| Name | ado-core JSON |
| Version |
1.2.0
JSON |
| download |
| home_page | None |
| Summary | ado is a unified platform for executing computational experiments at scale and analysing their results. It can be easily extended with new experiments or new analysis tools. It allows distributed teams of researchers and engineers to collaborate on projects, execute experiments, and share data. |
| upload_time | 2025-11-06 15:21:09 |
| maintainer | None |
| docs_url | None |
| author | None |
| requires_python | <3.13,>=3.10 |
| license | None |
| keywords |
discovery
exploration
optimization
orchestration
ray
|
| VCS |
 |
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
|
# Introduction
This is the repository for the **a**ccelerated **d**iscovery **o**rchestrator
(**`ado`**).
**`ado`** is a unified platform for **executing computational experiments at
scale** and **analysing their results**. It can be extended with new experiments
or new analysis tools. It allows distributed teams of researchers and engineers
to collaborate on projects, execute experiments, and share data.
You can run the experiments and analysis tools already available in **`ado`** in
a distributed, shared, environment with your team. You can also use **`ado`** to
get features like data-tracking, data-sharing, tool integration and a CLI, for
your analysis method or experiment for free.
🧑💻 Using **`ado`** assumes familiarity with command line tools.
🛠️ Developing **`ado`** requires knowledge of python.
## Key Features
- 💻 _CLI_: Our human-centric CLI follows [best practices](https://clig.dev)
- 🤝 _Projects_: Allow distributed groups of users to
[collaborate and share data](https://ibm.github.io/ado/resources/metastore)
- 🔌 _Extendable_: Easily
[add new experiments](https://ibm.github.io/ado/actuators/creating-custom-experiments),
[optimizers or other tools.](https://ibm.github.io/ado/operators/creating-operators)
- ⚙️ _Scalable_: We use [ray](https://ray.io) as our execution engine allowing
experiments and tools to easily scale
- ♻️ _Automatic data-reuse_: Avoid repeating work with
[transparent reuse of experiment results](https://ibm.github.io/ado/core-concepts/data-sharing).
`ado` internal protocols ensure this happens only when it makes sense
- 🔗 _Provenance_: As you work, the relationship between the data you create and
operations you perform are
[automatically tracked](https://ibm.github.io/ado/getting-started/ado#ado-show-related)
- 🔎 _Optimization and sampling_: Out-of-the-box, leverage powerful optimization
methods
[via `raytune`](https://ibm.github.io/ado/operators/optimisation-with-ray-tune)
or use our
[flexible in built sampler](https://ibm.github.io/ado/operators/random-walk)
### Foundation Model Experimentation
We have developed `ado` plugins providing advanced experiments for testing
foundation-models:
- ⏱️
[fine-tuning performance benchmarking](https://ibm.github.io/ado/actuators/sft-trainer)
- ⏱️ inference performance benchmarking (using the
[vLLM performance benchmark](https://docs.vllm.ai/en/stable/api/vllm/benchmarks/serve.html))
- **COMING SOON** 🔮 inference and fine-tuning prediction
## Requirements
A basic installation of `ado` only requires a recent Python version (3.10+).
This will allow you to run
[many of our examples](https://ibm.github.io/ado/examples/examples) and explore
ado features.
### Additional Requirements
Some advanced features have additional requirements:
<!-- markdownlint-disable descriptive-link-text -->
- **Distributed Projects** **_(Optional)_**: To support projects with multiple
users you will need a remote, accessible, MySQL database. See
[here](https://ibm.github.io/ado/getting-started/installing-backend-services#using-the-distributed-mysql-backend-for-ado)
for more
- **Multi-Node Execution** **_(Optional)_**: To support multi-node or scaling
execution you may need a multi-node RayCluster. See
[here](https://ibm.github.io/ado/getting-started/installing-backend-services#deploying-kuberay-and-creating-a-raycluster)
for more details
<!-- markdownlint-enable descriptive-link-text -->
In addition `ado` plugins may have additional requirements for executing
**_realistic_** experiments. For example,
- **_Fine-Tuning Benchmarking_**: Requires a
[RayCluster with GPUs](https://ibm.github.io/ado/actuators/sft-trainer#configure-your-raycluster)
- **_vLLM Performance Benchmarking_**: Requires an OpenShift cluster with GPUs
## Install
To install you can execute the following (we recommend you set up a virtual
environment)
```commandline
git clone https://github.com/IBM/ado.git
cd ado
pip install .
```
Alternate instructions to install `ado` can be found here:
<https://ibm.github.io/ado/getting-started/install/>
## Development
Instructions for developing ado are available in [DEVELOPING](DEVELOPING.md).
