[GitHub] szha closed issue #11501: Tracking issue for tests with fixed random seed
szha closed issue #11501: Tracking issue for tests with fixed random seed URL: https://github.com/apache/incubator-mxnet/issues/11501 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11741: test_sparse_ndarray.test_sparse_nd_save_load has fixed seed that can mask flakiness
szha opened a new issue #11741: test_sparse_ndarray.test_sparse_nd_save_load has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11741 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11740: test_sparse_ndarray.test_sparse_nd_pickle has fixed seed that can mask flakiness
szha opened a new issue #11740: test_sparse_ndarray.test_sparse_nd_pickle has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11740 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11739: test_quantization.test_get_optimal_thresholds has fixed seed that can mask flakiness
szha opened a new issue #11739: test_quantization.test_get_optimal_thresholds has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11739 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11738: test_optimizer.test_signum has fixed seed that can mask flakiness
szha opened a new issue #11738: test_optimizer.test_signum has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11738 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11736: test_optimizer.test_nadam has fixed seed that can mask flakiness
szha opened a new issue #11736: test_optimizer.test_nadam has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11736 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11737: test_optimizer.test_nag has fixed seed that can mask flakiness
szha opened a new issue #11737: test_optimizer.test_nag has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11737 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11734: test_operator_gpu.test_psroipooling_with_type has fixed seed that can mask flakiness
szha opened a new issue #11734: test_operator_gpu.test_psroipooling_with_type has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11734 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11735: test_optimizer.test_ftml has fixed seed that can mask flakiness
szha opened a new issue #11735: test_optimizer.test_ftml has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11735 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11733: test_operator_gpu.test_pooling_with_type has fixed seed that can mask flakiness
szha opened a new issue #11733: test_operator_gpu.test_pooling_with_type has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11733 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11732: test_operator_gpu.test_pooling_versions has fixed seed that can mask flakiness
szha opened a new issue #11732: test_operator_gpu.test_pooling_versions has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11732 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11731: test_operator_gpu.test_ifft has fixed seed that can mask flakiness
szha opened a new issue #11731: test_operator_gpu.test_ifft has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11731 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11730: test_operator_gpu.test_fft has fixed seed that can mask flakiness
szha opened a new issue #11730: test_operator_gpu.test_fft has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11730 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] NEWPLAN edited a comment on issue #11684: fix compile error in debug model
NEWPLAN edited a comment on issue #11684: fix compile error in debug model URL: https://github.com/apache/incubator-mxnet/pull/11684#issuecomment-404730258 Hi, My environment: OS: CentOS Linux release 7.3.1611 (Core) GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-28) CPU: Intel(R) Xeon(R) CPU E5-2620 v4 @ 2.10GHz DRAM: 64GB 2133MHZ G++: g++ (GCC) 4.8.5 20150623 (Red Hat 4.8.5-28) NVCC: Built on Fri_Sep__1_21:08:03_CDT_2017 + Cuda compilation tools, release 9.0, V9.0.176 NCCL: 2.1 CUDA: 9 cuDNN: 7.0 command for compiling: make -j $(nproc) USE_OPENCV=1 USE_BLAS=openblas USE_CUDA=1 USE_CUDA_PATH=/usr/local/cuda USE_CUDNN=1 USE_DIST_KVSTORE=1 USE_NCCL=1 USE_CPP_PACKAGE=1 DEBUG=1 there are two errors: 1: src/kvstore/./kvstore_nccl.h:345:28: error: ‘struct mxnet::kvstore::KVStoreNCCL::BufferEntry’ has no member named ‘ctx’ assert(root == buf.ctx().dev_id); When i fixed this error with uncomment the assert code and complie again, got the second error: a - build/src/kvstore/kvstore_utils_gpu.o /usr/lib/gcc/x86_64-redhat-linux/4.8.5/../../../../lib64/crti.o: In function `_init': (.init+0x7): relocation truncated to fit: R_X86_64_GOTPCREL against undefined symbol `__gmon_start__' /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::NumResponse(int)': /usr/include/c++/4.8.2/x86_64-redhat-linux/bits/gthr-default.h:747:(.text+0x5): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::AddResponse(int, int)': /usr/include/c++/4.8.2/x86_64-redhat-linux/bits/gthr-default.h:747:(.text+0x7d): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::WaitRequest(int)': /usr/include/c++/4.8.2/x86_64-redhat-linux/bits/gthr-default.h:747:(.text+0xfe): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::Receiving()': /home/yang/software/incubator-mxnet/3rdparty/ps-lite/src/customer.cc:49:(.text+0x1d2): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `std::string::_Rep::_S_empty_rep_storage@@GLIBCXX_3.4' defined in .bss section in /usr/lib/gcc/x86_64-redhat-linux/4.8.5/libstdc++.so /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::Receiving()': /usr/include/c++/4.8.2/x86_64-redhat-linux/bits/gthr-default.h:747:(.text+0x2dd): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::Receiving()': /usr/include/c++/4.8.2/bits/stl_construct.h:102:(.text+0x46f): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /usr/include/c++/4.8.2/bits/stl_construct.h:102:(.text+0x555): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /usr/include/c++/4.8.2/bits/stl_construct.h:102:(.text+0x61a): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /usr/include/c++/4.8.2/bits/stl_construct.h:102:(.text+0x6ce): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::Receiving()': /usr/include/c++/4.8.2/x86_64-redhat-linux/bits/gthr-default.h:747:(.text+0x773): additional relocation overflows omitted from the output collect2: error: ld returned 1 exit status make: *** [lib/libmxnet.so] Error 1 make: *** Waiting for unfinished jobs /usr/lib/gcc/x86_64-redhat-linux/4.8.5/../../../../lib64/crt1.o:(.eh_frame+0x20): relocation truncated to fit: R_X86_64_PC32 against `.text' /usr/lib/gcc/x86_64-redhat-linux/4.8.5/../../../../lib64/crti.o: In function `_init': (.init+0x7): relocation truncated to fit: R_X86_64_GOTPCREL against undefined symbol `__gmon_start__' /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In
[GitHub] NEWPLAN edited a comment on issue #11684: fix compile error in debug model
NEWPLAN edited a comment on issue #11684: fix compile error in debug model URL: https://github.com/apache/incubator-mxnet/pull/11684#issuecomment-404730258 Hi, My environment: OS: CentOS Linux release 7.3.1611 (Core) GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-28) CPU: Intel(R) Xeon(R) CPU E5-2620 v4 @ 2.10GHz DRAM: 64GB 2133MHZ G++: g++ (GCC) 4.8.5 20150623 (Red Hat 4.8.5-28) NVCC: Built on Fri_Sep__1_21:08:03_CDT_2017 + Cuda compilation tools, release 9.0, V9.0.176 NCCL: 2.1 CUDA: 9 cuDNN: 7.0 command for compiling: make -j $(nproc) USE_OPENCV=1 USE_BLAS=openblas USE_CUDA=1 USE_CUDA_PATH=/usr/local/cuda USE_CUDNN=1 USE_DIST_KVSTORE=1 USE_NCCL=1 USE_CPP_PACKAGE=1 DEBUG=1 there are two errors: 1: src/kvstore/./kvstore_nccl.h:345:28: error: ‘struct mxnet::kvstore::KVStoreNCCL::BufferEntry’ has no member named ‘ctx’ assert(root == buf.ctx().dev_id); When i fixed this error as uncommend and complie again, got the second error: a - build/src/kvstore/kvstore_utils_gpu.o /usr/lib/gcc/x86_64-redhat-linux/4.8.5/../../../../lib64/crti.o: In function `_init': (.init+0x7): relocation truncated to fit: R_X86_64_GOTPCREL against undefined symbol `__gmon_start__' /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::NumResponse(int)': /usr/include/c++/4.8.2/x86_64-redhat-linux/bits/gthr-default.h:747:(.text+0x5): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::AddResponse(int, int)': /usr/include/c++/4.8.2/x86_64-redhat-linux/bits/gthr-default.h:747:(.text+0x7d): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::WaitRequest(int)': /usr/include/c++/4.8.2/x86_64-redhat-linux/bits/gthr-default.h:747:(.text+0xfe): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::Receiving()': /home/yang/software/incubator-mxnet/3rdparty/ps-lite/src/customer.cc:49:(.text+0x1d2): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `std::string::_Rep::_S_empty_rep_storage@@GLIBCXX_3.4' defined in .bss section in /usr/lib/gcc/x86_64-redhat-linux/4.8.5/libstdc++.so /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::Receiving()': /usr/include/c++/4.8.2/x86_64-redhat-linux/bits/gthr-default.h:747:(.text+0x2dd): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::Receiving()': /usr/include/c++/4.8.2/bits/stl_construct.h:102:(.text+0x46f): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /usr/include/c++/4.8.2/bits/stl_construct.h:102:(.text+0x555): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /usr/include/c++/4.8.2/bits/stl_construct.h:102:(.text+0x61a): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /usr/include/c++/4.8.2/bits/stl_construct.h:102:(.text+0x6ce): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::Receiving()': /usr/include/c++/4.8.2/x86_64-redhat-linux/bits/gthr-default.h:747:(.text+0x773): additional