**Multi-Agent Accelerator for Data Science: Batch AutoML (MAADSBML)
*Revolutionizing Data Science with Artificial Intelligence*
**Overview**
*MAADSBML combines Artificial Intelligence, Docker, Machine Learning. It automates the machine learning process, and finds the BEST algorithm for your data. It also
produces a very detailed PDF report. MAADSBML is integrated with Docker Container.
This library allows users to harness the power of agent-based computing using hundreds of advanced linear and non-linear algorithms.
FOR MORE INFORMATION GO TO: https://maadsbml.readthedocs.io/en/latest/
**Compatibility**
- Python 3.8 or greater
- Minimal Python skills needed
**Copyright**
- Author: Sebastian Maurice, PhD
**Installation**
- At the command prompt write:
**pip install maadsbml**
- This assumes you have [Downloaded Python](https://www.python.org/downloads/) and installed it on your computer.
- You will also need to pull the MAADSBML Docker container: **maadsdocker/maads-batch-automl-otics**
**Syntax**
- There are literally two lines of code you need to write to train your data and make predictions:
**Main functions:**
- **hypertraining**
Executes hundreds of agents, running hundreds of advanced algorithms and completes in minutes. A master agent then chooses the BEST algorithm that best models your data.
- **hyperpredictions**
After training, make high quality predictions - takes less than half a second (about ~100 milliseconds). Users can also generate predictions using **non-python code**
such as JAVA.
- **algodescription**
Get detailed information on the optimal algorithm found during hypertraining
- **abort**
Abort the training process.
- **rundemo**
To run canned demo of the system to see how it works.
- **finddistribution**
Finds the best distribution for your continuous data
**First import the Python library.**
**import maadsbml**
1. **maadsbml.hypertraining(host,port,filename,dependentvariable,removeoutliers=0,hasseasonality=0,summer='6,7,8',winter='11,12,1,2',shoulder='3,4,5,9,10',
trainingpercentage=70,shuffle=0,deepanalysis=0,username='admin',timeout=1200,company='otics',password='123',email='support@otics.ca',usereverseproxy=0,
microserviceid='',maadstoken='123',mode=0)**
**Parameters:**
*host* : string, required
- This is the IP address of the running Docker container - it is usually http://localhost
*port* : int, required
- This is the TRAINING PORT in the container. The default is port==5595
*filename* : string, required
- This is the raw data file in csv format - Note this file is stored on your host machine - the DOCKER container needs to be mapped to this volume using -v
*dependentvariable* : string, required
- This is the dependent variable in your csv file.
*removeoutliers* : int, optional, 1 or 0
- If 1, then outliers will be removed from your data. If 0, no outliers are removed.
*hasseasonality* : int, optional, 1 or 0
- If 1, then your data will be modeled for seasonality: Winter, Summer, Shoulder. If 0, then your data will
not be modeled for seasonality. If modeling for seasonality, ensure you have enough data points that
covers the seasons, usually 1 year of data.
*summer* : string, optional
- Definition for summer months. This can be changed.
*winter* : string, optional
- Definition for winter months. This can be changed.
*shoulder* : string, optional
- Definition for shoulder months. This can be changed.
*trainingpercentage* : int, optional, Default=70
- This is the split percentage between Training and Test data sets. It is defaulted to 70 (70% for training, 30% test).
*shuffle* : number, 0 or 1, optional
- Indicates whether to shuffle the training dataset or not, default=0.
*deepanalysis* : int, optional
- This will force MAADSBML to perform deeper analysis on your data. This could take 30-40 minutes. Set to 1 for deepanalysis, 0 for no deep analysis.
*username* : string, optional
- This identifies a user. You may want to change this if multiple users are running the same file.
*company* : string, optional
- This identifies your company. You may want to change this for the Report.
*timeout* : int, optional
- You can increase this if you receive a timeout error before the training is taking too long. The setting is in seconds.
*password* : string, optional
- leave as is
*email* : string, optional
- leave as is
*usereverseproxy* : int, optional
- leave as is
*microserviceid* : string, optional
- leave as is if not using a pass through service.
*mode* : int, optional
- leave as is
*maadstoken* : string, optional
- leave as is
**Returns:** string JSON buffer, with the algorithm key (PKEY) and other details:
- PKEY: : This is the key to the BEST algorithm and must be used when making predictions.
