Skip to content

chore: update examples and readme #40

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Apr 26, 2024
Merged

chore: update examples and readme #40

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
131 changes: 87 additions & 44 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,129 +1,172 @@
# TaskingAI-client

The TaskingAI Python client for creating and managing AI-driven applications.
The official TaskingAI Python client.

For more information, see the docs at [TaskingAI Documentation](https://docs.tasking.ai/)

## Prerequisites

The TaskingAI client is compatible with Python 3.8 and above.

## Installation

Install the latest released version using pip:
Use `pip` to install the TaskingAI Python client.

```shell
# Install the latest version
pip install taskingai

# Install a specific version
pip install taskingai==0.2.2
```

## Usage

### Initialization

Before you can use the TaskingAI SDK, you must have your TaskingAI project set up and running. For community version, visit [TaskingAI Community](https://www.github.com/taskingai/taskingai) to get started. For cloud version, visit [TaskingAI Cloud](https://www.tasking.ai) to sign up first.

You need to initialize the TaskingAI Python client with an API key you obtain from the TaskingAI console. You can set the API key as an environment variable or pass it directly to the `init` function.

#### Using environment variables (Recommended)

Set it as an environment variable on your local system, and the SDK will automatically load the key without passing the `api_key` parameter in the `init` function.

```shell
pip3 install taskingai
export TASKINGAI_API_KEY=$YOUR_API_KEY
```

When you run your Python script, the SDK will automatically pick up the API key from the environment variable.

```python
import taskingai
# taskingai.init()
# No need to initialize the SDK with the API key
```

#### Passing the API key directly

You can also specify an API key to the SDK by passing it as a parameter to the init function:

```python
import taskingai
taskingai.init(api_key="YOUR_API_KEY")
```

## Quickstart
If you use community version, you can set the base URL to the TaskingAI server by passing it to the `init` function:

Here's how you can quickly start building and managing AI-driven applications using the TaskingAI client.
```python
import taskingai
taskingai.init(api_key="YOUR_API_KEY", host="http://localhost:8080")
```

### Assistants

Explore the ease of creating and customizing your own AI assistants with TaskingAI to enhance user interactions.
The Assistant system in TaskingAI represents a sophisticated framework designed to create and manage AI agents with customizable functionalities.

Here is an example of how to create, update, and delete an assistant:

```python
import taskingai
from taskingai.assistant import *
from taskingai.assistant.memory import AssistantNaiveMemory

# Initialize your API key if you haven't already set it in the environment
taskingai.init(api_key="YOUR_API_KEY")

# Create an assistant
assistant = create_assistant(
asst = taskingai.assistant.create_assistant(
model_id="YOUR_MODEL_ID",
memory=AssistantNaiveMemory(),
memory={"type": "naive"},
system_prompt_template=["You are a professional assistant."],
)
print(f"Assistant created: {assistant.id}")
print(f"Assistant created: {asst.assistant_id}")

# Get details about the assistant
assistant_details = get_assistant(assistant_id=assistant.id)
assistant_details = taskingai.assistant.get_assistant(assistant_id=asst.assistant_id)
print(f"Assistant details: {assistant_details}")

# Update the assistant's description
update_assistant(
assistant_id=assistant.id,
description="An updated description for my assistant."
taskingai.assistant.update_assistant(
assistant_id=asst.assistant_id,
description="Updated description"
)
print(f"Assistant updated.")

# Delete the assistant when done
delete_assistant(assistant_id=assistant.id)
taskingai.assistant.delete_assistant(assistant_id=asst.assistant_id)
print("Assistant deleted successfully.")
```

