# Arcee
## *The OptScale ML profiling tool by Hystax*
Arcee is a tool that hepls you to integrate ML tasks with [OptScale](https://my.optscale.com/).
This tool can automatically collect executor metadata from cloud and process stats.
## Installation
Arcee requires python 3.7+ to run.
```sh
pip install optscale-arcee
```
## Usage
First of all you need to import and init arcee in your code:
```sh
import optscale_arcee as arcee
```
```sh
# init arcee using context manager syntax
with arcee.init('token', 'model_key'):
# some code
```
To use custom endpoint and enable\disable ssl checks (supports using self-signed ssl certificates):
```sh
with arcee.init('token', 'model_key', endpoint_url='https://my.custom.endpoint:443/arcee/v2', ssl=False):
# some code
```
Alternatively arcee can be initialized via function call. However manual finish is required:
```sh
arcee.init('token', 'model_key')
# some code
arcee.finish()
```
Or in error case:
```sh
arcee.init('token', 'model_key')
# some code
arcee.error()
```
To send stats:
```sh
arcee.send({"loss": 2.0012, "iter": 2, "epoch": 1})
```
(key should be string, value - int or float, multiple values can be sent)
To add tags to model run (key, value):
```sh
arcee.tag("project", "torchvision demo")
```
To add milestones:
```sh
arcee.milestone("Download test data")
```
To add stages:
```sh
arcee.stage("calculation")
```
To add hyperparameters:
```sh
arcee.hyperparam("epochs", 5)
```
## Logging datasets
To log a dataset, use the dataset method with the following parameter:
- path (str): the path of the dataset.
```
arcee.dataset("dataset_path")
```
Raw data
{
"_id": null,
"home_page": "https://my.optscale.com/",
"name": "optscale-arcee",
"maintainer": "",
"docs_url": null,
"requires_python": "<4,>=3.7",
"maintainer_email": "",
"keywords": "arcee,ml,optscale,finops,mlops",
"author": "Hystax",
"author_email": "",
"download_url": "https://files.pythonhosted.org/packages/7b/e5/7810fbab09c5e8070d72723530c8fec0263bd6956eaa7bf58455f3053b6c/optscale_arcee-0.1.37.tar.gz",
"platform": null,
"description": "# Arcee\n## *The OptScale ML profiling tool by Hystax*\n\nArcee is a tool that hepls you to integrate ML tasks with [OptScale](https://my.optscale.com/).\nThis tool can automatically collect executor metadata from cloud and process stats.\n\n## Installation\nArcee requires python 3.7+ to run.\n```sh\npip install optscale-arcee\n```\n\n## Usage\nFirst of all you need to import and init arcee in your code:\n```sh\nimport optscale_arcee as arcee\n```\n\n```sh\n# init arcee using context manager syntax\nwith arcee.init('token', 'model_key'):\n # some code\n```\n\nTo use custom endpoint and enable\\disable ssl checks (supports using self-signed ssl certificates):\n```sh\nwith arcee.init('token', 'model_key', endpoint_url='https://my.custom.endpoint:443/arcee/v2', ssl=False):\n # some code\n```\n\nAlternatively arcee can be initialized via function call. However manual finish is required:\n```sh\narcee.init('token', 'model_key')\n# some code\narcee.finish()\n```\n\nOr in error case:\n```sh\narcee.init('token', 'model_key')\n# some code\narcee.error()\n```\n\nTo send stats:\n```sh\narcee.send({\"loss\": 2.0012, \"iter\": 2, \"epoch\": 1})\n```\n(key should be string, value - int or float, multiple values can be sent)\n\nTo add tags to model run (key, value):\n```sh\narcee.tag(\"project\", \"torchvision demo\")\n```\n\nTo add milestones:\n```sh\narcee.milestone(\"Download test data\")\n```\n\nTo add stages:\n```sh\narcee.stage(\"calculation\")\n```\n\nTo add hyperparameters:\n```sh\narcee.hyperparam(\"epochs\", 5)\n```\n\n## Logging datasets\nTo log a dataset, use the dataset method with the following parameter:\n- path (str): the path of the dataset.\n```\narcee.dataset(\"dataset_path\")\n```\n",
"bugtrack_url": null,
"license": "\"Apache License 2.0\"",
"summary": "ML profiling tool for OptScale",
"version": "0.1.37",
"project_urls": {
"Homepage": "https://my.optscale.com/",
"Source": "https://github.com/hystax/optscale_arcee"
},
"split_keywords": [
"arcee",
"ml",
"optscale",
"finops",
"mlops"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "7be57810fbab09c5e8070d72723530c8fec0263bd6956eaa7bf58455f3053b6c",
"md5": "0cb8d832fdf7066ac3d060d84105fbab",
"sha256": "b4991a8dad042bf206c32c7389649a04ad1e735b7f07d9c7a2185ac281c6706d"
},
"downloads": -1,
"filename": "optscale_arcee-0.1.37.tar.gz",
"has_sig": false,
"md5_digest": "0cb8d832fdf7066ac3d060d84105fbab",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "<4,>=3.7",
"size": 19149,
"upload_time": "2023-12-26T17:30:44",
"upload_time_iso_8601": "2023-12-26T17:30:44.678093Z",
"url": "https://files.pythonhosted.org/packages/7b/e5/7810fbab09c5e8070d72723530c8fec0263bd6956eaa7bf58455f3053b6c/optscale_arcee-0.1.37.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-12-26 17:30:44",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "hystax",
"github_project": "optscale_arcee",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"tox": true,
"lcname": "optscale-arcee"
}
JSON