**2. maadsbml.hyperpredictions(pkey,theinputdata,host,port,username,algoname='',seasonname='',usereverseproxy=0,microserviceid='',password='123',company='otics',
email='support@otics.ca',maadstoken='123')**
**Parameters:**
*pkey* : string, required
- This is the PKEY you received from the hypertraining function.
*theinputdata* : string, required
- These are the Xs for your model: For example if my model had 3 Xs then inputdata='5/21/2010,-14.3,-32.0,-12.0', with the first entry as Date: Date
must be in the format: MM/DD/YYYY
*host* : string, required
- This is the IP address of the running Docker container - it is usually http://localhost
*port* : int, required
- This is the PREDICTION PORT in the container. The default is port==5495 (or 5595)
*username* : string, required
- The username you used in the hypertraining functions. Default is admin.
*algoname* : string, optional
- Enter the name of the algorithm to use, this can be retrieved from the hypertraining function. If this is empty, the BEST algorithm will be used by default.
*seasonname* : string, optional
- Enter the season to use (winter,summer,shoulder), this can be retrieved from the hypertraining function. If this is empty, the default season is used.
*usereverseproxy* : int, optional
- leave as is
*microserviceid* : int, optional
- leave as is
*password* : string, optional
- leave as is
*company* : string, optional
- change for reporting.
*email* : string, optional
- leave as is
*maadstoken* : string, optional
- leave as is
**Returns:** string buffer containing the prediction, and other details.
**3. maadsbml.abort(host,port=10000)**
**Parameters:**
*host* : string, required
- This is the IP address of the Docker container: http://localhost
*port* : string, optional
- Port is fixed at 10000
**Returns:** Abort will shutdown and re-start your system.
**4. maadsbml.rundemo(host,port,demotype=1,timeout=1200,usereverseproxy=0,microserviceid='')**
**Parameters:**
*host* : string, required
- This is the IP address of the Docker container: http://localhost
*port* : string, required
- This is the TRAININGPORT, it is usually 5595.
*demotype* : int, required
- If demotype is 1, then a regression models is run; if demotype is 0 then a classification model is run.
*timeout* : int, optional
- The connection timeout between Python and the container, in seconds
*usereverseproxy* : int, optional
- leave as is
*microserviceid* : string, optional
- leave as is
**Returns:** null
**5. maadsbml.algodescription(host,port,pkey,timeout=300,usereverseproxy=0,microserviceid='')**
**Parameters:**
*host* : string, required
- This is the IP address of the Docker container: http://localhost
*port* : string, required
- This is the TRAININGPORT, it is usually 5595.
*pkey* : string, required
- This is the PKEY from hypertraining.
*timeout* : int, optional
- The connection timeout between Python and the container, in seconds
*usereverseproxy* : int, optional
- leave as is
*microserviceid* : string, optional
- leave as is
**Returns:** null
**6. maadsbml.finddistribution(filename,varname,dataarray=[],folderpath='',imgname='distimage',common=1,topdist=5)**
**Parameters:**
*filename* : string, required
- Filename containing the raw data. This must be a CSV file.
*varname* : string, required
- Name of the variable for your data.
*dataarray* : array_like, optional
- Numpy array. Rather than pass a filename, you can pass in an array.
*folderpath* : string, optional
- Folder path to store the output data, and distribution image file.
*imgname* : string, optional
- Name of the image and json data.
*common* : int, optional
- If set to 1, this will apply common distributions to your data.
If Set to 0, it will iterate through roughly 80 distributions.
*topdist* : int, optional
- The number of the TOP distributions to print.
