Skip to content
Snippets Groups Projects
Select Git revision
  • bootstrap_theme
  • ci-experiments
  • colab
  • master default
  • os-travis
  • remove-lbm
  • test-travis-test
  • textures
  • torch_native-rewrite
  • waldwiesen-simulation
  • weilin
  • 0.3.3
  • 0.3.2
  • 0.3.1
  • 0.3.0
  • 0.2.2
  • 0.2.1
  • 0.2.0
  • 0.1.4
  • 0.1.3
  • 0.1.2
  • 0.1.1
  • 0.1.0
23 results
You can move around the graph by using the arrow keys.
Created with Raphaël 2.2.017Nov1330Oct267615Sep1114Aug23Jun2214Mar1228Feb27262417131211654331Jan302923211715133230Dec191716229Nov2825241811531Oct302928252423171514126Sep2523201918171615131253228Aug212019181715141312987611JulApply dirty hack when adjoint calculation depends on size of forward0.3.3 master re…0.3.3 master remote_mirror_258c1b3866dc038ea96bda0a5c5b0613/masterRemove diagnostic pragmasGive torch ops nice namesAdd get_code_strUpdateAutomatically don't differentiate for indexVectorDIRTY HACK!!Allow return types for WrapperFunctionAllow injecting class definitions into Python bindingsAdd printing for BlockWithOutBracesSupport dicts in ast_dictFix imports of pystencils_autodiff.lbmSupport symbolic index assignmentsSet module_name in TorchModuleTry to generate more swapsRemove encoding headerLintFix code for indexarray autodiffAlways skip weird tf_mad_gradient_checkAutoformatRemove pyconrad from a testAuto-formatFix indentation bugAdapt module.tmpl.cpp to MSVC __restrictIgnore complexity warnings in flake8 configLintAdd TODOfix requires_gradAllow calling Torch kernels with dynamic shapeChange back to data_ptr (data is deprecatedUse cudaFix gpuAssert for MicrosoftAvoid making AdjointField custom if forward field CUSTOMFix test custom printerSkip test_dynamic_functionFix test_module_printing_globalsFix bug in TorchModule.compile that requires `compile_module_name` to be setRemove show_code_debug (DebugFrameworkPrinter is a hack!)Use suffix in destructuring for WalberlaFieldUse suffix in destructuring for WalberlaFieldsTrigger magic