### Testing
To run unit-tests read [tests/README.md](tests/README.md).
## Example
This video shows listing
[actuators](website/docs/actuators/working-with-actuators.md) and getting the
details of an experiment.
Check [demo](https://ibm.github.io/ado/getting-started/demo) for more videos.
[](https://github.com/user-attachments/assets/fc4862f3-763b-4967-ab3c-4bd359900a50)
## Technical Report
For more details on the Discovery Spaces concept underlying ado, please refer to
this [technical report](https://arxiv.org/abs/2506.21467).
## Acknowledgement
This project is partially funded by the European Union through the Smart
Networks and Services Joint Undertaking (SNS JU) under grant agreement No.
101192750 (Project 6G-DALI).
Raw data
{
"_id": null,
"home_page": null,
"name": "ado-core",
"maintainer": null,
"docs_url": null,
"requires_python": "<3.13,>=3.10",
"maintainer_email": "\"Michael A. Johnston\" <michaelj@ie.ibm.com>",
"keywords": "discovery, exploration, optimization, orchestration, ray",
"author": null,
"author_email": "\"Michael A. Johnston\" <michaelj@ie.ibm.com>",
"download_url": "https://files.pythonhosted.org/packages/74/13/4b4f3624c0d08c8f775e29288c12f789142989b668d2a00ea6f6fe6a6f60/ado_core-1.2.0.tar.gz",
"platform": null,
"description": "# Introduction\n\nThis is the repository for the **a**ccelerated **d**iscovery **o**rchestrator\n(**`ado`**).\n\n**`ado`** is a unified platform for **executing computational experiments at\nscale** and **analysing their results**. It can be extended with new experiments\nor new analysis tools. It allows distributed teams of researchers and engineers\nto collaborate on projects, execute experiments, and share data.\n\nYou can run the experiments and analysis tools already available in **`ado`** in\na distributed, shared, environment with your team. You can also use **`ado`** to\nget features like data-tracking, data-sharing, tool integration and a CLI, for\nyour analysis method or experiment for free.\n\n\ud83e\uddd1\u200d\ud83d\udcbb Using **`ado`** assumes familiarity with command line tools.\n\n\ud83d\udee0\ufe0f Developing **`ado`** requires knowledge of python.\n\n## Key Features\n\n- \ud83d\udcbb _CLI_: Our human-centric CLI follows [best practices](https://clig.dev)\n- \ud83e\udd1d _Projects_: Allow distributed groups of users to\n [collaborate and share data](https://ibm.github.io/ado/resources/metastore)\n- \ud83d\udd0c _Extendable_: Easily\n [add new experiments](https://ibm.github.io/ado/actuators/creating-custom-experiments),\n [optimizers or other tools.](https://ibm.github.io/ado/operators/creating-operators)\n- \u2699\ufe0f _Scalable_: We use [ray](https://ray.io) as our execution engine allowing\n experiments and tools to easily scale\n- \u267b\ufe0f _Automatic data-reuse_: Avoid repeating work with\n [transparent reuse of experiment results](https://ibm.github.io/ado/core-concepts/data-sharing).\n `ado` internal protocols ensure this happens only when it makes sense\n- \ud83d\udd17 _Provenance_: As you work, the relationship between the data you create and\n operations you perform are\n [automatically tracked](https://ibm.github.io/ado/getting-started/ado#ado-show-related)\n- \ud83d\udd0e _Optimization and sampling_: Out-of-the-box, leverage powerful optimization\n methods\n [via `raytune`](https://ibm.github.io/ado/operators/optimisation-with-ray-tune)\n or use our\n [flexible in built sampler](https://ibm.github.io/ado/operators/random-walk)\n\n### Foundation Model Experimentation\n\nWe have developed `ado` plugins providing advanced experiments for testing\nfoundation-models:\n\n- \u23f1\ufe0f\n [fine-tuning performance benchmarking](https://ibm.github.io/ado/actuators/sft-trainer)\n- \u23f1\ufe0f inference performance benchmarking (using the\n [vLLM performance benchmark](https://docs.vllm.ai/en/stable/api/vllm/benchmarks/serve.html))\n- **COMING SOON** \ud83d\udd2e inference and fine-tuning prediction\n\n## Requirements\n\nA basic installation of `ado` only requires a recent Python version (3.10+).\nThis will allow you to run\n[many of our examples](https://ibm.github.io/ado/examples/examples) and explore\nado features.\n\n### Additional Requirements\n\nSome advanced features have additional requirements:\n\n<!