**Returns:** status,dist dataframe,name of best distribution,all JSON data
Raw data
{
"_id": null,
"home_page": "https://github.com/smaurice101/acnsmauricedsmas",
"name": "maadsbml",
"maintainer": null,
"docs_url": null,
"requires_python": null,
"maintainer_email": null,
"keywords": "multi-agent, data science, batch automl, optimization, prescriptive analytics, machine learning, automl, auto-ml, artificial intelligence, predictive analytics, advanced analytics",
"author": "Sebastian Maurice",
"author_email": "sebastian.maurice@otics.ca",
"download_url": "https://files.pythonhosted.org/packages/bd/43/550970a78bc83d274de5aca3729f5d87bfee665d33ca4e6f61b0f25dfb4e/maadsbml-1.3.19.tar.gz",
"platform": null,
"description": "**Multi-Agent Accelerator for Data Science: Batch AutoML (MAADSBML)\r\n\r\n*Revolutionizing Data Science with Artificial Intelligence*\r\n\r\n**Overview**\r\n\r\n*MAADSBML combines Artificial Intelligence, Docker, Machine Learning. It automates the machine learning process, and finds the BEST algorithm for your data. It also\r\nproduces a very detailed PDF report. MAADSBML is integrated with Docker Container.\r\n \r\nThis library allows users to harness the power of agent-based computing using hundreds of advanced linear and non-linear algorithms. \r\n\r\nFOR MORE INFORMATION GO TO: https://maadsbml.readthedocs.io/en/latest/\r\n\r\n**Compatibility**\r\n - Python 3.8 or greater\r\n - Minimal Python skills needed\r\n\r\n**Copyright**\r\n - Author: Sebastian Maurice, PhD\r\n\r\n**Installation**\r\n - At the command prompt write:\r\n **pip install maadsbml**\r\n - This assumes you have [Downloaded Python](https://www.python.org/downloads/) and installed it on your computer. \r\n\t - You will also need to pull the MAADSBML Docker container: **maadsdocker/maads-batch-automl-otics**\r\n\r\n**Syntax**\r\n - There are literally two lines of code you need to write to train your data and make predictions:\r\n\r\n**Main functions:**\r\n - **hypertraining**\r\n Executes hundreds of agents, running hundreds of advanced algorithms and completes in minutes. A master agent then chooses the BEST algorithm that best models your data.\r\n - **hyperpredictions**\r\n After training, make high quality predictions - takes less than half a second (about ~100 milliseconds). Users can also generate predictions using **non-python code** \r\n\t such as JAVA. \t \r\n - **algodescription**\r\n Get detailed information on the optimal algorithm found during hypertraining\r\n - **abort**\r\n Abort the training process.\r\n - **rundemo**\r\n To run canned demo of the system to see how it works.\r\n - **finddistribution**\r\n Finds the best distribution for your continuous data\r\n\t \r\n**First import the Python library.**\r\n\r\n**import maadsbml**\r\n\r\n1. **maadsbml.hypertraining(host,port,filename,dependentvariable,removeoutliers=0,hasseasonality=0,summer='6,7,8',winter='11,12,1,2',shoulder='3,4,5,9,10',\r\n trainingpercentage=70,shuffle=0,deepanalysis=0,username='admin',timeout=1200,company='otics',password='123',email='support@otics.ca',usereverseproxy=0,\r\n microserviceid='',maadstoken='123',mode=0)**\r\n\r\n**Parameters:**\t\r\n\r\n*host* : string, required\r\n\r\n- This is the IP address of the running Docker container - it is usually http://localhost\r\n\r\n*port* : int, required\r\n\r\n- This is the TRAINING PORT in the container. The default is port==5595\r\n\r\n*filename* : string, required\r\n \r\n - This is the raw data file in csv format - Note this file is stored on your host machine - the DOCKER container needs to be mapped to this volume using -v\r\n\r\n*dependentvariable* : string, required\r\n\r\n- This is the dependent variable in your csv file.\r\n\r\n*removeoutliers* : int, optional, 1 or 0\r\n\r\n- If 1, then outliers will be removed from your data. If 0, no outliers are removed.\r\n\r\n*hasseasonality* : int, optional, 1 or 0\r\n \r\n- If 1, then your data will be modeled for seasonality: Winter, Summer, Shoulder. If 0, then your data will \r\n not be modeled for seasonality. If modeling for seasonality, ensure you have enough data points that \r\n covers the seasons, usually 1 year of data.