:
```
</CodeGroup>
#### How to contact support?
</Accordion>
</AccordionGroup>
#### Need more help?
If docs aren't sufficient, please feel free to reach out to us using one of the following methods:
<Snippet file="get-help.mdx" />

4
docs/get-started/introduction.mdx
View File

@@ -105,7 +105,7 @@ app.deploy()
# ✅ Data of type: web_page, value: https://www.forbes.com/profile/elon-musk added successfully.
```
## 🚀 How it works?
## 🛠️ How it works?
Embedchain abstracts out the following steps from you to easily create LLM powered apps:
@@ -129,3 +129,5 @@ The process of loading the dataset and querying involves multiple steps, each wi
- How should I find similar documents for a query? Which ranking model should I use?
Embedchain takes care of all these nuances and provides a simple interface to create apps on any data.
## [🚀 Get started](https://docs.embedchain.ai/get-started/quickstart)

110
docs/get-started/quickstart.mdx
View File

@@ -12,79 +12,73 @@ pip install embedchain
```
<Tip>
Embedchain now supports OpenAI's latest `gpt-4-turbo` model. Checkout the [docs here](/get-started/faq#how-to-use-gpt-4-turbo-model-released-on-openai-devday) on how to use it.
Embedchain now supports OpenAI's latest `gpt-4-turbo` model. Checkout the [FAQs](/get-started/faq#how-to-use-gpt-4-turbo-model-released-on-openai-devday).
</Tip>
Creating an app involves 3 steps:
<Steps>
<Step title="⚙️ Import app instance">
```python
from embedchain import Pipeline as App
app = App()
```
```python
from embedchain import Pipeline as App
app = App()
```
<Accordion title="Customize your app by a simple YAML config" icon="gear-complex">
Embedchain provides a wide range of options to customize your app. You can customize the model, data sources, and much more.
Explore the custom configurations [here](https://docs.embedchain.ai/advanced/configuration).
```python
from embedchain import Pipeline as App
app = App(yaml_config="config.yaml")
```
</Accordion>
</Step>
<Step title="🗃️ Add data sources">
```python
# Add different data sources
app.add("https://en.wikipedia.org/wiki/Elon_Musk")
app.add("https://www.forbes.com/profile/elon-musk")
# You can also add local data sources such as pdf, csv files etc.
# app.add("/path/to/file.pdf")
```
```python
app.add("https://en.wikipedia.org/wiki/Elon_Musk")
app.add("https://www.forbes.com/profile/elon-musk")
# app.add("path/to/file/elon_musk.pdf")
```
<Accordion title="Embedchain supports adding data from many data sources." icon="files">
Embedchain supports adding data from many data sources including web pages, PDFs, databases, and more.
Explore the list of supported [data sources](https://docs.embedchain.ai/data-sources/overview).
</Accordion>
</Step>
<Step title="💬 Query or chat or search context on your data">
```python
app.query("What is the net worth of Elon Musk today?")
# Answer: The net worth of Elon Musk today is $258.7 billion.
```
<Step title="💬 Ask questions, chat, or search through your data with ease">
```python
app.query("What is the net worth of Elon Musk today?")
# Answer: The net worth of Elon Musk today is $258.7 billion.
```
<Accordion title="Want to chat with your app?" icon="face-thinking">
Embedchain provides a wide range of features to interact with your app. You can chat with your app, ask questions, search through your data, and much more.
```python
app.chat("How many companies does Elon Musk run? Name those")
# Answer: Elon Musk runs 3 companies: Tesla, SpaceX, and Neuralink.
app.chat("What is his net worth today?")
# Answer: The net worth of Elon Musk today is $258.7 billion.
```
To learn about other features, click [here](https://docs.embedchain.ai/get-started/introduction)
</Accordion>
</Step>
<Step title="🚀 (Optional) Deploy your pipeline to Embedchain Platform">
```python
app.deploy()
# 🔑 Enter your Embedchain API key. You can find the API key at https://app.embedchain.ai/settings/keys/
# ec-xxxxxx
<Step title="🚀 Seamlessly launch your App on the Embedchain Platform!">
```python
app.deploy()
# 🔑 Enter your Embedchain API key. You can find the API key at https://app.embedchain.ai/settings/keys/
# ec-xxxxxx
# 🛠️ Creating pipeline on the platform...
# 🎉🎉🎉 Pipeline created successfully! View your pipeline: https://app.embedchain.ai/pipelines/xxxxx
# 🛠️ Creating pipeline on the platform...
# 🎉🎉🎉 Pipeline created successfully! View your pipeline: https://app.embedchain.ai/pipelines/xxxxx
# 🛠️ Adding data to your pipeline...
# ✅ Data of type: web_page, value: https://www.forbes.com/profile/elon-musk added successfully.
```
# 🛠️ Adding data to your pipeline...
# ✅ Data of type: web_page, value: https://www.forbes.com/profile/elon-musk added successfully.
```
<Accordion title="Share your app with others" icon="laptop-mobile">
You can now share your app with others from our platform.
