{ "info": { "author": "The Swiss AI Lab IDSIA", "author_email": "mailstorm@googlemail.com", "bugtrack_url": null, "classifiers": [ "Development Status :: 4 - Beta", "Intended Audience :: Developers", "Intended Audience :: Education", "Intended Audience :: Science/Research", "License :: OSI Approved :: MIT License", "Natural Language :: English", "Programming Language :: Python :: 2.7", "Programming Language :: Python :: 3.3" ], "description": "Documentation\n-------------\n\nThe full documentation is at http://brainstorm.rtfd.org.\n\n\n\nHistory\n-------\n\n0.5 (2015-12-01)\n++++++++++++++++\nChanged Behaviour\n\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\"\n* examples now run on CPU by default\n* added ``brainstorm.tools.shuffle_data`` and ``brainstorm.tools.split`` to help with data preparation\n* ``SigmoidCE`` and ``SquaredDifference`` layers now outputs a loss for each dimension instead of summing over features.\n* ``SquaredDifference`` layer does no longer scale by one half.\n* Added a ``SquaredLoss`` layer that computes half the squared difference and\n has an interface that is compatible with the ``SigmoidCE`` and ``SigmoidCE`` layers.\n* Output `probabilities` renamed to `predictions` in ``SigmoidCE`` and ``SigmoidCE`` layers.\n\nNew Features\n\"\"\"\"\"\"\"\"\"\"\"\"\n* added a `use_conv` option to ``brainstorm.tools.create_net_from_spec``\n* added `criterion` option to ``brainstorm.hooks.EarlyStopper`` hook\n* added ``brainstorm.tools.get_network_info`` function that returns information\n about the network as a string\n* added ``brainstorm.tools.extract`` function that applies a network to some\n data and saves a set of requested buffers.\n* ``brainstorm.layers.mask`` layer now supports masking individual features\n* added ``brainstorm.hooks.StopAfterThresholdReached`` hook\n\nImprovements\n\"\"\"\"\"\"\"\"\"\"\"\"\n* EarlyStopper now works for any timescale and interval\n* Recurrent, Lstm, Clockwork, and ClockworkLstm layers now accept inputs of\n arbitrary shape by implicitly flattening them.\n* several fixes to make building the docs easier\n* some performance improvements of NumpyHandler operations ``binarize_t`` and ``index_m_by_v``\n* sped up tests\n* several improvements to installation scripts\n\nBugfixes\n\"\"\"\"\"\"\"\"\n* fixed `sqrt` operation for ``PyCudaHandler``. This should fix problems with BatchNormalization on GPU.\n* fixed a bug for task_type='regression' in ``brainstorm.tools.get_in_out_layers``\n and ``brainstorm.tools.create_net_from_spec``\n* removed defunct name argument from input layer\n* fixed a crash when applying ``brainstorm.hooks.SaveBestNetwork`` to `rolling_training` loss\n* various minor fixes of the ``brainstorm.hooks.BokehVisualizer``\n* fixed a problem with ``sum_t`` operation in ``brainstorm.handlers.PyCudaHandler``\n* fixed a blocksize problem in convolutional and pooling operations in ``brainstorm.handlers.PyCudaHandler``\n\n\n0.5b0 (2015-10-25)\n++++++++++++++++++\n* First release on PyPI.", "description_content_type": null, "docs_url": null, "download_url": "UNKNOWN", "downloads": { "last_day": -1, "last_month": -1, "last_week": -1 }, "home_page": "https://github.com/IDSIA/brainstorm", "keywords": null, "license": "MIT", "maintainer": null, "maintainer_email": null, "name": "brainstorm", "package_url": "https://pypi.org/project/brainstorm/", "platform": "UNKNOWN", "project_url": "https://pypi.org/project/brainstorm/", "project_urls": { "Download": "UNKNOWN", "Homepage": "https://github.com/IDSIA/brainstorm" }, "release_url": "https://pypi.org/project/brainstorm/0.5/", "requires_dist": null, "requires_python": null, "summary": "Fast, flexible and fun neural networks.", "version": "0.5" }, "last_serial": 1841521, "releases": { "0.5": [ { "comment_text": "", "digests": { "md5": "348510a6a73f39a44760da1f1bf56fb5", "sha256": "e8fde7eb1edda32b528db80f36c90abe59647e0ce1e43ebc74ce0f7d285728fd" }, "downloads": -1, "filename": "brainstorm-0.5.tar.gz", "has_sig": false, "md5_digest": "348510a6a73f39a44760da1f1bf56fb5", "packagetype": "sdist", "python_version": "source", "requires_python": null, "size": 246309, "upload_time": "2015-12-01T14:06:18", "url": "https://files.pythonhosted.org/packages/6b/ce/2f4cac3346bb75fbd284d33502df1bfb7a027e9d54c11409af372bc0f7f8/brainstorm-0.5.tar.gz" } ], "0.5b0": [], "0.5b1": [ { "comment_text": "", "digests": { "md5": "71af23a0a93e9ec77e351841efeb5c05", "sha256": "8ac07d61d983ccd00d2d877f42fd90ca29eb66efb3a0a7c9b6e74295723cb99a" }, "downloads": -1, "filename": "brainstorm-0.5b1.tar.gz", "has_sig": false, "md5_digest": "71af23a0a93e9ec77e351841efeb5c05", "packagetype": "sdist", "python_version": "source", "requires_python": null, "size": 240184, "upload_time": "2015-10-27T14:23:22", "url": "https://files.pythonhosted.org/packages/be/91/4716b8773f9a57bfae3f3bfc803790410181cce5cd574063f8820e645e59/brainstorm-0.5b1.tar.gz" } ] }, "urls": [ { "comment_text": "", "digests": { "md5": "348510a6a73f39a44760da1f1bf56fb5", "sha256": "e8fde7eb1edda32b528db80f36c90abe59647e0ce1e43ebc74ce0f7d285728fd" }, "downloads": -1, "filename": "brainstorm-0.5.tar.gz", "has_sig": false, "md5_digest": "348510a6a73f39a44760da1f1bf56fb5", "packagetype": "sdist", "python_version": "source", "requires_python": null, "size": 246309, "upload_time": "2015-12-01T14:06:18", "url": "https://files.pythonhosted.org/packages/6b/ce/2f4cac3346bb75fbd284d33502df1bfb7a027e9d54c11409af372bc0f7f8/brainstorm-0.5.tar.gz" } ] }