-- markdownlint-disable descriptive-link-text -->\n- **Distributed Projects** **_(Optional)_**: To support projects with multiple\n users you will need a remote, accessible, MySQL database. See\n [here](https://ibm.github.io/ado/getting-started/installing-backend-services#using-the-distributed-mysql-backend-for-ado)\n for more\n- **Multi-Node Execution** **_(Optional)_**: To support multi-node or scaling\n execution you may need a multi-node RayCluster. See\n [here](https://ibm.github.io/ado/getting-started/installing-backend-services#deploying-kuberay-and-creating-a-raycluster)\n for more details\n<!-- markdownlint-enable descriptive-link-text -->\n\nIn addition `ado` plugins may have additional requirements for executing\n**_realistic_** experiments. For example,\n\n- **_Fine-Tuning Benchmarking_**: Requires a\n [RayCluster with GPUs](https://ibm.github.io/ado/actuators/sft-trainer#configure-your-raycluster)\n- **_vLLM Performance Benchmarking_**: Requires an OpenShift cluster with GPUs\n\n## Install\n\nTo install you can execute the following (we recommend you set up a virtual\nenvironment)\n\n```commandline\ngit clone https://github.com/IBM/ado.git\ncd ado\npip install .\n```\n\nAlternate instructions to install `ado` can be found here:\n<https://ibm.github.io/ado/getting-started/install/>\n\n## Development\n\nInstructions for developing ado are available in [DEVELOPING](DEVELOPING.md).\n\n### Testing\n\nTo run unit-tests read [tests/README.md](tests/README.md).\n\n## Example\n\nThis video shows listing\n[actuators](website/docs/actuators/working-with-actuators.md) and getting the\ndetails of an experiment.\n\nCheck [demo](https://ibm.github.io/ado/getting-started/demo) for more videos.\n\n[](https://github.com/user-attachments/assets/fc4862f3-763b-4967-ab3c-4bd359900a50)\n\n## Technical Report\n\nFor more details on the Discovery Spaces concept underlying ado, please refer to\nthis [technical report](https://arxiv.org/abs/2506.21467).\n\n## Acknowledgement\n\nThis project is partially funded by the European Union through the Smart\nNetworks and Services Joint Undertaking (SNS JU) under grant agreement No.\n101192750 (Project 6G-DALI).\n",
"bugtrack_url": null,
"license": null,
"summary": "ado is a unified platform for executing computational experiments at scale and analysing their results. It can be easily extended with new experiments or new analysis tools. It allows distributed teams of researchers and engineers to collaborate on projects, execute experiments, and share data.",
"version": "1.2.0",
"project_urls": {
"Changelog": "https://github.com/ibm/ado/CHANGELOG.md",
"Homepage": "https://ibm.github.io/ado/",
"Issues": "https://github.com/ibm/ado/issues",
"Repository": "https://github.com/ibm/ado.git"
},
"split_keywords": [
"discovery",
" exploration",
" optimization",
" orchestration",
" ray"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "8bbf778c55d66c15e397276089c5f3f875cdd2fdaab403e3cdd122c7fb9fea10",
"md5": "afc60e6c1f2fb0a56ad0dc50841e0e35",
"sha256": "42d9d395e9329840dca487ad6e9ff9a5d6674e3292ae82d2f9861614ec228047"
},
"downloads": -1,
"filename": "ado_core-1.2.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "afc60e6c1f2fb0a56ad0dc50841e0e35",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<3.13,>=3.10",
"size": 313781,
"upload_time": "2025-11-06T15:21:02",
"upload_time_iso_8601": "2025-11-06T15:21:02.545088Z",
"url": "https://files.pythonhosted.org/packages/8b/bf/778c55d66c15e397276089c5f3f875cdd2fdaab403e3cdd122c7fb9fea10/ado_core-1.2.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "74134b4f3624c0d08c8f775e29288c12f789142989b668d2a00ea6f6fe6a6f60",
"md5": "2cbbccd1c23b941d76822a6a4e75aae8",
"sha256": "dade374295caa3bbe45ef4684e93ff77ef312964fcbc1a6bc3bdd774dd3b626c"
},
"downloads": -1,
"filename": "ado_core-1.2.0.tar.gz",
"has_sig": false,
"md5_digest": "2cbbccd1c23b941d76822a6a4e75aae8",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "<3.13,>=3.10",
"size": 17405494,
"upload_time": "2025-11-06T15:21:09",
"upload_time_iso_8601": "2025-11-06T15:21:09.139554Z",
"url": "https://files.pythonhosted.org/packages/74/13/4b4f3624c0d08c8f775e29288c12f789142989b668d2a00ea6f6fe6a6f60/ado_core-1.2.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-11-06 15:21:09",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "ibm",
"github_project": "ado",
"travis_ci": false,
"coveralls": true,
"github_actions": false,
"requirements": [],
"tox": true,
"lcname": "ado-core"
}
