# What is this Library about?
Easy-to-use (4 lines of code, actually) framework for training powerful predictive models!
# Description
We made a mother-model, which consists of multiple layers of predictive models: ewma is used as trend, Prophet is used for getting seasonality, CatBoost is used for predicting residuals.
Why did we do that? Because we needed a out-of-the-box solution, which could be used by non-ML users.
# How-to-install?
You can install this framework via pypi:
```
pip install TeremokTSLib
```
# How-to-use?
You can watch an example in TeremokTSLib/tests foulder. All you need is dataframe with 2 columns: date, consumption.
Then you can initiate mother-model and train it with just 2 rows of code:
```
import TeremokTSLib as tts
model = tts.Model()
model.train(data=data)
```
# Maintained by
Library is developed and being maintained by Teremok ML team
# Contacts
- Our website: https://teremok.ru/
- ML team: you can contact us via telegram channel @pivo_txt
Change Log
==========
0.1.0 (27.07.2024)
------------------
- First release
1.1.0 (28.07.2024)
------------------
- Beta verison release
- Visualisation of itertest added
1.1.1 (06.08.2024)
------------------
- Fixed some bugs
1.1.2 (09.08.2024)
------------------
- Added parallel training for Prophet
1.1.3 (16.08.2024)
------------------
- Now predict_order method returns dict with predicted orders and cons
- Added visualisation of optuna trials
1.1.4 (17.08.2024)
------------------
- Optimized Prophet inference. 54% reduction of inference time.
1.1.5 (21.08.2024)
------------------
- Fixed bug with Optuna beta optimisation.
1.1.6 (21.08.2024)
------------------
- Fixed bug with ewma shift.
1.1.7 - YANKED - (22.08.2024)
------------------
- Added regularisation for orders in time of surges in consumption.
1.1.8 (23.08.2024)
------------------
- Uploaded fixed seasonality;
- Added WAPE metric calculation in itertest.
1.1.9 (24.08.2024)
------------------
- finally fixed beta optimization;
- added regularization parameter to optuna;
- added safe stock coef to optuna;
1.2.0 (24.08.2024)
------------------
- added support for lower-than-predicted orders.
1.2.1 (26.08.2024)
------------------
- fixed itertest;
- added NeuralProphet option.
1.2.2 (28.08.2024)
------------------
- deleted NeuralProphet option;
- added MinMaxModel for modelling long-living items.
1.2.4 (11.09.2024)
------------------
- added minmax models;
- now output of predict_order method is dict of np arrays;
1.2.5 (29.09.2024)
------------------
- switched from floor-cap minmax models to just floor;
Raw data
{
"_id": null,
"home_page": "https://teremok.ru/",
"name": "TeremokTSLib",
"maintainer": null,
"docs_url": null,
"requires_python": null,
"maintainer_email": null,
"keywords": "forcasting",
"author": "Alexander Nikitin",
"author_email": "sniknickitin@yandex.ru",
"download_url": "https://files.pythonhosted.org/packages/8f/27/ead3a576380c8128b9baab3ab707b35ad5d3ff0586252085ec564b3a6a16/TeremokTSLib-1.2.5.tar.gz",
"platform": null,
"description": "\r\n\r\n# What is this Library about?\r\nEasy-to-use (4 lines of code, actually) framework for training powerful predictive models!\r\n\r\n# Description\r\nWe made a mother-model, which consists of multiple layers of predictive models: ewma is used as trend, Prophet is used for getting seasonality, CatBoost is used for predicting residuals.\r\nWhy did we do that? Because we needed a out-of-the-box solution, which could be used by non-ML users. \r\n\r\n# How-to-install?\r\nYou can install this framework via pypi: \r\n```\r\npip install TeremokTSLib\r\n```\r\n\r\n# How-to-use?\r\nYou can watch an example in TeremokTSLib/tests foulder. All you need is dataframe with 2 columns: date, consumption.