{
"info": {
"author": "Kenneth S Goodman",
"author_email": "kennethsgoodman@yahoo.com",
"bugtrack_url": null,
"classifiers": [
"License :: OSI Approved :: MIT License",
"Operating System :: OS Independent",
"Programming Language :: Python :: 3"
],
"description": "\n[](https://travis-ci.org/kennethgoodman/lazy_numpy)\n[](https://codecov.io/gh/kennethgoodman/lazy_numpy)\n\n# lazynumpy\na lazy evaluated wrapper around numpy\n\nWhat is gained?\n\n* Chained matrix multiplication will be minimized by keeping the values of the other arrays in memory and solving the associative problem that minimizes the number of computations.\n - only keeps one copy of each matrix [Memory optimization in progress]\n* Allow partial matrix returns withou calculating the entire matrix [In Progress]\n\n\nIf you have three matrices with dimensions as below there are two ways to do the matrix multiplication to find the answer:\n\n![](\"https://latex.codecogs.com/gif.latex?A_{1000\\&space;*\\1}&space;*&space;B_{1\\&space;*\\1000}&space;*&space;C_{1000\\&space;*\\1000}&space;=&space;D_{1000\\&space;*\\1000}\" "\\"A_{1000\\")
\n\nEither:\n\n![](\"https://latex.codecogs.com/gif.latex?[1]\\&space;\\&space;(A_{1000\\&space;*\\1}&space;*&space;B_{1\\&space;*\\1000})&space;*&space;C_{1000\\&space;*\\1000}&space;=&space;D_{1000\\&space;*\\1000}\" "\\"[1]\\")
\n\nor\n\n![](\"https://latex.codecogs.com/gif.latex?[2]\\&space;\\&space;A_{1000\\&space;*\\1}&space;*&space;(B_{1\\&space;*\\1000}&space;*&space;C_{1000\\&space;*\\1000})&space;=&space;D_{100\\10001000}\" "\\"[2]\\")
\n\n[1] will take `1000 * 1 * 1000` operations to calculate `A * B` plus `1000 * 1000 * 1000` operations to calculate `(A * B) * C`. The total sum to calculate `A * B * C` is equal to `1000^3 + 1000^2`.\n\n[2] will take `1 * 1000 * 1000` operations to calculate `B * C` plus `1000 * 1 * 1000` operations to calculate `A * (B * C)`. The total sum to calculate `A * B * C` is equal to `1000^2 + 1000^2` which means the optimal multiplication order will be ~500 faster.\n\nIf you run [the simple example](https://github.com/kennethgoodman/lazynumpy/blob/master/examples/simple_faster_calculation.py) you should see a significant speed up. On my computer there is a 50x speedup with only three matrix calculations.\n\n\n",
"description_content_type": "text/markdown",
"docs_url": null,
"download_url": "",
"downloads": {
"last_day": -1,
"last_month": -1,
"last_week": -1
},
"home_page": "https://github.com/kennethgoodman/lazynumpy",
"keywords": "",
"license": "",
"maintainer": "",
"maintainer_email": "",
"name": "lazynumpy",
"package_url": "https://pypi.org/project/lazynumpy/",
"platform": "",
"project_url": "https://pypi.org/project/lazynumpy/",
"project_urls": {
"Homepage": "https://github.com/kennethgoodman/lazynumpy"
},
"release_url": "https://pypi.org/project/lazynumpy/0.0.1/",
"requires_dist": null,
"requires_python": "",
"summary": "A wrapper around numpy that does lazy evaluations to optimize for chained matrix multiplication",
"version": "0.0.1"
},
"last_serial": 4723359,
"releases": {
"0.0.1": [
{
"comment_text": "",
"digests": {
"md5": "a96d43eb047bea30b9cfd5795e4416e0",
"sha256": "be3bc493a48ed1e580f60ffaf1579da58f3fd27934952b80bfc053e9c6758458"
},
"downloads": -1,
"filename": "lazynumpy-0.0.1-py3-none-any.whl",
"has_sig": false,
"md5_digest": "a96d43eb047bea30b9cfd5795e4416e0",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": null,
"size": 8657,
"upload_time": "2019-01-21T21:26:22",
"url": "https://files.pythonhosted.org/packages/f5/58/71b57c8d75bd4a95d882fe4e64e1d69d16aa34bd77c763e88ecd40718ae3/lazynumpy-0.0.1-py3-none-any.whl"
},
{
"comment_text": "",
"digests": {
"md5": "a9e6556176f035128ab0ad2891817698",
"sha256": "02b44607b7070bb02e769962a818bbeed7c2be1b1ca333227d8a90f2c90c9f01"
},
"downloads": -1,
"filename": "lazynumpy-0.0.1.tar.gz",
"has_sig": false,
"md5_digest": "a9e6556176f035128ab0ad2891817698",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 5812,
"upload_time": "2019-01-21T21:26:24",
"url": "https://files.pythonhosted.org/packages/d6/fc/dc9cfc701faa2b080da06ba62c2fddff00e6041928de86c15fa5cb8e128d/lazynumpy-0.0.1.tar.gz"
}
]
},
"urls": [
{
"comment_text": "",
"digests": {
"md5": "a96d43eb047bea30b9cfd5795e4416e0",
"sha256": "be3bc493a48ed1e580f60ffaf1579da58f3fd27934952b80bfc053e9c6758458"
},
"downloads": -1,
"filename": "lazynumpy-0.0.1-py3-none-any.whl",
"has_sig": false,
"md5_digest": "a96d43eb047bea30b9cfd5795e4416e0",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": null,
"size": 8657,
"upload_time": "2019-01-21T21:26:22",
"url": "https://files.pythonhosted.org/packages/f5/58/71b57c8d75bd4a95d882fe4e64e1d69d16aa34bd77c763e88ecd40718ae3/lazynumpy-0.0.1-py3-none-any.whl"
},
{
"comment_text": "",
"digests": {
"md5": "a9e6556176f035128ab0ad2891817698",
"sha256": "02b44607b7070bb02e769962a818bbeed7c2be1b1ca333227d8a90f2c90c9f01"
},
"downloads": -1,
"filename": "lazynumpy-0.0.1.tar.gz",
"has_sig": false,
"md5_digest": "a9e6556176f035128ab0ad2891817698",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 5812,
"upload_time": "2019-01-21T21:26:24",
"url": "https://files.pythonhosted.org/packages/d6/fc/dc9cfc701faa2b080da06ba62c2fddff00e6041928de86c15fa5cb8e128d/lazynumpy-0.0.1.tar.gz"
}
]
}