### Retrieval

Leverage TaskingAI's retrieval capabilities to store, manage, and extract information, making your applications smarter and more responsive.
TaskingAI offers comprehensive tools for the retrieval system, ranging from straightforward to intricate setups. Here is an example of how to create, add, retrieve, and delete a record in a collection:

```python
import taskingai
from taskingai.retrieval import *

# Create a collection for storing and retrieving data
collection = create_collection(
embedding_model_id="YOUR_MODEL_ID",
coll = taskingai.retrieval.create_collection(
embedding_model_id="YOUR_EMBEDDING_MODEL_ID",
capacity=1000
)
print(f"Collection created: {collection.id}")
print(f"Collection created: {coll.collection_id}")

# Add a record to the collection
record = create_record(
collection_id=collection.id,
content="Example text for machine learning.",
text_splitter=TokenTextSplitter(chunk_size=200, chunk_overlap=20),
record = taskingai.retrieval.create_record(
collection_id=coll.collection_id,
type="text",
content="Machine learning is ...",
text_splitter={"type": "token", "chunk_size": 200, "chunk_overlap": 20}
)
print(f"Record added to collection: {record.id}")
print(f"Record added to collection: {record.record_id}")

# Retrieve the record from the collection
retrieved_record = get_record(
collection_id=collection.id,
record_id=record.id
retrieved_record = taskingai.retrieval.get_record(
collection_id=coll.collection_id,
record_id=record.record_id
)
print(f"Record retrieved: {retrieved_record.text}")
print(f"Record retrieved: {retrieved_record.content}")

# Delete the record
delete_record(
collection_id=collection.id,
record_id=record.id
taskingai.retrieval.delete_record(
collection_id=coll.collection_id,
record_id=record.record_id
)
print("Record deleted.")

# Delete the collection
delete_collection(collection_id=collection.id)
taskingai.retrieval.delete_collection(collection_id=coll.collection_id)
print("Collection deleted.")
```

### Tools

Utilize TaskingAI's tools to create actions that enable your assistant to interact with external APIs and services, enriching the user experience.
The Tools module in TaskingAI is an essential suite designed to augment the capabilities of TaskingAI agents. Here is an example of how to create, run, and delete a tool action:

```python
import taskingai
from taskingai.tool import *

# Define a schema for the tool action
NUMBERS_API_SCHEMA = {
OPENAPI_SCHEMA = {
# Schema definition goes here
}

# Create a tool action based on the defined schema
actions = bulk_create_actions(
openapi_schema=NUMBERS_API_SCHEMA,
authentication=ActionAuthentication(type=ActionAuthenticationType.NONE)
actions = taskingai.tool.bulk_create_actions(
openapi_schema=OPENAPI_SCHEMA,
authentication={"type": "none"},
)
action = actions[0]
print(f"Action created: {action.id}")
print(f"Action created: {action.action_id}")

# Run the action for a test purpose
result = run_action(
action_id=action.id,
result = taskingai.tool.run_action(
action_id=action.action_id,
parameters={"number": 42}
)
print(f"Action result: {result}")

# Delete the action when done
delete_action(action_id=action.id)
taskingai.tool.delete_action(action_id=action.action_id)
print("Action deleted.")
```