\r\n \r\n*summer* : string, optional\r\n \r\n- Definition for summer months. This can be changed.\r\n\r\n*winter* : string, optional\r\n \r\n- Definition for winter months. This can be changed.\r\n\r\n*shoulder* : string, optional\r\n \r\n- Definition for shoulder months. This can be changed.\r\n\r\n*trainingpercentage* : int, optional, Default=70\r\n \r\n- This is the split percentage between Training and Test data sets. It is defaulted to 70 (70% for training, 30% test).\r\n\r\n*shuffle* : number, 0 or 1, optional\r\n\r\n- Indicates whether to shuffle the training dataset or not, default=0.\r\n\r\n*deepanalysis* : int, optional\r\n\r\n- This will force MAADSBML to perform deeper analysis on your data. This could take 30-40 minutes. Set to 1 for deepanalysis, 0 for no deep analysis.\r\n\r\n*username* : string, optional\r\n \r\n - This identifies a user. You may want to change this if multiple users are running the same file.\r\n\r\n*company* : string, optional\r\n \r\n - This identifies your company. You may want to change this for the Report.\r\n\r\n*timeout* : int, optional\r\n\r\n- You can increase this if you receive a timeout error before the training is taking too long. The setting is in seconds.\r\n\r\n*password* : string, optional \r\n\r\n - leave as is\r\n\r\n*email* : string, optional \r\n\r\n - leave as is\r\n \r\n*usereverseproxy* : int, optional\r\n\r\n- leave as is\r\n \r\n*microserviceid* : string, optional\r\n\r\n- leave as is if not using a pass through service.\r\n\r\n*mode* : int, optional\r\n\r\n- leave as is\r\n\r\n*maadstoken* : string, optional\r\n\r\n- leave as is\r\n\r\n**Returns:** string JSON buffer, with the algorithm key (PKEY) and other details:\r\n \r\n- PKEY: : This is the key to the BEST algorithm and must be used when making predictions.\r\n\t\t\t\r\n**2. maadsbml.hyperpredictions(pkey,theinputdata,host,port,username,algoname='',seasonname='',usereverseproxy=0,microserviceid='',password='123',company='otics',\r\n email='support@otics.ca',maadstoken='123')**\r\n\r\n**Parameters:**\t\r\n\r\n*pkey* : string, required\r\n\r\n- This is the PKEY you received from the hypertraining function.\r\n\r\n*theinputdata* : string, required\r\n\r\n- These are the Xs for your model: For example if my model had 3 Xs then inputdata='5/21/2010,-14.3,-32.0,-12.0', with the first entry as Date: Date \r\nmust be in the format: MM/DD/YYYY\r\n\r\n*host* : string, required\r\n\r\n- This is the IP address of the running Docker container - it is usually http://localhost\r\n\r\n*port* : int, required\r\n\r\n- This is the PREDICTION PORT in the container. The default is port==5495 (or 5595)\r\n\r\n*username* : string, required\r\n\r\n- The username you used in the hypertraining functions. Default is admin.\r\n\r\n*algoname* : string, optional\r\n\r\n- Enter the name of the algorithm to use, this can be retrieved from the hypertraining function. If this is empty, the BEST algorithm will be used by default.\r\n\r\n*seasonname* : string, optional\r\n\r\n- Enter the season to use (winter,summer,shoulder), this can be retrieved from the hypertraining function. If this is empty, the default season is used.\r\n\r\n*usereverseproxy* : int, optional\r\n\r\n- leave as is\r\n\r\n*microserviceid* : int, optional\r\n\r\n- leave as is\r\n\r\n*password* : string, optional\r\n\r\n- leave as is\r\n\r\n*company* : string, optional\r\n\r\n- change for reporting.\r\n\r\n*email* : string, optional\r\n\r\n- leave as is\r\n\r\n*maadstoken* : string, optional\r\n\r\n- leave as is\r\n\t \r\n**Returns:** string buffer containing the prediction, and other details.\r\n \r\n\r\n**3. maadsbml.abort(host,port=10000)**\r\n\r\n**Parameters:**\t\r\n\r\n*host* : string, required\r\n\r\n- This is the IP address of the Docker container: http://localhost\r\n \r\n*port* : string, optional\r\n \r\n- Port is fixed at 10000\r\n\r\n**Returns:** Abort will shutdown and re-start your system.\r\n \r\n \r\n**4. maadsbml.rundemo(host,port,demotype=1,timeout=1200,usereverseproxy=0,microserviceid='')**\r\n\r\n**Parameters:**\t\r\n\r\n*host* : string, required\r\n\r\n- This is the IP address of the Docker container: http://localhost\r\n \r\n*port* : string, required\r\n \r\n- This is the TRAININGPORT, it is usually 5595.