{
"info": {
"author": "George Linderman, Gioele La Manno",
"author_email": "george.linderman@gmail.com, gioelelamanno@gmail.com",
"bugtrack_url": null,
"classifiers": [],
"description": "FFT-accelerated Interpolation-based t-SNE (FIt-SNE)\n===================================================\n\nIntroduction\n------------\n\nt-Stochastic Neighborhood Embedding\n(`t-SNE `__) is a highly successful\nmethod for dimensionality reduction and visualization of high\ndimensional datasets. A popular\n`implementation `__ of t-SNE uses\nthe Barnes-Hut algorithm to approximate the gradient at each iteration\nof gradient descent. We modified this implementation as follows:\n\n- Computation of the N-body Simulation: Instead of approximating the\n N-body simulation using Barnes-Hut, we interpolate onto an equispaced\n grid and use FFT to perform the convolution, dramatically reducing\n the time to compute the gradient at each iteration of gradient\n descent. See `this\n `__\n post for some intuition on how it works.\n- Computation of Input Similiarities: Instead of computing nearest\n neighbors using vantage-point trees, we approximate nearest neighbors\n using the `Annoy `__ library. The\n neighbor lookups are multithreaded to take advantage of machines with\n multiple cores. Using \"near\" neighbors as opposed to strictly\n \"nearest\" neighbors is faster, but also has a smoothing effect, which\n can be useful for embedding some datasets (see `Linderman et al.\n (2017) `__). If subtle detail is required\n (e.g. in identifying small clusters), then use vantage-point trees (which is\n also multithreaded in this implementation). \n- Early exaggeration: In `Linderman and Steinerberger\n (2017) `__, we showed that\n appropriately choosing the early exaggeration coefficient can lead to\n improved embedding of swissrolls and other synthetic datase ts\n- Late exaggeration: By increasing the exaggeration coefficient late in\n the optimization process (e.g. after 800 of 1000 iterations) can\n improve separation of the clusters\n\nCheck out our `preprint `__ for more\ndetails and some benchmarks.\n\nThis PyPI package is a Cython wrapper for `FIt-SNE `_\nand was written by `Gioele La Manno `_.\n\nInstallation\n------------\nThe only prerequisite is `FFTW `__. FFTW and fitsne can be installed as follows:\n\n.. code:: bash\n \n conda config --add channels conda-forge #if not already in your channels. Needed for fftw.\n conda install cython numpy fftw \n pip install fitsne\n\nAnd you're good to go!\n\nBug reports, feature requests, etc.\n-------------------------------------\nIf you have any problems with this package, please open an issue on the Github `repository `__.\n\nReferences\n----------\n\nIf you use our software, please cite:\n\nGeorge C. Linderman, Manas Rachh, Jeremy G. Hoskins, Stefan\nSteinerberger, Yuval Kluger. (2017). Efficient Algorithms for\nt-distributed Stochastic Neighborhood Embedding. (2017)\n*arXiv:1712.09005* (`link `__)\n\nOur implementation is derived from the Barnes-Hut implementation:\n\nLaurens van der Maaten (2014). Accelerating t-SNE using tree-based\nalgorithms. Journal of Machine Learning Research, 15(1):3221\u20133245.\n(`link `__)\n",
"description_content_type": "",
"docs_url": null,
"download_url": "https://github.com/KlugerLab/pyFIt-SNE/archive/1.0.1.tar.gz",
"downloads": {
"last_day": -1,
"last_month": -1,
"last_week": -1
},
"home_page": "https://github.com/KlugerLab/FIt-SNE",
"keywords": "tSNE",
"license": "BSD3",
"maintainer": "",
"maintainer_email": "",
