- 02 12月, 2020 1 次提交
-
-
由 Shenghang Tsai 提交于
Former-commit-id: bdb27546064ce15707b176159885fc7a1777a6fc
-
- 30 11月, 2020 4 次提交
-
-
由 guo ran 提交于
* add combined margin loss * refine * skip_unless_1n4d * skip if cpu only Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: f4bf35f7
-
由 OuYang Yu 提交于
* Replace the py instruction with CFG Instruction * move RunInstruction to pybind & refactor EagerOneflow's interface by cfg * use forward declaration * Remove unused import * fix code style * Adjust import order * replace py parallel_conf proto to cfg * fix test_cpu_only_user_op parallel_conf * move RunInstruction api to oneflow_api.vm * fix MakeMachineId2DeviceIdList * remove useless line in oneflow_internal.i * replace args str to cfg_obj in python callback * add forward declear of InstructionListProto * cancel forward declear of InstructionListProto * fix a name spelling mistake * use the CFG object in the Python Callback * use cfg in the py Callback * fix redundant conversions * fix template bug * fix template * virtual const tmplate cfg * fix template.cfg.h * update cfg * update cfg Co-authored-by: clackhan <han_binbin@163.com> Co-authored-by: Noneflow-bot <69100618+oneflow-bot@users.noreply.github.com> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: cad594bd
-
由 Juncheng 提交于
* Add NaiveB2PSubTskGphBuilder * refine * refine * refine * refine Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: 9f504dfc
-
- 29 11月, 2020 1 次提交
-
-
由 daquexian 提交于
* disable new checkpoint by default temporarily Signed-off-by: Ndaquexian <daquexian566@gmail.com> * disable test Signed-off-by: Ndaquexian <daquexian566@gmail.com> Former-commit-id: c4bbf8c0
-
- 28 11月, 2020 4 次提交
-
-
由 guo ran 提交于
* indexed_slices_model_update handle empty tensor * indexed_slices_sgd Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: 6e71b007
-
由 Shenghang Tsai 提交于
* fix oss list file 100 limit * fix cu100_xla Former-commit-id: 8fb3048c
-
由 Shenghang Tsai 提交于
Former-commit-id: 790d41f3
- 27 11月, 2020 8 次提交
-
-
由 Shenghang Tsai 提交于
Former-commit-id: d535fc05
-
由 Shenghang Tsai 提交于
Former-commit-id: ff9976b2
-
由 Shenghang Tsai 提交于
Former-commit-id: c665f1d9
-
由 Shenghang Tsai 提交于
* add dir * create multiple index * rm cu111_xla * fix format * rename index file * rm legacy 9.0 * update url * rename Former-commit-id: 79eed7fa
-
由 qq_22305325 提交于
* replace ErrorProto with cfg::ErrorProto * fix macro name * optimize error * optimize error * fix code style * Organize the code * fix code style * fix code style * copy cfg head file Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: dc8eb372
-
由 cheng cheng 提交于
* using new chain aglorithm * fix bug of chain merge * fix bug of bfs search * fix order of rm empty adn chain merge * Try NOT merge in MemChain * using DfsTopoForEachNodeSortByDistanceToSink for set order in graph * fix compile err * rollback for topo order * using area id split optimizer with fw/bw chain * NOT consider tick in merge chain * use area id to split optimizer chain and fw/bw chain * remove note * refine code for review * make docker container stay live 1 hour Co-authored-by: NOuYang Yu <xuanjiuye@gmail.com> Co-authored-by: NShenghang Tsai <jackalcooper@gmail.com> Former-commit-id: 65c75854
-
由 Shenghang Tsai 提交于
* add cron * update once a day * fix cuda args * port better error msg * fix gh env * fix cuda version arg * change dist * add flush * allow CI run up to 20 hours when buiding release * check in ensure img * use matrix * call ensure img * fix yaml syntax * use full sha because short not consistent * turn off continue-on-error * add matrix_extra_flags for cpu * add matrix * refactor matrix * check env var * add exist check * fix condition * fix exist path * add more log * Revert "add more log" This reverts commit 85b1649829405d1996247e3425c0dbf3d95e6331 [formerly 6e6c494f1b9bfa3104ece69c6ce35d6328a2c231]. * Update version.py Co-authored-by: NYour Name <you@example.com> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: 0f163678
-
由 guo ran 提交于
Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: 9bf1b0fe
-
- 26 11月, 2020 4 次提交
-
-
由 daquexian 提交于
* flow.load/save/get_all_variables without large tensor and multi machine support * add lazy blob cache and disable blob_cache after writing * update checkpoint to call the potential slice_assign and read_slice_from_blob method * reformat * new checkpoint supports eager * split mut bn into mutable input bn and output bn * work in eager mode. deprecate checkpoint.init() * slice_assign implementation * new slice op * check step > 0, add more tests, refine the code * revert the initializer changes * remove print * set y to 0 for partialsum * check sbp, fix incorrect attr check * add more tests * rename slice2->logical_slice * update tests * extract common python code into a function * get_size_in_slice -> GetSizeInSlice, rm unused test file * minor update about step > 0 * minor update on tests * add WITH_CUDA guard * integrate with logical slice/slice_assign * set scope according