apstone

Name	apstone JSON
Version	0.0.8 JSON
	download
home_page	https://github.com/ykk648/apstone
Summary	ai_power base stone
upload_time	2024-01-09 02:12:13
maintainer
docs_url	None
author	ykk648
requires_python
license
keywords
VCS
bugtrack_url
requirements	No requirements were recorded.
Travis-CI	No Travis.
coveralls test coverage	No coveralls.

            ### Introduction

Base stone of AI_power, maintain all inference of AI_Power models.

#### Wrapper

- Supply different model infer wrapper, including ONNX/TensorRT/Torch JIT;
- Support onnx different Execution Providers (EP) , including cpu/gpu/trt/trt16/int8;
- High level mmlab model (converted) infer wrapper, including MMPose/MMDet;

#### Model Convert

- torch2jit torch2onnx etc.
- detectron2 to onnx
- modelscope to onnx
- onnx2simple2trt
- tf2pb2onnx

#### Model Tools

- torch model edit
- onnx model shape/speed test (different EP)
- common scripts from onnxruntime

### Usage

#### onnx model speed test
```python
from apstone import ONNXModel

onnx_p = 'pretrain_models/sr_lib/realesr-general-x4v3-dynamic.onnx'
input_dynamic_shape = (1, 3, 96, 72)  # None
# cpu gpu trt trt16 int8
ONNXModel(onnx_p, provider='cpu', debug=True, input_dynamic_shape=input_dynamic_shape).speed_test()
```

### Install

```sh
pip install apstone
```

#### Envs

| Execution Providers | Needs                                                       |
| ------------------- | ----------------------------------------------------------- |
| cpu                 | pip install onnxruntime                                     |
| gpu                 | pip install onnxruntime-gpu                                 |
| trt/trt16/int8      | onnxruntime-gpu compiled with tensorrt EP                   |
| TensorRT            | pip install tensorrt pycuda                                 |
| torch JIT           | install [pytorch](https://pytorch.org/get-started/locally/) |


            

Raw data

            {
    "_id": null,
    "home_page": "https://github.com/ykk648/apstone",
    "name": "apstone",
    "maintainer": "",
    "docs_url": null,
    "requires_python": "",
    "maintainer_email": "",
    "keywords": "",
    "author": "ykk648",
    "author_email": "ykk648@gmail.com",
    "download_url": "https://files.pythonhosted.org/packages/75/5b/6ac2ffe7fd2dee1e8f522c904acdaa517ab58bd46db9549404f533113eed/apstone-0.0.8.tar.gz",
    "platform": null,
    "description": "### Introduction\n\nBase stone of AI_power, maintain all inference of AI_Power models.\n\n#### Wrapper\n\n- Supply different model infer wrapper, including ONNX/TensorRT/Torch JIT;\n- Support onnx different Execution Providers (EP) , including cpu/gpu/trt/trt16/int8;\n- High level mmlab model (converted) infer wrapper, including MMPose/MMDet;\n\n#### Model Convert\n\n- torch2jit torch2onnx etc.\n- detectron2 to onnx\n- modelscope to onnx\n- onnx2simple2trt\n- tf2pb2onnx\n\n#### Model Tools\n\n- torch model edit\n- onnx model shape/speed test (different EP)\n- common scripts from onnxruntime\n\n### Usage\n\n#### onnx model speed test\n```python\nfrom apstone import ONNXModel\n\nonnx_p = 'pretrain_models/sr_lib/realesr-general-x4v3-dynamic.onnx'\ninput_dynamic_shape = (1, 3, 96, 72)  # None\n# cpu gpu trt trt16 int8\nONNXModel(onnx_p, provider='cpu', debug=True, input_dynamic_shape=input_dynamic_shape).speed_test()\n```\n\n### Install\n\n```sh\npip install apstone\n```\n\n#### Envs\n\n| Execution Providers | Needs                                                       |\n| ------------------- | ----------------------------------------------------------- |\n| cpu                 | pip install onnxruntime                                     |\n| gpu                 | pip install onnxruntime-gpu                                 |\n| trt/trt16/int8      | onnxruntime-gpu compiled with tensorrt EP                   |\n| TensorRT            | pip install tensorrt pycuda                                 |\n| torch JIT           | install [pytorch](https://pytorch.org/get-started/locally/) |\n\n",
    "bugtrack_url": null,
    "license": "",
    "summary": "ai_power base stone",
    "version": "0.0.8",
    "project_urls": {
        "Bug Tracker": "https://github.com/ykk648/apstone/issues",
        "Homepage": "https://github.com/ykk648/apstone"
    },
    "split_keywords": [],
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "blake2b_256": "755b6ac2ffe7fd2dee1e8f522c904acdaa517ab58bd46db9549404f533113eed",
                "md5": "8a820b749c0b159508e5934469a3bd4a",
                "sha256": "0e93f99275affbe0abf733f2ec065a23cfe6bc64365d11e666c3b3f374bc6f60"
            },
            "downloads": -1,
            "filename": "apstone-0.0.8.tar.gz",
            "has_sig": false,
            "md5_digest": "8a820b749c0b159508e5934469a3bd4a",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": null,
            "size": 41309,
            "upload_time": "2024-01-09T02:12:13",
            "upload_time_iso_8601": "2024-01-09T02:12:13.769820Z",
            "url": "https://files.pythonhosted.org/packages/75/5b/6ac2ffe7fd2dee1e8f522c904acdaa517ab58bd46db9549404f533113eed/apstone-0.0.8.tar.gz",
            "yanked": false,
            "yanked_reason": null
        }
    ],
    "upload_time": "2024-01-09 02:12:13",
    "github": true,
    "gitlab": false,
    "bitbucket": false,
    "codeberg": false,
    "github_user": "ykk648",
    "github_project": "apstone",
    "travis_ci": false,
    "coveralls": false,
    "github_actions": false,
    "lcname": "apstone"
}