"name": "fitsne",
"package_url": "https://pypi.org/project/fitsne/",
"platform": "",
"project_url": "https://pypi.org/project/fitsne/",
"project_urls": {
"Download": "https://github.com/KlugerLab/pyFIt-SNE/archive/1.0.1.tar.gz",
"Homepage": "https://github.com/KlugerLab/FIt-SNE"
},
"release_url": "https://pypi.org/project/fitsne/1.0.1/",
"requires_dist": null,
"requires_python": "",
"summary": "Fast Fourier Transform-accelerated Interpolation-based t-SNE (FIt-SNE)",
"version": "1.0.1"
},
"last_serial": 4496931,
"releases": {
"0.1.10": [
{
"comment_text": "",
"digests": {
"md5": "8874a3d7e7470466ab8d5f0a16fba52e",
"sha256": "cbf0f526718488d006167dcee00ac9d40199b64f869831b8227a1f2d0aed8698"
},
"downloads": -1,
"filename": "fitsne-0.1.10.tar.gz",
"has_sig": false,
"md5_digest": "8874a3d7e7470466ab8d5f0a16fba52e",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 137266,
"upload_time": "2017-12-29T19:28:53",
"url": "https://files.pythonhosted.org/packages/7a/db/2bc55d2ed08824dbf823b281b61b9963464f418f2deaf7d6c01a47ea9d3d/fitsne-0.1.10.tar.gz"
}
],
"0.1.4": [
{
"comment_text": "",
"digests": {
"md5": "3285f08bc529c7373eeb57db217a8d8b",
"sha256": "c7bed9b4799406fe326ffb46d1c3e89d0cfd216a4628561c99881755115e0b57"
},
"downloads": -1,
"filename": "fitsne-0.1.4.tar.gz",
"has_sig": false,
"md5_digest": "3285f08bc529c7373eeb57db217a8d8b",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 135972,
"upload_time": "2017-12-29T14:56:33",
"url": "https://files.pythonhosted.org/packages/9e/eb/0437a9d120f975d0f3fa8bfc09381ff42f24236af80110958b1b6deaf74c/fitsne-0.1.4.tar.gz"
}
],
"0.1.5": [
{
"comment_text": "",
"digests": {
"md5": "7fa3815216920fb10a927cc024831dad",
"sha256": "b0a06dec0b9ed36dd21ccab0002c6795f3942e77ae79c89bb9b8e88f0a71f404"
},
"downloads": -1,
"filename": "fitsne-0.1.5.tar.gz",
"has_sig": false,
"md5_digest": "7fa3815216920fb10a927cc024831dad",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 135992,
"upload_time": "2017-12-29T15:58:32",
"url": "https://files.pythonhosted.org/packages/d3/1f/d4654f9d8bcc3e3f26d825269cee6ebfc1f595c2dada50d91ebd9331e591/fitsne-0.1.5.tar.gz"
}
],
"0.1.6": [
{
"comment_text": "",
"digests": {
"md5": "f10b0fa441cf2d157ac54ac16cdaddf5",
"sha256": "d7a62e09624c9d8db837308ce25d70e6f12b201fa0ffa8787973161a4bb6446f"
},
"downloads": -1,
"filename": "fitsne-0.1.6.tar.gz",
"has_sig": false,
"md5_digest": "f10b0fa441cf2d157ac54ac16cdaddf5",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 135997,
"upload_time": "2017-12-29T16:14:18",
"url": "https://files.pythonhosted.org/packages/c8/15/ed36cacdddb08d5765616f602b70ad8bf8daa95ce3903da3271ab7bfca65/fitsne-0.1.6.tar.gz"
}
],
"0.1.7": [
{
"comment_text": "",
"digests": {
"md5": "8826983c90bdcb9b119d9d37b740c0c4",
"sha256": "f64fb882e826f2ba6ab2a00bbc69b2d1126f019f01c6ef479f2550f47229c521"
},
"downloads": -1,
"filename": "fitsne-0.1.7.tar.gz",
"has_sig": false,
"md5_digest": "8826983c90bdcb9b119d9d37b740c0c4",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 137430,
"upload_time": "2017-12-29T18:50:41",
"url": "https://files.pythonhosted.org/packages/1d/b0/1507b43aacce04621cb808362b57f9e0f374af8adef3ed259f8a5cf4d40e/fitsne-0.1.7.tar.gz"
}
],
"0.1.8": [
{
"comment_text": "",
"digests": {
"md5": "beb2a311372cd69c900c08216bcc2bfd",
"sha256": "52e461caca48f63f4df27eb7ad2974eaa23a993270e505598ec6f48c6718fce1"
},
"downloads": -1,
"filename": "fitsne-0.1.8.tar.gz",
"has_sig": false,
"md5_digest": "beb2a311372cd69c900c08216bcc2bfd",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 137427,