Access your app on our [platform](https://app.embedchain.ai/).
</Accordion>
</Step>
</Steps>
Putting it together, you can run your first app using the following code. Make sure to set the `OPENAI_API_KEY` 🔑 environment variable in the code.
```python
import os
from embedchain import Pipeline as App
os.environ["OPENAI_API_KEY"] = "xxx"
app = App()
# Add different data sources
app.add("https://en.wikipedia.org/wiki/Elon_Musk")
app.add("https://www.forbes.com/profile/elon-musk")
# You can also add local data sources such as pdf, csv files etc.
# app.add("/path/to/file.pdf")
response = app.query("What is the net worth of Elon Musk today?")
print(response)
# Answer: The net worth of Elon Musk today is $258.7 billion.
app.deploy()
# 🔑 Enter your Embedchain API key. You can find the API key at https://app.embedchain.ai/settings/keys/
# ec-xxxxxx
# 🛠️ Creating pipeline on the platform...
# 🎉🎉🎉 Pipeline created successfully! View your pipeline: https://app.embedchain.ai/pipelines/xxxxx
# 🛠️ Adding data to your pipeline...
# ✅ Data of type: web_page, value: https://www.forbes.com/profile/elon-musk added successfully.
```
You can try it out yourself using the following Google Colab notebook:
Putting it together, you can run your first app using the following Google Colab. Make sure to set the `OPENAI_API_KEY` 🔑 environment variable in the code.
<a href="https://colab.research.google.com/drive/17ON1LPonnXAtLaZEebnOktstB_1cJJmh?usp=sharing">
<img src="https://camo.githubusercontent.com/84f0493939e0c4de4e6dbe113251b4bfb5353e57134ffd9fcab6b8714514d4d1/68747470733a2f2f636f6c61622e72657365617263682e676f6f676c652e636f6d2f6173736574732f636f6c61622d62616467652e737667" alt="Open in Colab" />

4
embedchain/apps/app.py
View File

@@ -12,7 +12,7 @@ from embedchain.factory import EmbedderFactory, LlmFactory, VectorDBFactory
from embedchain.helpers.json_serializable import register_deserializable
from embedchain.llm.base import BaseLlm
from embedchain.llm.openai import OpenAILlm
from embedchain.utils import validate_yaml_config
from embedchain.utils import validate_config
from embedchain.vectordb.base import BaseVectorDB
from embedchain.vectordb.chroma import ChromaDB
@@ -134,7 +134,7 @@ class App(EmbedChain):
config_data = yaml.safe_load(file)
try:
validate_yaml_config(config_data)
validate_config(config_data)
except Exception as e:
raise Exception(f"❌ Error occurred while validating the YAML config. Error: {str(e)}")

5
embedchain/llm/huggingface.py
View File

@@ -1,4 +1,5 @@
import importlib
import logging
import os
from typing import Optional
@@ -42,9 +43,11 @@ class HuggingFaceLlm(BaseLlm):
else:
raise ValueError("`top_p` must be > 0.0 and < 1.0")
model = config.model or "google/flan-t5-xxl"
logging.info(f"Using HuggingFaceHub with model {model}")
llm = HuggingFaceHub(
huggingfacehub_api_token=os.environ["HUGGINGFACE_ACCESS_TOKEN"],
repo_id=config.model or "google/flan-t5-xxl",
repo_id=model,
model_kwargs=model_kwargs,
)

77
embedchain/pipeline.py
View File

@@ -4,6 +4,7 @@ import logging
import os
import sqlite3
import uuid
from typing import Any, Dict, Optional
import requests
import yaml
@@ -19,7 +20,7 @@ from embedchain.helpers.json_serializable import register_deserializable
from embedchain.llm.base import BaseLlm
from embedchain.llm.openai import OpenAILlm
from embedchain.telemetry.posthog import AnonymousTelemetry
from embedchain.utils import validate_yaml_config
from embedchain.utils import validate_config
from embedchain.vectordb.base import BaseVectorDB
from embedchain.vectordb.chroma import ChromaDB
@@ -43,7 +44,7 @@ class Pipeline(EmbedChain):
db: BaseVectorDB = None,
embedding_model: BaseEmbedder = None,
llm: BaseLlm = None,
yaml_path: str = None,
config_data: dict = None,
log_level=logging.WARN,
auto_deploy: bool = False,
chunker: ChunkerConfig = None,
@@ -59,15 +60,15 @@ class Pipeline(EmbedChain):
:type embedding_model: BaseEmbedder, optional
:param llm: The LLM model used to calculate embeddings, defaults to None
:type llm: BaseLlm, optional
:param yaml_path: Path to the YAML configuration file, defaults to None
:type yaml_path: str, optional
:param config_data: Config dictionary, defaults to None
:type config_data: dict, optional
:param log_level: Log level to use, defaults to logging.WARN
:type log_level: int, optional
:param auto_deploy: Whether to deploy the pipeline automatically, defaults to False
:type auto_deploy: bool, optional
:raises Exception: If an error occurs while creating the pipeline
"""
if id and yaml_path:
if id and config_data:
raise Exception("Cannot provide both id and config. Please provide only one of them.")
