Aimped is a unique library that provides classes and functions for only exclusively business-tailored AI-based NLP models.
Project description
aimped
Welcome to Aimped Inc.: an LLMs-in-the-loop AI Platform.
Aimped is a distinctive AI platform offering model inference and development, harnessing the power of LLMs. Our innovative approach ensures quick, cost-effective, and efficient model training and evaluation. Additionally, we provide user-friendly UI-based AI applications and LLM Agents, making advanced AI accessible and easy to use.
Moreover, we will soon allow model uploads and enable users to earn money. This will encourage the open-source community to share models, reducing redundant training and computing costs while fostering collaboration and innovation.
aimped
is a unique python library that provides classes and functions for only exclusively business-tailored AI-based models.
In this version, we provide the following features:
API service, Sound processing tools and functions, NLP tools and functions, and a pipeline class for NLP tasks.
Installation
pip install aimped
API Usage
Configuration of the Library
from aimped.services.api import AimpedAPI
# Create new instance Aimped
user_key = '' # user_key received from A3M.
user_secret = '' # user_secret received from A3M.
BASE_URL = 'https://aimped.ai' # Aimped domain url
api_service = AimpedAPI(user_key, user_secret, {
base_url: BASE_URL
})
Preparation of the model input data
model_id = "" # ID of the model run. The model ID is available on the model description page under API usage.
payload = {...} # Model input examples (payload) are available in the api usage tab on the Model description page.
Usage of API Function
result = api_service.run_mode(model_id, input_data)
Usage of API Callback Function
# return callback function
def callback(event, message, time, data=None):
if event == 'start':
print(f'Start event at {time}: {message}')
elif event == 'proccess':
print(f'Progress event at {time}: {message}')
elif event == 'error':
print(f'Error event at {time}: {message}')
elif event == 'end':
print(f'End event at {time}: {message}. Data: {data}')
result = api_service.run_model_callback(model_id, payload, callback)
Usage of API File Upload
Some of the models supports file inputs. These inputs are accepted as URIs. Here is the usage of API for file uploads.
input = api_service.file_upload(
model_id,
'/Users/joe/Downloads/xyz.pdf' # sample file path to upload
)
Usage of API File Download
Some of the models supports file outputs as result. These outputs are created as URIs. Here is the usage of API for file downloads.
output_file = api_service.file_download_and_save(
'input/application/model_{{modelId}}/user_{{userId}}/file_name', # URI of the model output file in the result
'/Users/joe/Downloads/123_file_name' # sample local file path to save
)
Contributing
Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.
Please make sure to update tests as appropriate.
License
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.