"upload_time": "2017-12-29T18:52:04",
"url": "https://files.pythonhosted.org/packages/33/b3/7ce9ede6a7696c1462c65ecc319ad06cdbe3de1391b230d5c72bbb87a58f/fitsne-0.1.8.tar.gz"
}
],
"0.1.9": [
{
"comment_text": "",
"digests": {
"md5": "0ddc7e855dde3b255372fad54a4aa6e6",
"sha256": "f44ba77aaaeb346aecad24ea67982190e2bb395363bbc3af14d230114e6dbdf9"
},
"downloads": -1,
"filename": "fitsne-0.1.9.tar.gz",
"has_sig": false,
"md5_digest": "0ddc7e855dde3b255372fad54a4aa6e6",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 137397,
"upload_time": "2017-12-29T19:02:37",
"url": "https://files.pythonhosted.org/packages/e9/84/024295c526b9852e708aa07af5335fdea3b834b82217c9993bc41612c303/fitsne-0.1.9.tar.gz"
}
],
"0.2.0": [
{
"comment_text": "",
"digests": {
"md5": "eb72b4dcb7901c64baa778ef50e7935a",
"sha256": "1de1f029cd7b4b03cae4013025b3dbb59a6644bdd6a83ec1d8718c45e0613888"
},
"downloads": -1,
"filename": "fitsne-0.2.0.tar.gz",
"has_sig": false,
"md5_digest": "eb72b4dcb7901c64baa778ef50e7935a",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 135246,
"upload_time": "2018-01-16T22:52:22",
"url": "https://files.pythonhosted.org/packages/50/10/b4496d66c4fb8337a65857ee2f02a15ced8cb04080c2781c3ab0bf4f6878/fitsne-0.2.0.tar.gz"
}
],
"0.2.1": [
{
"comment_text": "",
"digests": {
"md5": "af579df2caab9be0d24da4c952a65eee",
"sha256": "0a097b4377e4a4cc5ac44da480f202f6c38a09fce3a62b46a8c79fed97777250"
},
"downloads": -1,
"filename": "fitsne-0.2.1.tar.gz",
"has_sig": false,
"md5_digest": "af579df2caab9be0d24da4c952a65eee",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 141036,
"upload_time": "2018-02-20T03:38:19",
"url": "https://files.pythonhosted.org/packages/0f/0e/c3daeabf812927eba24a84716a4d40af4ee77e526464b7d1d629e1074c57/fitsne-0.2.1.tar.gz"
}
],
"0.2.2": [
{
"comment_text": "",
"digests": {
"md5": "8217adcfcdd9a4806d8521a715a7de51",
"sha256": "ae6238bb39ce000077ed26534942d51e0cdb520de32d3707a6c44b68082285c1"
},
"downloads": -1,
"filename": "fitsne-0.2.2.tar.gz",
"has_sig": false,
"md5_digest": "8217adcfcdd9a4806d8521a715a7de51",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 141143,
"upload_time": "2018-03-22T13:36:31",
"url": "https://files.pythonhosted.org/packages/fb/5d/fea125898f35a254c013d35aae76f181d37389b644a3dfb19e745f0ea48b/fitsne-0.2.2.tar.gz"
}
],
"0.2.3": [
{
"comment_text": "",
"digests": {
"md5": "b24ef45f510c5fd65152745d8308414e",
"sha256": "7eb9ceaab1c585d99fbbd941983325b664e68d2aea841dfc8f969ab19a0296fe"
},
"downloads": -1,
"filename": "fitsne-0.2.3.tar.gz",
"has_sig": false,
"md5_digest": "b24ef45f510c5fd65152745d8308414e",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 138444,
"upload_time": "2018-05-23T18:25:40",
"url": "https://files.pythonhosted.org/packages/d1/7c/dad9c910cc7a7bb8f54f49c98e3150c031142900e62862779574d6a3d0e4/fitsne-0.2.3.tar.gz"
}
],
"0.2.4": [
{
"comment_text": "",
"digests": {
"md5": "004afd07af0f943be0d29be2ba6f363a",
"sha256": "34901724d5bae01396e61a5d095c733243c3c9a3d5598eb80f91c4e29a6ba729"
},
"downloads": -1,
"filename": "fitsne-0.2.4.tar.gz",
"has_sig": false,
"md5_digest": "004afd07af0f943be0d29be2ba6f363a",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 138447,
"upload_time": "2018-07-04T17:25:05",
"url": "https://files.pythonhosted.org/packages/03/c8/629278116e9ece13527d8010dbbf85ad15e0521ddcdedd6be488194fb0d9/fitsne-0.2.4.tar.gz"
}
],
"0.2.5": [
{
"comment_text": "",
"digests": {
"md5": "4389433caffa9c45278bd7c4902e2863",
"sha256": "b0cb8191302773924091f8cd86b094c5ea0ca734944194f717ff9035de2508d2"
},