if id and name:
@@ -79,8 +80,8 @@ class Pipeline(EmbedChain):
logging.basicConfig(level=log_level, format="%(asctime)s - %(name)s - %(levelname)s - %(message)s")
self.logger = logging.getLogger(__name__)
self.auto_deploy = auto_deploy
# Store the yaml config as an attribute to be able to send it
self.yaml_config = None
# Store the dict config as an attribute to be able to send it
self.config_data = config_data if (config_data and validate_config(config_data)) else None
self.client = None
# pipeline_id from the backend
self.id = None
@@ -92,11 +93,6 @@ class Pipeline(EmbedChain):
self.name = self.config.name
self.config.id = self.local_id = str(uuid.uuid4()) if self.config.id is None else self.config.id
if yaml_path:
with open(yaml_path, "r") as file:
config_data = yaml.safe_load(file)
self.yaml_config = config_data
if id is not None:
# Init client first since user is trying to fetch the pipeline
# details from the platform
@@ -187,9 +183,9 @@ class Pipeline(EmbedChain):
Create a pipeline on the platform.
"""
print("🛠️ Creating pipeline on the platform...")
# self.yaml_config is a dict. Pass it inside the key 'yaml_config' to the backend
# self.config_data is a dict. Pass it inside the key 'yaml_config' to the backend
payload = {
"yaml_config": json.dumps(self.yaml_config),
"yaml_config": json.dumps(self.config_data),
"name": self.name,
"local_id": self.local_id,
}
@@ -346,24 +342,57 @@ class Pipeline(EmbedChain):
self.telemetry.capture(event_name="deploy", properties=self._telemetry_props)
@classmethod
def from_config(cls, yaml_path: str, auto_deploy: bool = False):
def from_config(
cls,
config_path: Optional[str] = None,
config: Optional[Dict[str, Any]] = None,
auto_deploy: bool = False,
yaml_path: Optional[str] = None,
):
"""
Instantiate a Pipeline object from a YAML configuration file.
Instantiate a Pipeline object from a configuration.
:param yaml_path: Path to the YAML configuration file.
:type yaml_path: str
:param config_path: Path to the YAML or JSON configuration file.
:type config_path: Optional[str]
:param config: A dictionary containing the configuration.
:type config: Optional[Dict[str, Any]]
:param auto_deploy: Whether to deploy the pipeline automatically, defaults to False
:type auto_deploy: bool, optional
:param yaml_path: (Deprecated) Path to the YAML configuration file. Use config_path instead.
:type yaml_path: Optional[str]
:return: An instance of the Pipeline class.