to variable op_conf * initial support of stream init * read_slice_from_blob/as_numpy return nd_idx and set the cpu:0 placement for created variable * extract a 'for_every_slice' function * initializer registration * one meta file per variable * remove mis-added file * code clean * create model io jobs only if legacy model io enabled, update legacy api * add legacy model io test * slice operation optimization * add and update tests * barrier for multi node eager * make sync as a cluster instruction * update test * fix life cycle problem * add python api vm_util.Sync() * make initializer receive a random_seed * Add vm_util.Sync(), remove debug code * resolve TODO, remove __repr__ for now * use compiled op_conf for getting random_seed * UserOpAttrVal -> AttrValue, remove debug code * test another dtype * remove mis-added ) * fix dtype error when shape[axis+1:] is empty * add initializers to check_point * code clean, enable a temporary default checkpoint for test * move legacy implementation to deprecated/ * update deprecated implementation * fix bug in eager, add eager tests and some other minor updates * remove name field in FileBackendBlob, update Load for single variable, and some other minor updates * remove mis-added file * move initializer implementation, some minor changes * disable some bn tests missing checkpoint.init() * fix dtype conversion bug * relex the tolerance of layer_norm test * reformat * minor code clean * use new pybind11 eager sync api * add assignment between memory test * disable optimizers test for now * code clean * reuse CreateEagerVariableBlob * remove mis-added file * unify two read slice function * minor code clean * add initializer_updated to check_point.py * fix typo * resolve merge conflict * restore bn tests * add type annotations, add some comments and minor code clean * add some comments, remove 'need_root_path' parameter * fixup * get parallel_conf from job_set instead of op_attribute * disable two tests involving legacy model io in eager mode * add InitialzierImpl * add InitializerImpl * support load from numpy array, add test * rename and format * Add necessary docs and TODO, improve warning message * ParallelConf4InterfaceOpName->ParallelConf4LazyInterfaceOpName * address some comments * rename api * fix problems * add test_initializer.py * remove unused initializers * remove quantinfo, move new checkpoint to check_point_v2.py * fix crash on checkpoint.init() Signed-off-by: Ndaquexian <daquexian566@gmail.com> * restore optimizer test Signed-off-by: Ndaquexian <daquexian566@gmail.com> * Add GetOpAttributes api Signed-off-by: Ndaquexian <daquexian566@gmail.com> * restore ParallelConf4LazyOp as parallel desc symbol id in op attr doesn't align with that in job set Signed-off-by: Ndaquexian <daquexian566@gmail.com> * Add TestResumeTraining, shrink the large model size Signed-off-by: Ndaquexian <daquexian566@gmail.com> * restore 2n4c ci test Signed-off-by: Ndaquexian <daquexian566@gmail.com> * code clean Signed-off-by: Ndaquexian <daquexian566@gmail.com> * add snapshot_done Signed-off-by: Ndaquexian <daquexian566@gmail.com> * add test_mixed_model, update test_load_numpy Signed-off-by: Ndaquexian <daquexian566@gmail.com> * add flow.sync_default_session in api implementation Signed-off-by: Ndaquexian <daquexian566@gmail.com> * change the default value of ignore_mismatch from False to True to align with existing behavior Signed-off-by: Ndaquexian <daquexian566@gmail.com> * fix wrong initializer in test_mseloss.py and test_bce_loss.py Signed-off-by: Ndaquexian <daquexian566@gmail.com> * ForEachOpNode -> ForEachNode Signed-off-by: Ndaquexian <daquexian566@gmail.com> * fix test_partially_load_numpy Signed-off-by: Ndaquexian <daquexian566@gmail.com> Co-authored-by: Nwanghongsheng <2496533749@qq.com> Co-authored-by: Ncheng cheng <472491134@qq.com> Former-commit-id: 6a1b2253
-
由 ShawnXuan 提交于
* bak * rm usless lines * fix skip * better blob name * check in fix * refine code * fix fmt * fix fmt * rm include Co-authored-by: NTsai <caishenghang@oneflow.org> Co-authored-by: NShenghang Tsai <jackalcooper@gmail.com> Former-commit-id: 3e6f8895
-
由 Shenghang Tsai 提交于
Former-commit-id: 6ef63c96
-
- 25 11月, 2020 7 次提交
-
-
由 Mardino 提交于
* add margin loss * add margin rank loss and test case * add broadcast and grad evalutation * refine * add triplet loss * add docs for triplet loss and margin loss * add test case * add backward test * fix format * add docs * fix gradient check * fix reduce axis and fix test case * fix * fix format * fix format * fix name and todo * fix other name * fix name * add example for bceloss Co-authored-by: NLiang Depeng <liangdepeng@gmail.com> Former-commit-id: 267860bc
-
由 Zhenhua 提交于
* Add typename U for UnsortedSegmentSumGpu * using NdIndexOffsetHelper * Delete GetOutOffset & IsSafeUseIndex32 * debug half kernel * fix (#3755) * update test_unsorted_segment_sum * update tolerance * add specialized case * fix UnsortedSegmentSumKernel * switch typename T and U * Refactor function * Fix CPU build failed Co-authored-by: Nguo ran <360112263@qq.com> Co-authored-by: NJuncheng <liujuncheng1022@gmail.com> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: 6fca270f