"downloads": -1,
"filename": "fitsne-0.2.5.tar.gz",
"has_sig": false,
"md5_digest": "4389433caffa9c45278bd7c4902e2863",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 138383,
"upload_time": "2018-07-04T18:40:51",
"url": "https://files.pythonhosted.org/packages/a7/55/b5d36bce446920e17efe6e558cd19894204dc6bdb2678776e309aeb73a58/fitsne-0.2.5.tar.gz"
}
],
"0.2.6": [
{
"comment_text": "",
"digests": {
"md5": "2c3742ab35a0b5f6624bd9f1b7fb54ab",
"sha256": "7f614de8349942b4b325f1b47622897f1748eaa7b0569e9c0d5b2426f7d12b87"
},
"downloads": -1,
"filename": "fitsne-0.2.6.tar.gz",
"has_sig": false,
"md5_digest": "2c3742ab35a0b5f6624bd9f1b7fb54ab",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 168567,
"upload_time": "2018-09-23T19:44:38",
"url": "https://files.pythonhosted.org/packages/11/8d/1d499fec426cd94f5654a7916c1490701aa5521314ccf5216a3c3e014dd4/fitsne-0.2.6.tar.gz"
}
],
"0.2.7": [
{
"comment_text": "",
"digests": {
"md5": "2a2c2ab8c3ba95d48bba1b83c4c2bf2d",
"sha256": "a6a0a5ff03f34353e56cba378e7fed8be732c56a55591ce134cd126526c40f18"
},
"downloads": -1,
"filename": "fitsne-0.2.7.tar.gz",
"has_sig": false,
"md5_digest": "2a2c2ab8c3ba95d48bba1b83c4c2bf2d",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 171901,
"upload_time": "2018-09-23T20:04:04",
"url": "https://files.pythonhosted.org/packages/6c/a1/9199394182cfb8fe9dafc3e4600397a03c9911d6287e969acf7660f2c55d/fitsne-0.2.7.tar.gz"
}
],
"0.2.8": [
{
"comment_text": "",
"digests": {
"md5": "9deebb52086a3c8ee0e2028a5c1c443f",
"sha256": "22793dd89475d384d001473e744f1d844eb136f3cba1821cd231acef52e33675"
},
"downloads": -1,
"filename": "fitsne-0.2.8.tar.gz",
"has_sig": false,
"md5_digest": "9deebb52086a3c8ee0e2028a5c1c443f",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 169966,
"upload_time": "2018-09-26T21:39:24",
"url": "https://files.pythonhosted.org/packages/e3/11/68c09830d0ded56031c523cb7cf1c83c5f86e0b7c02810792ce97fbef0d0/fitsne-0.2.8.tar.gz"
}
],
"0.2.9": [
{
"comment_text": "",
"digests": {
"md5": "9759853dd0185114d21ddad4235839cd",
"sha256": "2912313e6591bfd9d7c92b0e06f4a3c40f636d83a3b856b0b44aca1aecf0271c"
},
"downloads": -1,
"filename": "fitsne-0.2.9.tar.gz",
"has_sig": false,
"md5_digest": "9759853dd0185114d21ddad4235839cd",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 170185,
"upload_time": "2018-09-26T21:52:19",
"url": "https://files.pythonhosted.org/packages/46/1c/1f28558d49fbc8b719103cf999d12fa76f9e80959b1066803b0bc56bb015/fitsne-0.2.9.tar.gz"
}
],
"1.0.1": [
{
"comment_text": "",
"digests": {
"md5": "38123fbb2d9d2320196ca28750b5c7e2",
"sha256": "5658cbe8f98ce05b5a17636233c9dd5476c7444e285ca198de504b42ee5e6f29"
},
"downloads": -1,
"filename": "fitsne-1.0.1.tar.gz",
"has_sig": false,
"md5_digest": "38123fbb2d9d2320196ca28750b5c7e2",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 171635,
"upload_time": "2018-11-17T11:54:49",
"url": "https://files.pythonhosted.org/packages/b6/84/7f8ea3c3837228ed6eb4f2773e764443e4a92d4e5ffd4f6e85517def667e/fitsne-1.0.1.tar.gz"
}
]
},
"urls": [
{
"comment_text": "",
"digests": {
"md5": "38123fbb2d9d2320196ca28750b5c7e2",
"sha256": "5658cbe8f98ce05b5a17636233c9dd5476c7444e285ca198de504b42ee5e6f29"
},
"downloads": -1,
"filename": "fitsne-1.0.1.tar.gz",
"has_sig": false,
"md5_digest": "38123fbb2d9d2320196ca28750b5c7e2",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 171635,
"upload_time": "2018-11-17T11:54:49",
"url": "https://files.pythonhosted.org/packages/b6/84/7f8ea3c3837228ed6eb4f2773e764443e4a92d4e5ffd4f6e85517def667e/fitsne-1.0.1.tar.gz"
}
]
}