In this post today I am going to explaining about the python packages for accessing cohere.ai natural language API in Python 3. The main core principle is functionality developed to simplify interfacing. The main motive for Python Packages For Accessing Cohere Ai Through APIs.
Documentation:
See the API’s documentation.
Also, see some code examples for the SDK here.
Installation:
If you want the package, you can install it through pip
:
pip install --upgrade cohere
Install from source:
python setup.py install
Requirements:
Python 3.6+
Usage:
import cohere
# initialize the Cohere Client with an API Key
co = cohere.CohereClient('YOUR_API_KEY')
# generate a prediction for a prompt
prediction = co.generate(
model="baseline-shrimp",
prompt="co:here",
max_tokens=10)
# print the predicted text
print('prediction: {}'.format(prediction.text))
More usage examples can be found here.
Endpoints:
For a full breakdown of endpoints and arguments, please consult the Cohere Docs.
Cohere Endpoint | Function |
---|---|
/generate | co.generate() |
/similarity | co.similarity() |
/choose-best | co.choose_best() |
/embed | co.embed() |
/likelihood | co.likelihood() |
Models
To view an up-to-date list of available models please consult the Cohere CLI. To get started try out baseline-shrimp
or baseline-seal
.
GitHub:
Please comment and share the post if you find it helpful and WhatsApp for More repositories.