Expand Down
6 changes: 1 addition & 5 deletions examples/assistant/chat_with_assistant.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -111,7 +111,7 @@
"from taskingai.assistant.memory import AssistantMessageWindowMemory\n",
"\n",
"# choose an available chat_completion model from your project\n",
"model_id = \"YOUR_MODEL_ID\"\n",
"model_id = \"YOUR_CHAT_COMPLETION_MODEL_ID\"\n",
"\n",
"assistant: Assistant = taskingai.assistant.create_assistant(\n",
" model_id=model_id,\n",
Expand All @@ -129,10 +129,6 @@
" ToolRef(\n",
" type=ToolType.ACTION,\n",
" id=action.action_id,\n",
" ), \n",
" ToolRef(\n",
" type=ToolType.ACTION,\n",
" id=action.action_id,\n",
" )\n",
" ],\n",
" retrievals=[],\n",
Expand Down
64 changes: 30 additions & 34 deletions examples/crud/assistant_crud.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,11 +25,8 @@
"metadata": {},
"outputs": [],
"source": [
"from taskingai.assistant import Assistant, Chat\n",
"from taskingai.assistant.memory import AssistantNaiveMemory\n",
"\n",
"# choose an available chat_completion model from your project\n",
"model_id = \"YOUR_MODEL_ID\""
"model_id = \"YOUR_CHAT_COMPLETION_MODEL_ID\""
]
},
{
Expand All @@ -47,24 +44,27 @@
"metadata": {},
"outputs": [],
"source": [
"from taskingai.assistant import RetrievalConfig, RetrievalMethod\n",
"\n",
"# create an assistant\n",
"def create_assistant() -> Assistant:\n",
" assistant: Assistant = taskingai.assistant.create_assistant(\n",
"def create_assistant():\n",
" assistant = taskingai.assistant.create_assistant(\n",
" model_id=model_id,\n",
" name=\"Customer Service Assistant\",\n",
" description=\"A professional assistant for customer service.\",\n",
" system_prompt_template=[\"You are a professional customer service assistant speaking {{language}}.\"],\n",
" memory=AssistantNaiveMemory(),\n",
" memory={\"type\": \"naive\",},\n",
" tools=[],\n",
" retrievals=[],\n",
" retrieval_configs=RetrievalConfig(top_k=3, max_tokens=4096, method=RetrievalMethod.USER_MESSAGE),\n",
" retrieval_configs={\n",
" \"top_k\": 3, \n",
" \"max_tokens\": 4096,\n",
" \"method\": \"user_message\"\n",
" },\n",
" metadata={\"foo\": \"bar\"},\n",
" )\n",
" return assistant\n",
"\n",
"assistant: Assistant = create_assistant()\n",
"assistant = create_assistant()\n",
"assistant_id: str = assistant.assistant_id\n",
"print(f\"created assistant: {assistant}\\n\")"
]
},
Expand All @@ -77,11 +77,7 @@
"outputs": [],
"source": [
"# get assistant\n",
"assistant_id: str = assistant.assistant_id\n",
"assistant: Assistant = taskingai.assistant.get_assistant(\n",
" assistant_id=assistant_id\n",
")\n",
"\n",
"assistant = taskingai.assistant.get_assistant(assistant_id)\n",
"print(f\"got assistant: {assistant}\\n\")"
]
},
Expand All @@ -94,13 +90,16 @@
"outputs": [],
"source": [
"# update assistant\n",
"assistant: Assistant = taskingai.assistant.update_assistant(\n",
"assistant = taskingai.assistant.update_assistant(\n",
" assistant_id=assistant_id,\n",
" name=\"New Assistant\",\n",
" retrieval_configs=RetrievalConfig(top_k=4, max_tokens=8192, method=RetrievalMethod.USER_MESSAGE),\n",
" name=\"New Assistant Name\",\n",
" retrieval_configs={\n",
" \"top_k\": 5, \n",
" \"max_tokens\": 2048,\n",
" \"method\": \"user_message\"\n",
" },\n",
")\n",
"\n",
"print(f\"updated assistant: {assistant}\\n\")\n"
"print(f\"updated assistant: {assistant}\\n\")"
]
},
{
Expand All @@ -112,7 +111,7 @@
"outputs": [],
"source": [
"# delete assistant\n",
"taskingai.assistant.delete_assistant(assistant_id=assistant_id)\n",
"taskingai.assistant.delete_assistant(assistant_id)\n",
"print(f\"deleted assistant: {assistant_id}\\n\")"
]
},
Expand Down Expand Up @@ -149,10 +148,10 @@
"outputs": [],
"source": [
"# create a new assistant\n",
"assistant: Assistant = create_assistant()\n",
"assistant = create_assistant()\n",
"\n",
"# create a chat\n",
"chat: Chat = taskingai.assistant.create_chat(\n",
"chat = taskingai.assistant.create_chat(\n",
" assistant_id=assistant.assistant_id,\n",
")\n",
"print(f\"created chat: {chat.chat_id} for assistant: {assistant.assistant_id}\\n\")"
Expand All @@ -168,7 +167,7 @@
"source": [
"# get chat\n",
"chat_id: str = chat.chat_id\n",
"chat: Chat = taskingai.assistant.get_chat(\n",
"chat = taskingai.assistant.get_chat(\n",
" assistant_id=assistant.assistant_id,\n",
" chat_id=chat_id,\n",
")\n",
Expand All @@ -184,7 +183,7 @@
"outputs": [],
"source": [
"# update chat\n",
"chat: Chat = taskingai.assistant.update_chat(\n",
"chat = taskingai.assistant.update_chat(\n",
" assistant_id=assistant.assistant_id,\n",
" chat_id=chat_id,\n",
" name=\"New Chat\",\n",
Expand Down Expand Up @@ -217,15 +216,12 @@
},
"outputs": [],
"source": [
"# list chats \n",
"# create chats \n",
"for _ in range(3):\n",
" taskingai.assistant.create_chat(\n",
" assistant_id=assistant.assistant_id,\n",
" )\n",
" taskingai.assistant.create_chat(assistant.assistant_id)\n",
"\n",
"chats = taskingai.assistant.list_chats(\n",
" assistant_id=assistant.assistant_id,\n",
")\n",
"# list chats\n",
"chats = taskingai.assistant.list_chats(assistant.assistant_id)\n",
"print(f\"num chats = {len(chats)}\\n\")"
]
},
Expand All @@ -238,7 +234,7 @@
"outputs": [],
"source": [
"# delete assistant\n",
"taskingai.assistant.delete_assistant(assistant_id=assistant.assistant_id)"
"taskingai.assistant.delete_assistant(assistant.assistant_id)"
]
}
],
Expand Down
Loading
Loading