An integration package connecting IBM watsonx.ai and GigaChain
Project description
langchain-ibm
This package provides the integration between LangChain and IBM watsonx.ai through the ibm-watsonx-ai
SDK.
Installation
To use the langchain-ibm
package, follow these installation steps:
pip install gigachain-ibm
Usage
Setting up
To use IBM's models, you must have an IBM Cloud user API key. Here's how to obtain and set up your API key:
- Obtain an API Key: For more details on how to create and manage an API key, refer to IBM's documentation.
- Set the API Key as an Environment Variable: For security reasons, it's recommended to not hard-code your API key directly in your scripts. Instead, set it up as an environment variable. You can use the following code to prompt for the API key and set it as an environment variable:
import os
from getpass import getpass
watsonx_api_key = getpass()
os.environ["WATSONX_APIKEY"] = watsonx_api_key
In alternative, you can set the environment variable in your terminal.
-
Linux/macOS: Open your terminal and execute the following command:
export WATSONX_APIKEY='your_ibm_api_key'
To make this environment variable persistent across terminal sessions, add the above line to your
~/.bashrc
,~/.bash_profile
, or~/.zshrc
file. -
Windows: For Command Prompt, use:
set WATSONX_APIKEY=your_ibm_api_key
Loading the model
You might need to adjust model parameters for different models or tasks. For more details on the parameters, refer to IBM's documentation.
parameters = {
"decoding_method": "sample",
"max_new_tokens": 100,
"min_new_tokens": 1,
"temperature": 0.5,
"top_k": 50,
"top_p": 1,
}
Initialize the WatsonxLLM class with the previously set parameters.
from langchain_ibm import WatsonxLLM
watsonx_llm = WatsonxLLM(
model_id="PASTE THE CHOSEN MODEL_ID HERE",
url="PASTE YOUR URL HERE",
project_id="PASTE YOUR PROJECT_ID HERE",
params=parameters,
)
Note:
- You must provide a
project_id
orspace_id
. For more information refer to IBM's documentation. - Depending on the region of your provisioned service instance, use one of the urls described here.
- You need to specify the model you want to use for inferencing through
model_id
. You can find the list of available models here.
Alternatively you can use Cloud Pak for Data credentials. For more details, refer to IBM's documentation.
watsonx_llm = WatsonxLLM(
model_id="ibm/granite-13b-instruct-v2",
url="PASTE YOUR URL HERE",
username="PASTE YOUR USERNAME HERE",
password="PASTE YOUR PASSWORD HERE",
instance_id="openshift",
version="4.8",
project_id="PASTE YOUR PROJECT_ID HERE",
params=parameters,
)
Create a Chain
Create PromptTemplate
objects which will be responsible for creating a random question.
from langchain.prompts import PromptTemplate
template = "Generate a random question about {topic}: Question: "
prompt = PromptTemplate.from_template(template)
Provide a topic and run the LLMChain.
from langchain.chains import LLMChain
llm_chain = LLMChain(prompt=prompt, llm=watsonx_llm)
response = llm_chain.invoke("dog")
print(response)
Calling the Model Directly
To obtain completions, you can call the model directly using a string prompt.
# Calling a single prompt
response = watsonx_llm.invoke("Who is man's best friend?")
print(response)
# Calling multiple prompts
response = watsonx_llm.generate(
[
"The fastest dog in the world?",
"Describe your chosen dog breed",
]
)
print(response)
Streaming the Model output
You can stream the model output.
for chunk in watsonx_llm.stream(
"Describe your favorite breed of dog and why it is your favorite."
):
print(chunk, end="")
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for gigachain_ibm-0.1.8-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | e7c85f4d0988c1e3a54cc99af2326256e6d1f0f274c632e69245e884950af82f |
|
MD5 | aa7f719380a344770b913c7700664cc5 |
|
BLAKE2b-256 | 2ff3af875713e94a34e8162a4df67464a1b04f44fda0c1f6c7bcbcf0ceb50e4a |