[{"_1":2,"_155":-5,"_156":-5},"loaderData",{"_3":4,"_13":14,"_21":22,"_134":135},"root",{"_5":-5,"_6":-7,"_7":8,"_9":-7,"_10":11,"_12":-7},"currentUser","primaryEmail","siteVerification","6LeIxAcTAAAAAJcZVRqyHh71UMIEGNQ_MXjiZKhI","toast","enableDev",false,"incidentStatus","routes/_layout",{"_15":-5,"_16":17,"_18":17,"_19":11,"_20":11},"username","isAskGordonEnabled",true,"isUsingNewTemplate","showMinimalLanding","showHoneypotTrap","routes/_layout.r.$namespace.$reponame",{"_23":24,"_25":-5,"_26":11,"_27":11,"_28":29,"_30":31,"_32":-5,"_33":11,"_34":35,"_84":-5,"_85":86,"_125":-5,"_126":11,"_127":17,"_128":11,"_129":-5,"_130":11,"_131":-5,"_132":133},"canonicalUrl","https://hub.docker.com/r/tobegit3hub/hplearn/tags","mcpServer","canSeeDockerfile","canSeeAutobuilds","namespace","tobegit3hub","reponame","hplearn","cloudBuildSource","isAutobuildsAvailable","repo",{"_36":37,"_38":39,"_41":42,"_43":44,"_45":11,"_46":47,"_48":49,"_50":51,"_53":31,"_28":29,"_54":55,"_56":40,"_57":58,"_59":60,"_61":62,"_63":-5,"_64":58,"_65":66,"_67":11,"_68":29,"_69":70,"_75":17,"_76":77,"_81":29,"_82":83},"categories",[],"content_types",[40],"image","date_registered","2017-08-26T12:53:13.059703Z","description","HPLearn is the high performance machine learning system in pure C++.","is_private","last_updated","2017-09-18T02:02:16.683012Z","last_modified","2024-10-16T13:48:34.145251Z","media_types",[52],"application/vnd.docker.container.image.v1+json","name","pull_count",23019,"repository_type","star_count",0,"status",1,"status_description","active","source","collaborator_count","full_description","# HPLearn\n\n## Introduction\n\nHPLearn is the high performance machine learning system in pure C++.\n\n- [x] Basic mathematics operations\n- [x] Automatic partial derivative with chain rule \n- [x] Imperative and declarative computations\n- [x] Override +-*/ operations for graph building\n- [x] CMake integration for Linux/Mac/Windows\n- [x] Bazel integration for build and run\n- [x] Op/Graph/Session TensorFlow-like APIs\n- [ ] GPU integration with Nividia CUDA library\n\n## Installation\n\nBuild from scratch with `cmake`. \n\n```\ncmake .\n\nmake\n```\n\nOr compile code with `bazel`.\n\n```\nbazel build hplearn:main\n```\n\nOr run directly with `docker`.\n\n```\ndocker run -it tobegit3hub/hplearn bash\n```\n\n## Usage\n\n### Import hplearn libraries\n\n```\n#include \"op.h\"\n#include \"graph.h\"\n#include \"session.h\"\n#include \"optimizer.h\"\n\nusing namespace hplearn;\n```\n\n### Basic operations\n\n```\nVariableOp* firstOp = new VariableOp(20.2);\nVariableOp* secondOp = new VariableOp(10.1);\nAddOp* addOp = new AddOp(firstOp, secondOp);\n\ncout << \"Basic operation result: \" << addOp->forward() << endl;\n```\n\n### Overrided operations\n\n```\nGraph* graph = new Graph();\nVariableOp* variableOp1 = new VariableOp(20.2);\nVariableOp* variableOp2 = new VariableOp(10.1);\nAddOp* addOp = (AddOp*) (*variableOp1 + *variableOp2);\nMinusOp* minusOp = (MinusOp*) (*variableOp1 - *variableOp2);\nMultipleOp* multipleOp = (MultipleOp*) (*variableOp1 * *variableOp2);\nDivideOp* divideOp = (DivideOp*) (*variableOp1 / *variableOp2);\n\nSession* session = new Session(graph);\ncout << \"Overrided + operator result: \" << to_string(session->run(addOp->getName())) << endl;\ncout << \"Overrided - operator result: \" << to_string(session->run(minusOp->getName())) << endl;\ncout << \"Overrided * operator result: \" << to_string(session->run(multipleOp->getName())) << endl;\ncout << \"Overrided / operator result: \" << to_string(session->run(divideOp->getName())) << endl;\n```\n\n### Use placeholder\n\n```\nGraph* graph = new Graph();\nPlaceholderOp* placeholderOp1 = new PlaceholderOp();\nPlaceholderOp* placeholderOp2 = new PlaceholderOp();\nAddOp* addOp = new AddOp(placeholderOp1, placeholderOp2);\n\nSession* session = new Session(graph);\nmap feedDict;\nfeedDict[placeholderOp1->getName()] = 20.2;\nfeedDict[placeholderOp2->getName()] = 10.1;\ndouble result = session->run(addOp->getName(), feedDict);\ncout << \"Use placeholder result: \" << to_string(result) << endl;\n```\n\n### Linear model\n\n```\n// Define graph \nGraph* graph = new Graph();\nVariableOp* weights = new VariableOp(0.0);\nVariableOp* bias = new VariableOp(0.0);\nPlaceholderOp* x = new PlaceholderOp();\nPlaceholderOp* y = new PlaceholderOp();\n\nOp* multipleOp = *weights * *x;\nOp* predictOp = *multipleOp + *bias;\nOp* minusOp = *y - *predictOp;\nSquareOp* lossOp = new SquareOp(minusOp);\nGradientDescentOptimizer* optimizer = new GradientDescentOptimizer(graph, learningRate);\nOptimizerMinimizeOp* trainOp = (OptimizerMinimizeOp*) optimizer->minimize(lossOp);\n\n// Define session\nSession* sess = new Session(graph);\nmap feedDict;\n\nfor (int i=0; igetName()] = feature;\n feedDict[y->getName()] = label;\n sess->run(trainOp->getName(), feedDict);\n\n // Print statistics\n double lossValue = sess->run(lossOp->getName(), feedDict);\n double weightValue = sess->run(weights->getName());\n double biasValue = sess->run(bias->getName());\n cout << \"Epoch: \" << to_string(i) << \", loss: \" << to_string(lossValue) << \", weight: \" \n << to_string(weightValue) << \", bias: \" << to_string(biasValue) << endl;\n```\n\n## Performance\n\nBenchmark with [TensorFlow](https://github.com/tensorflow/tensorflow) while running 100000 epochs of [add operation](./benchmark/benchmark_add_operation.cpp), [multiple operation](./benchmark/benchmark_multiple_operation.cpp) and [linear regression](benchmark/benchmark_linear_regression.cpp). \n\n![](./images/benchmark_add_operation.png)\n\n![](./images/benchmark_multiple_operation.png)\n\n![](./images/benchmark_linear_regression.png)\n\n## Contribution\n\nThe HPLearn project is mostly inspired by [TensorFlow](https://github.com/tensorflow/tensorflow) and [MiniFlow](https://github.com/tobegit3hub/miniflow).\n\nGitHub issues and pull-requests are highly appreciated and feel free to make your contribution.\n","has_starred","hub_user","immutable_tags_settings",{"_71":11,"_72":73},"enabled","rules",[74],".*","is_automated","permissions",{"_78":17,"_79":11,"_80":11},"read","write","admin","user","storage_size",1637691383,"orgData","recentTags",[87],{"_88":40,"_89":90,"_91":92,"_93":94,"_95":96,"_46":114,"_115":90,"_116":29,"_117":52,"_53":118,"_119":120,"_121":105,"_122":114,"_123":62,"_124":17},"content_type","creator",20953,"full_size",151042231,"id",14531127,"images",[97],{"_98":99,"_100":101,"_102":103,"_104":105,"_106":107,"_108":109,"_110":103,"_111":-5,"_112":92,"_59":62,"_113":-5},"architecture","amd64","digest","sha256:8778ec6f9288a36de3375ab79f61163348acd653dbbb942ff75b86bbd9f32b5a","features","","last_pulled","2026-09-22T14:29:25.543946786Z","last_pushed","2017-09-18T02:02:17Z","os","linux","os_features","os_version","size","variant","2017-09-18T02:02:16.353406Z","last_updater","last_updater_username","media_type","latest","repository",1767969,"tag_last_pulled","tag_last_pushed","tag_status","v2","sbxKitMetadata","isPotentialSpam","isHubSbxKitsEnabled","showSbxLogoOverride","sbxTrademarkText","displaySecurityMetadata","tagAttestations","tagsHaveAttestations",[],"routes/_layout.r.$namespace.$reponame.tags",{"_34":136,"_143":144,"_132":153,"_154":11},{"_36":137,"_38":138,"_41":42,"_43":44,"_45":11,"_46":47,"_48":49,"_50":139,"_53":31,"_28":29,"_54":55,"_56":40,"_57":58,"_59":60,"_61":62,"_63":-5,"_64":58,"_65":66,"_67":11,"_68":29,"_69":140,"_75":17,"_76":142,"_81":29,"_82":83},[],[40],[52],{"_71":11,"_72":141},[74],{"_78":17,"_79":11,"_80":11},"tags",{"_145":60,"_146":-5,"_147":-5,"_148":149},"count","next","previous","results",[150],{"_88":40,"_89":90,"_91":92,"_93":94,"_95":151,"_46":114,"_115":90,"_116":29,"_117":52,"_53":118,"_119":120,"_121":105,"_122":114,"_123":62,"_124":17},[152],{"_98":99,"_100":101,"_102":103,"_104":105,"_106":107,"_108":109,"_110":103,"_111":-5,"_112":92,"_59":62,"_113":-5},[],"canDeleteTags","actionData","errors"]

tobegit3hub/hplearn

Tags for tobegit3hub/hplearn