A restful client library, designed to access predictnow restful api.
Project description
Usage:
from pnow.pdapi import PredictNowClient
Setup client along with its api key
api_key = "<YOUR_API_KEY>" client = PredictNowClient(api_key)
Train demo
train_input_path = 'C:/Users/devstack/Documents/example_input_train.csv' train_params = { "username": "welly", "email": "welly@predictnow.ai", "label": "futreturn", "timeseries": "yes", "type": "classification", "feature_selection": "shap", "analysis": "small", "boost": "gbdt", "mode": "train", "testsize": "0.2", "weights": "no", "prob_calib": "no", "suffix": "myfirstsuffix", "eda": "no", }
response = client.train(train_input_path, train_params) print(response)
Predict demo
live_input_path = 'C:/Users/devstack/Documents/example_input_live.csv' username = train_params["username"] suffix = train_params["suffix"] path = "../" + train_params["username"] predict_params = { "username": username, "model_name": "saved_model_" + suffix + ".pkl", # TODO proper model name "eda": "no", } response = client.predict(live_input_path, params=predict_params) print(response)
Save Result demo
response = client.save_to_output({"username": "welly", "output": "C:/Users/devstack/Documents"}) print(response)
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 pnow_client-0.1.2-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | e4fa5e65fa305e8f38e76d52f366b088d6649a048ba74e4e56a23e811ee42aac |
|
MD5 | 018e0308281cde97026c02d62977597e |
|
BLAKE2b-256 | f30795bd224907862a6368673fe529fc8b921e6e0d9d81307847a80188c9daf6 |