\r\n\r\n*demotype* : int, required\r\n\r\n- If demotype is 1, then a regression models is run; if demotype is 0 then a classification model is run.\r\n\r\n*timeout* : int, optional\r\n\r\n- The connection timeout between Python and the container, in seconds\r\n\r\n*usereverseproxy* : int, optional\r\n\r\n- leave as is \r\n\r\n*microserviceid* : string, optional\r\n\r\n- leave as is\r\n \r\n**Returns:** null\r\n \r\n \r\n**5. maadsbml.algodescription(host,port,pkey,timeout=300,usereverseproxy=0,microserviceid='')**\r\n\r\n**Parameters:**\t\r\n\r\n*host* : string, required\r\n\r\n- This is the IP address of the Docker container: http://localhost\r\n \r\n*port* : string, required\r\n \r\n- This is the TRAININGPORT, it is usually 5595.\r\n\r\n*pkey* : string, required\r\n\r\n- This is the PKEY from hypertraining.\r\n\r\n*timeout* : int, optional\r\n\r\n- The connection timeout between Python and the container, in seconds\r\n\r\n*usereverseproxy* : int, optional\r\n\r\n- leave as is \r\n\r\n*microserviceid* : string, optional\r\n\r\n- leave as is\r\n \r\n**Returns:** null\r\n \r\n \r\n**6. maadsbml.finddistribution(filename,varname,dataarray=[],folderpath='',imgname='distimage',common=1,topdist=5)**\r\n\r\n**Parameters:**\t\r\n\r\n*filename* : string, required\r\n\r\n- Filename containing the raw data. This must be a CSV file.\r\n \r\n*varname* : string, required\r\n \r\n- Name of the variable for your data.\r\n\r\n*dataarray* : array_like, optional\r\n\r\n- Numpy array. Rather than pass a filename, you can pass in an array.\r\n\r\n*folderpath* : string, optional\r\n\r\n- Folder path to store the output data, and distribution image file.\r\n\r\n*imgname* : string, optional\r\n\r\n- Name of the image and json data.\r\n\r\n*common* : int, optional\r\n\r\n- If set to 1, this will apply common distributions to your data. \r\n \r\n If Set to 0, it will iterate through roughly 80 distributions.\r\n\r\n*topdist* : int, optional\r\n\r\n- The number of the TOP distributions to print.\r\n \r\n**Returns:** status,dist dataframe,name of best distribution,all JSON data\r\n \r\n",
"bugtrack_url": null,
"license": "MIT License",
"summary": "Multi-Agent Accelerator for Data Science (MAADS) Batch AutoML (MAADSBML)",
"version": "1.3.19",
"project_urls": {
"Homepage": "https://github.com/smaurice101/acnsmauricedsmas"
},
"split_keywords": [
"multi-agent",
" data science",
" batch automl",
" optimization",
" prescriptive analytics",
" machine learning",
" automl",
" auto-ml",
" artificial intelligence",
" predictive analytics",
" advanced analytics"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "1f6666c6ef221725c822c74aa0aa439b69e6fd59e2bb8f9edc353b284a02b79d",
"md5": "e40e58491ac34d8c7e1fdc414af40c09",
"sha256": "64724b5689d0667bddb90c55f5b2e93e2471488e67056a42372d0d437a51f58a"
},
"downloads": -1,
"filename": "maadsbml-1.3.19-py3-none-any.whl",
"has_sig": false,
"md5_digest": "e40e58491ac34d8c7e1fdc414af40c09",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": null,
"size": 13850,
"upload_time": "2024-10-08T01:42:37",
"upload_time_iso_8601": "2024-10-08T01:42:37.642912Z",
"url": "https://files.pythonhosted.org/packages/1f/66/66c6ef221725c822c74aa0aa439b69e6fd59e2bb8f9edc353b284a02b79d/maadsbml-1.3.19-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "bd43550970a78bc83d274de5aca3729f5d87bfee665d33ca4e6f61b0f25dfb4e",
"md5": "056c44bfb69556e654a4b9db59927e69",
"sha256": "587478a9b5dff9d0f46b8d74ea82ac2554e7e52c522e03c549d20a56d2c4f1b4"
},
"downloads": -1,
"filename": "maadsbml-1.3.19.tar.gz",
"has_sig": false,
"md5_digest": "056c44bfb69556e654a4b9db59927e69",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 15992,
"upload_time": "2024-10-08T01:42:39",
"upload_time_iso_8601": "2024-10-08T01:42:39.215545Z",
"url": "https://files.pythonhosted.org/packages/bd/43/550970a78bc83d274de5aca3729f5d87bfee665d33ca4e6f61b0f25dfb4e/maadsbml-1.3.19.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-10-08 01:42:39",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "smaurice101",
"github_project": "acnsmauricedsmas",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"lcname": "maadsbml"
}
JSON