:rtype: Pipeline
"""
with open(yaml_path, "r") as file:
config_data = yaml.safe_load(file)
# Backward compatibility for yaml_path
if yaml_path and not config_path:
config_path = yaml_path
if config_path and config:
raise ValueError("Please provide only one of config_path or config.")
config_data = None
if config_path:
file_extension = os.path.splitext(config_path)[1]
with open(config_path, "r") as file:
if file_extension in [".yaml", ".yml"]:
config_data = yaml.safe_load(file)
elif file_extension == ".json":
config_data = json.load(file)
else:
raise ValueError("config_path must be a path to a YAML or JSON file.")
elif config and isinstance(config, dict):
config_data = config
else:
logging.error(
"Please provide either a config file path (YAML or JSON) or a config dictionary. Falling back to defaults because no config is provided.", # noqa: E501
)
config_data = {}
try:
validate_yaml_config(config_data)
validate_config(config_data)
except Exception as e:
raise Exception(f"❌ Error occurred while validating the YAML config. Error: {str(e)}")
raise Exception(f"Error occurred while validating the config. Error: {str(e)}")
pipeline_config_data = config_data.get("app", {}).get("config", {})
db_config_data = config_data.get("vectordb", {})
@@ -388,7 +417,7 @@ class Pipeline(EmbedChain):
)
# Send anonymous telemetry
event_properties = {"init_type": "yaml_config"}
event_properties = {"init_type": "config_data"}
AnonymousTelemetry().capture(event_name="init", properties=event_properties)
return cls(
@@ -396,7 +425,7 @@ class Pipeline(EmbedChain):
llm=llm,
db=db,
embedding_model=embedding_model,
yaml_path=yaml_path,
config_data=config_data,
auto_deploy=auto_deploy,
chunker=chunker_config_data,
)

2
embedchain/store/assistants.py
View File

@@ -165,7 +165,7 @@ class AIAssistant:
self.instructions = instructions
self.assistant_id = assistant_id or str(uuid.uuid4())
self.thread_id = thread_id or str(uuid.uuid4())
self.pipeline = Pipeline.from_config(yaml_path=yaml_path) if yaml_path else Pipeline()
self.pipeline = Pipeline.from_config(config_path=yaml_path) if yaml_path else Pipeline()
self.pipeline.local_id = self.pipeline.config.id = self.thread_id
if self.instructions:

2
embedchain/utils.py
View File

@@ -355,7 +355,7 @@ def is_valid_json_string(source: str):
return False
def validate_yaml_config(config_data):
def validate_config(config_data):
schema = Schema(
{
Optional("app"): {

12
examples/rest-api/main.py
View File

@@ -108,7 +108,7 @@ async def get_datasources_associated_with_app_id(app_id: str, db: Session = Depe
if db_app is None:
raise HTTPException(detail=f"App with id {app_id} does not exist, please create it first.", status_code=400)
app = App.from_config(yaml_path=db_app.config)
app = App.from_config(config_path=db_app.config)
response = app.get_data_sources()
return {"results": response}
@@ -147,7 +147,7 @@ async def add_datasource_to_an_app(body: SourceApp, app_id: str, db: Session = D
if db_app is None:
raise HTTPException(detail=f"App with id {app_id} does not exist, please create it first.", status_code=400)
app = App.from_config(yaml_path=db_app.config)
app = App.from_config(config_path=db_app.config)
response = app.add(source=body.source, data_type=body.data_type)
return DefaultResponse(response=response)
@@ -185,7 +185,7 @@ async def query_an_app(body: QueryApp, app_id: str, db: Session = Depends(get_db
if db_app is None:
raise HTTPException(detail=f"App with id {app_id} does not exist, please create it first.", status_code=400)
app = App.from_config(yaml_path=db_app.config)
app = App.from_config(config_path=db_app.config)
response = app.query(body.query)
return DefaultResponse(response=response)
@@ -227,7 +227,7 @@ async def query_an_app(body: QueryApp, app_id: str, db: Session = Depends(get_db
# status_code=400
# )
# app = App.from_config(yaml_path=db_app.config)
# app = App.from_config(config_path=db_app.config)
# response = app.chat(body.message)
# return DefaultResponse(response=response)
@@ -264,7 +264,7 @@ async def deploy_app(body: DeployAppRequest, app_id: str, db: Session = Depends(
