AutoML time series forecasting
Project description
Auto Forecast
AutoML library for time series forecasting
Quick start
Upgrade pip
$ pip install pip --upgrade
Install autoforecast
$ pip install autoforecast
Try it out with your own dataset
- Preprocessing
from Autoforecast.preprocessing import preprocessing
X_train, y_train, X_test, y_test = preprocessing(
df=df,
target_name=['sales'],
categoricals=['store_id', 'dpt_id', 'holiday', 'zipcode', ...],
numericals=['employee_num', 'store_surface', ...],
date_col=['date'],
train_size=0.8,
engineering=True,
selection=True
)
- Fitting and predicting
from autoforecast.automl import AutoForecast
model = AutoForecast()
print('Autoforecast() model fitting...')
model.fit(X_train=X_train, y_train=y_train)
print('Autoforecast() model predicting...')
y_pred = model.predict(X_test=X_test)
print(f'y_pred={y_pred})
Run the example function
from autoforecast.examples import autoforecast_bitcoin
autoforecast_bitcoin.run()
Fetch historical cryptocurrency data
This function is a wrapper of https://developers.coinbase.com/api/v2#prices
- n: integer, number of days we want since today
- type: str, ['buy', 'sell', 'spot']
- currency_pair: str, crypto & currency
from autoforecast.datasets.import_bitcoin_price import get_price_for_last_n_days
crypto_df = get_price_for_last_n_days(
n=1, type='spot', currency_pair='BTC-USD')
)
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
autoforecast-0.0.9.tar.gz
(15.3 kB
view details)
Built Distribution
File details
Details for the file autoforecast-0.0.9.tar.gz
.
File metadata
- Download URL: autoforecast-0.0.9.tar.gz
- Upload date:
- Size: 15.3 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/3.4.1 importlib_metadata/4.6.0 pkginfo/1.7.0 requests/2.25.1 requests-toolbelt/0.9.1 tqdm/4.61.1 CPython/3.9.5
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | fa51de563c61db1490891cb84364c5ddf5f44309478e01388d93625888089d09 |
|
MD5 | 5963982f2e723e30bb8c28f2707a265a |
|
BLAKE2b-256 | 56d9a1b1a272f74b27dc22438a3cc5503dd428924a0dc00298a802589e896989 |
File details
Details for the file autoforecast-0.0.9-py3-none-any.whl
.
File metadata
- Download URL: autoforecast-0.0.9-py3-none-any.whl
- Upload date:
- Size: 25.6 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/3.4.1 importlib_metadata/4.6.0 pkginfo/1.7.0 requests/2.25.1 requests-toolbelt/0.9.1 tqdm/4.61.1 CPython/3.9.5
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 5bb9c5a52ab4c22e57590c6267e171195fba225c7f6497dae6461996198594af |
|
MD5 | c319712f9c33ec67f01bf5d8b128dd01 |
|
BLAKE2b-256 | 781597cd75a536a22ad34cae27a1fe276012f3aa9652a9fbeac5705deaf4852a |