-
由 Shenghang Tsai 提交于
* fix include files not copied * larger tol Co-authored-by: NTsai <caishenghang@oneflow.org> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: 282c8f49
-
由 guo ran 提交于
* split like add grad * fix name * fix * fix name * add axis0 test case Co-authored-by: NJuncheng <liujuncheng1022@gmail.com> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: 4cc08ba8
-
由 qq_22305325 提交于
* optimize cfg generator to save time * fix code format * use join to get file_path * use join to get file_path & replace CMAKE_CURRENT_SOURCE_DIR with PROJECT_SOURCE_DIR Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: 02bbfaf8
-
由 Zhenhua 提交于
* add python interface for polyval and test, currently only work for float * add test case for double type * fix polyval op * Add License * format polyval * update polyval op test case * Add polyval bw test * Fix 1n2d test case * rm tensorflow in test case Co-authored-by: Niamyf <yangf@zhejianglab.com> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: df877d3e
-
由 Juncheng 提交于
* multi_count_not_finite op * refine * add cpu * refine * refine * refine * format * add count not finite * fix * Dynamic loss scale * refine * merge * fix * refine * refine Co-authored-by: Nguoran <guoran@oneflow.org> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: 47304ff9
-
- 24 11月, 2020 4 次提交
-
-
由 Mardino 提交于
Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: 1374f045
-
由 guo ran 提交于
Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: 8a738777
-
由 guo ran 提交于
Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: 17f436e2
-
由 guo ran 提交于
Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: e841d975
-
- 23 11月, 2020 3 次提交
-
-
由 Mardino 提交于
* add check in deconv * add check * fix format * fix check in python * fix deconv2d params Co-authored-by: NShenghang Tsai <jackalcooper@gmail.com> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: 20348c38
-
由 Mardino 提交于
* add bceloss * optimize some steps * add bce loss diff evaluate * fix format * update test script * fix annotation * rebuild test code * fix gen_arg * fix arg * fix name * add bceloss * fix format * fix format * Update test.yml Co-authored-by: Ndoombeaker <later@usopp.net> Co-authored-by: Noneflow-bot <69100618+oneflow-bot@users.noreply.github.com> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Co-authored-by: NShenghang Tsai <jackalcooper@gmail.com> Former-commit-id: e2709ccd
- 22 11月, 2020 2 次提交
-
-
由 Li Xinqi 提交于
* rename UserOpAttrVal to AttrValue * Scope::GetAttrValue * add ssp variable proxy pass * AddSspVariableProxy * ssp_config_def.cpp * merge config_def from master * REGISTER_SCOPE_CONFIG_DEF * description for ssp_partition_strategy * fix return type of JobPass::HasState * FlexDef/FlexValue * support recursive flex def * remove field_number * more cfg files * instructions builder * forward declaration instead of include * more test for cfg * revert cfg files * InstructionsBuilder * using std::function as argument of IdCache::FindOrCreate * scope op_collection * include <functional> in framework/interpreter.h * puts more code into WithOptimizerOpCollectionScope * IsInOptimizerOpCollection * include <functional> in symbol_id_cache.h * calculation pass * IsInOptimizerOpCollection -> IsInOptimizerPass * minor refine about spp_config_def.cpp * test for add_ssp_variable_proxy * rm framework/flex * refine add ssp variable proxy pass * refine Error * refine Error * AddScopeToPyStorage * fix test_watch * get scope_symbol_id from current scope * fix assert bug * no longer use scope_proto.symbol_id Co-authored-by: binbinHan <han_binbin@163.com> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: ef6786d5
-
由 Li Xinqi 提交于
* more cfg files * instructions builder * forward declaration instead of include * more test for cfg * revert cfg files * InstructionsBuilder * using std::function as argument of IdCache::FindOrCreate * scope op_collection * include <functional> in framework/interpreter.h * puts more code into WithOptimizerOpCollectionScope * include <functional> in symbol_id_cache.h * calculation pass * refine Error * AddScopeToPyStorage * fix test_watch * get scope_symbol_id from current scope * fix assert bug Co-authored-by: binbinHan <han_binbin@163.com> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: 1467aecf
-
- 21 11月, 2020 2 次提交
-
-
由 guo ran 提交于
* multi_count_not_finite op * refine * add cpu * refine * refine * refine * format * add count not finite * fix Co-authored-by: Nguoran <guoran@oneflow.org> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Co-authored-by: Nliujuncheng <liujuncheng1022@gmail.com> Former-commit-id: 58627eec
-
由 guo ran 提交于
Co-authored-by: Nguoran <guoran@oneflow.org> Co-authored-by: Noneflow-ci-bot <69100618+oneflow-ci-bot@users.noreply.github.com> Former-commit-id: ab5cde94
-