if db_app is None:
raise HTTPException(detail=f"App with id {app_id} does not exist, please create it first.", status_code=400)
app = App.from_config(yaml_path=db_app.config)
app = App.from_config(config_path=db_app.config)
api_key = body.api_key
# this will save the api key in the embedchain.db
@@ -305,7 +305,7 @@ async def delete_app(app_id: str, db: Session = Depends(get_db)):
if db_app is None:
raise HTTPException(detail=f"App with id {app_id} does not exist, please create it first.", status_code=400)
app = App.from_config(yaml_path=db_app.config)
app = App.from_config(config_path=db_app.config)
# reset app.db
app.db.reset()

2
notebooks/anthropic.ipynb
View File

@@ -109,7 +109,7 @@
},
"outputs": [],
"source": [
"app = App.from_config(yaml_path=\"anthropic.yaml\")"
"app = App.from_config(config_path=\"anthropic.yaml\")"
]
},
{

2
notebooks/azure-openai.ipynb
View File

@@ -105,7 +105,7 @@
"metadata": {},
"outputs": [],
"source": [
"app = App.from_config(yaml_path=\"azure_openai.yaml\")"
"app = App.from_config(config_path=\"azure_openai.yaml\")"
]
},
{

2
notebooks/chromadb.ipynb
View File

@@ -105,7 +105,7 @@
},
"outputs": [],
"source": [
"app = App.from_config(yaml_path=\"chromadb.yaml\")"
"app = App.from_config(config_path=\"chromadb.yaml\")"
]
},
{

2
notebooks/cohere.ipynb
View File

@@ -114,7 +114,7 @@
},
"outputs": [],
"source": [
"app = App.from_config(yaml_path=\"cohere.yaml\")"
"app = App.from_config(config_path=\"cohere.yaml\")"
]
},
{

2
notebooks/elasticsearch.ipynb
View File

@@ -103,7 +103,7 @@
},
"outputs": [],
"source": [
"app = App.from_config(yaml_path=\"elasticsearch.yaml\")"
"app = App.from_config(config_path=\"elasticsearch.yaml\")"
]
},
{

2
notebooks/gpt4all.ipynb
View File

@@ -114,7 +114,7 @@
},
"outputs": [],
"source": [
"app = App.from_config(yaml_path=\"gpt4all.yaml\")"
"app = App.from_config(config_path=\"gpt4all.yaml\")"
]
},
{

2
notebooks/hugging_face_hub.ipynb
View File

@@ -114,7 +114,7 @@
},
"outputs": [],
"source": [
"app = App.from_config(yaml_path=\"huggingface.yaml\")"
"app = App.from_config(config_path=\"huggingface.yaml\")"
]
},
{

2
notebooks/jina.ipynb
View File

@@ -114,7 +114,7 @@
},
"outputs": [],
"source": [
"app = App.from_config(yaml_path=\"jina.yaml\")"
"app = App.from_config(config_path=\"jina.yaml\")"
]
},
{

2
notebooks/llama2.ipynb
View File

@@ -109,7 +109,7 @@
},
"outputs": [],
"source": [
"app = App.from_config(yaml_path=\"llama2.yaml\")"
"app = App.from_config(config_path=\"llama2.yaml\")"
]
},
{

2
notebooks/openai.ipynb
View File

@@ -115,7 +115,7 @@
},
"outputs": [],
"source": [
"app = App.from_config(yaml_path=\"openai.yaml\")"
"app = App.from_config(config_path=\"openai.yaml\")"
]
},
{

2
notebooks/opensearch.ipynb
View File

@@ -107,7 +107,7 @@
},
"outputs": [],
"source": [
"app = App.from_config(yaml_path=\"opensearch.yaml\")"
"app = App.from_config(config_path=\"opensearch.yaml\")"
]
},
{

2
notebooks/pinecone.ipynb
View File

@@ -104,7 +104,7 @@
},
"outputs": [],
"source": [
"app = App.from_config(yaml_path=\"pinecone.yaml\")"
"app = App.from_config(config_path=\"pinecone.yaml\")"
]
},
{

2
notebooks/vertex_ai.ipynb
View File

@@ -117,7 +117,7 @@
},
"outputs": [],
"source": [
"app = App.from_config(yaml_path=\"vertexai.yaml\")"
"app = App.from_config(config_path=\"vertexai.yaml\")"
]
},
{

2
pyproject.toml
View File

@@ -1,6 +1,6 @@
[tool.poetry]
name = "embedchain"
version = "0.1.22"
version = "0.1.23"
description = "Data platform for LLMs - Load, index, retrieve and sync any unstructured data"
authors = [
"Taranjeet Singh <taranjeet@embedchain.ai>",

4
tests/test_utils.py
View File

@@ -1,6 +1,6 @@
import yaml
from embedchain.utils import validate_yaml_config
from embedchain.utils import validate_config
CONFIG_YAMLS = [
"configs/anthropic.yaml",
@@ -30,7 +30,7 @@ def test_all_config_yamls():
assert config is not None
try:
validate_yaml_config(config)
validate_config(config)
except Exception as e:
print(f"Error in {config_yaml}: {e}")
raise e