my-project-public

repository

loading code, commits, and activity

repositories

loading repo index

#1	---
#2	title: "App"
#3	---
#4
#5	Create a RAG app object on Embedchain. This is the main entrypoint for a developer to interact with Embedchain APIs. An app configures the llm, vector database, embedding model, and retrieval strategy of your choice.
#6
#7	### Attributes
#8
#9	<ParamField path="local_id" type="str">
#10	App ID
#11	</ParamField>
#12	<ParamField path="name" type="str" optional>
#13	Name of the app
#14	</ParamField>
#15	<ParamField path="config" type="BaseConfig">
#16	Configuration of the app
#17	</ParamField>
#18	<ParamField path="llm" type="BaseLlm">
#19	Configured LLM for the RAG app
#20	</ParamField>
#21	<ParamField path="db" type="BaseVectorDB">
#22	Configured vector database for the RAG app
#23	</ParamField>
#24	<ParamField path="embedding_model" type="BaseEmbedder">
#25	Configured embedding model for the RAG app
#26	</ParamField>
#27	<ParamField path="chunker" type="ChunkerConfig">
#28	Chunker configuration
#29	</ParamField>
#30	<ParamField path="client" type="Client" optional>
#31	Client object (used to deploy an app to Embedchain platform)
#32	</ParamField>
#33	<ParamField path="logger" type="logging.Logger">
#34	Logger object
#35	</ParamField>
#36
#37	## Usage
#38
#39	You can create an app instance using the following methods:
#40
#41	### Default setting
#42
#43	```python Code Example
#44	from embedchain import App
#45	app = App()
#46	```
#47
#48
#49	### Python Dict
#50
#51	```python Code Example
#52	from embedchain import App
#53
#54	config_dict = {
#55	'llm': {
#56	'provider': 'gpt4all',
#57	'config': {
#58	'model': 'orca-mini-3b-gguf2-q4_0.gguf',
#59	'temperature': 0.5,
#60	'max_tokens': 1000,
#61	'top_p': 1,
#62	'stream': False
#63	}
#64	},
#65	'embedder': {
#66	'provider': 'gpt4all'
#67	}
#68	}
#69
#70	# load llm configuration from config dict
#71	app = App.from_config(config=config_dict)
#72	```
#73
#74	### YAML Config
#75
#76	<CodeGroup>
#77
#78	```python main.py
#79	from embedchain import App
#80
#81	# load llm configuration from config.yaml file
#82	app = App.from_config(config_path="config.yaml")
#83	```
#84
#85	```yaml config.yaml
#86	llm:
#87	provider: gpt4all
#88	config:
#89	model: 'orca-mini-3b-gguf2-q4_0.gguf'
#90	temperature: 0.5
#91	max_tokens: 1000
#92	top_p: 1
#93	stream: false
#94
#95	embedder:
#96	provider: gpt4all
#97	```
#98
#99	</CodeGroup>
#100
#101	### JSON Config
#102
#103	<CodeGroup>
#104
#105	```python main.py
#106	from embedchain import App
#107
#108	# load llm configuration from config.json file
#109	app = App.from_config(config_path="config.json")
#110	```
#111
#112	```json config.json
#113	{
#114	"llm": {
#115	"provider": "gpt4all",
#116	"config": {
#117	"model": "orca-mini-3b-gguf2-q4_0.gguf",
#118	"temperature": 0.5,
#119	"max_tokens": 1000,
#120	"top_p": 1,
#121	"stream": false
#122	}
#123	},
#124	"embedder": {
#125	"provider": "gpt4all"
#126	}
#127	}
#128	```
#129
#130	</CodeGroup>
#131

z6Mkq5mY3JWtxoxUobWcfNHm7AkRubgSWEZTkBVqZXJviFZ5/my-project-public