\r\nThen you can initiate mother-model and train it with just 2 rows of code:\r\n```\r\nimport TeremokTSLib as tts\r\nmodel = tts.Model()\r\nmodel.train(data=data)\r\n```\r\n\r\n# Maintained by\r\nLibrary is developed and being maintained by Teremok ML team\r\n\r\n# Contacts\r\n- Our website: https://teremok.ru/\r\n- ML team: you can contact us via telegram channel @pivo_txt\r\n\r\nChange Log\r\n==========\r\n\r\n0.1.0 (27.07.2024)\r\n------------------\r\n- First release\r\n\r\n1.1.0 (28.07.2024)\r\n------------------\r\n- Beta verison release\r\n- Visualisation of itertest added\r\n\r\n1.1.1 (06.08.2024)\r\n------------------\r\n- Fixed some bugs\r\n\r\n1.1.2 (09.08.2024)\r\n------------------\r\n- Added parallel training for Prophet\r\n\r\n1.1.3 (16.08.2024)\r\n------------------\r\n- Now predict_order method returns dict with predicted orders and cons\r\n- Added visualisation of optuna trials\r\n\r\n1.1.4 (17.08.2024)\r\n------------------\r\n- Optimized Prophet inference. 54% reduction of inference time.\r\n\r\n1.1.5 (21.08.2024)\r\n------------------\r\n- Fixed bug with Optuna beta optimisation.\r\n\r\n1.1.6 (21.08.2024)\r\n------------------\r\n- Fixed bug with ewma shift.\r\n\r\n1.1.7 - YANKED - (22.08.2024)\r\n------------------\r\n- Added regularisation for orders in time of surges in consumption.\r\n\r\n1.1.8 (23.08.2024)\r\n------------------\r\n- Uploaded fixed seasonality;\r\n- Added WAPE metric calculation in itertest.\r\n\r\n1.1.9 (24.08.2024)\r\n------------------\r\n- finally fixed beta optimization;\r\n- added regularization parameter to optuna;\r\n- added safe stock coef to optuna;\r\n\r\n1.2.0 (24.08.2024)\r\n------------------\r\n- added support for lower-than-predicted orders.\r\n\r\n1.2.1 (26.08.2024)\r\n------------------\r\n- fixed itertest;\r\n- added NeuralProphet option.\r\n\r\n1.2.2 (28.08.2024)\r\n------------------\r\n- deleted NeuralProphet option;\r\n- added MinMaxModel for modelling long-living items.\r\n\r\n1.2.4 (11.09.2024)\r\n------------------\r\n- added minmax models;\r\n- now output of predict_order method is dict of np arrays;\r\n\r\n1.2.5 (29.09.2024)\r\n------------------\r\n- switched from floor-cap minmax models to just floor;\r\n\r\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "Easy-to-use box ML solution for forcasting consumption",
"version": "1.2.5",
"project_urls": {
"Homepage": "https://teremok.ru/"
},
"split_keywords": [
"forcasting"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "672ff16ef7073fd1a5b6d23400a3f92e544bd3a46fa58956dfe2368410d09758",
"md5": "68d40e1d2bff25d73573b3e21641242f",
"sha256": "d16229868c9997da33d19e633bc63df03489d9b74e6c3cbe149e7095ae86d13a"
},
"downloads": -1,
"filename": "TeremokTSLib-1.2.5-py3-none-any.whl",
"has_sig": false,
"md5_digest": "68d40e1d2bff25d73573b3e21641242f",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": null,
"size": 23514,
"upload_time": "2024-09-29T18:18:48",
"upload_time_iso_8601": "2024-09-29T18:18:48.209392Z",
"url": "https://files.pythonhosted.org/packages/67/2f/f16ef7073fd1a5b6d23400a3f92e544bd3a46fa58956dfe2368410d09758/TeremokTSLib-1.2.5-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "8f27ead3a576380c8128b9baab3ab707b35ad5d3ff0586252085ec564b3a6a16",
"md5": "e70aaa12d882ed576ec78b9f715fef9b",
"sha256": "ebade37c4d5231af3ef73e057957665edc40e66fc90ebba0d5861ebfeea76ea6"
},
"downloads": -1,
"filename": "TeremokTSLib-1.2.5.tar.gz",
"has_sig": false,
"md5_digest": "e70aaa12d882ed576ec78b9f715fef9b",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 23980,
"upload_time": "2024-09-29T18:18:49",
"upload_time_iso_8601": "2024-09-29T18:18:49.671147Z",
"url": "https://files.pythonhosted.org/packages/8f/27/ead3a576380c8128b9baab3ab707b35ad5d3ff0586252085ec564b3a6a16/TeremokTSLib-1.2.5.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-09-29 18:18:49",
"github": false,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"lcname": "teremoktslib"
}
JSON