relocation overflows omitted from the output collect2: error: ld returned 1 exit status make: *** [lib/libmxnet.so] Error 1 make: *** Waiting for unfinished jobs /usr/lib/gcc/x86_64-redhat-linux/4.8.5/../../../../lib64/crt1.o:(.eh_frame+0x20): relocation truncated to fit: R_X86_64_PC32 against `.text' /usr/lib/gcc/x86_64-redhat-linux/4.8.5/../../../../lib64/crti.o: In function `_init': (.init+0x7): relocation truncated to fit: R_X86_64_GOTPCREL against undefined symbol `__gmon_start__' /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function
[GitHub] szha opened a new issue #11728: test_operator_gpu.test_deformable_convolution_with_type has fixed seed that can mask flakiness
szha opened a new issue #11728: test_operator_gpu.test_deformable_convolution_with_type has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11728 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11729: test_operator_gpu.test_deformable_psroipooling_with_type has fixed seed that can mask flakiness
szha opened a new issue #11729: test_operator_gpu.test_deformable_psroipooling_with_type has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11729 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11727: test_operator_gpu.test_convolution_with_type has fixed seed that can mask flakiness
szha opened a new issue #11727: test_operator_gpu.test_convolution_with_type has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11727 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] NEWPLAN commented on issue #11684: fix compile error in debug model
NEWPLAN commented on issue #11684: fix compile error in debug model URL: https://github.com/apache/incubator-mxnet/pull/11684#issuecomment-404730258 Hi, My environment: OS: CentOS Linux release 7.3.1611 (Core) GCC: gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-28) CPU: Intel(R) Xeon(R) CPU E5-2620 v4 @ 2.10GHz DRAM: 64GB 2133MHZ G++: g++ (GCC) 4.8.5 20150623 (Red Hat 4.8.5-28) NVCC: Built on Fri_Sep__1_21:08:03_CDT_2017 + Cuda compilation tools, release 9.0, V9.0.176 command for compiling: make -j $(nproc) USE_OPENCV=1 USE_BLAS=openblas USE_CUDA=1 USE_CUDA_PATH=/usr/local/cuda USE_CUDNN=1 USE_DIST_KVSTORE=1 USE_NCCL=1 USE_CPP_PACKAGE=1 DEBUG=1 there are two errors: 1: src/kvstore/./kvstore_nccl.h:345:28: error: ‘struct mxnet::kvstore::KVStoreNCCL::BufferEntry’ has no member named ‘ctx’ assert(root == buf.ctx().dev_id); When i fixed this error as uncommend and complie again, got the second error: a - build/src/kvstore/kvstore_utils_gpu.o /usr/lib/gcc/x86_64-redhat-linux/4.8.5/../../../../lib64/crti.o: In function `_init': (.init+0x7): relocation truncated to fit: R_X86_64_GOTPCREL against undefined symbol `__gmon_start__' /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::NumResponse(int)': /usr/include/c++/4.8.2/x86_64-redhat-linux/bits/gthr-default.h:747:(.text+0x5): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::AddResponse(int, int)': /usr/include/c++/4.8.2/x86_64-redhat-linux/bits/gthr-default.h:747:(.text+0x7d): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::WaitRequest(int)': /usr/include/c++/4.8.2/x86_64-redhat-linux/bits/gthr-default.h:747:(.text+0xfe): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::Receiving()': /home/yang/software/incubator-mxnet/3rdparty/ps-lite/src/customer.cc:49:(.text+0x1d2): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `std::string::_Rep::_S_empty_rep_storage@@GLIBCXX_3.4' defined in .bss section in /usr/lib/gcc/x86_64-redhat-linux/4.8.5/libstdc++.so /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::Receiving()': /usr/include/c++/4.8.2/x86_64-redhat-linux/bits/gthr-default.h:747:(.text+0x2dd): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::Receiving()': /usr/include/c++/4.8.2/bits/stl_construct.h:102:(.text+0x46f): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /usr/include/c++/4.8.2/bits/stl_construct.h:102:(.text+0x555): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /usr/include/c++/4.8.2/bits/stl_construct.h:102:(.text+0x61a): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /usr/include/c++/4.8.2/bits/stl_construct.h:102:(.text+0x6ce): relocation truncated to fit: R_X86_64_GOTPCREL against symbol `__pthread_key_create@@GLIBC_2.2.5' defined in .text section in /lib64/libpthread.so.0 /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::Receiving()': /usr/include/c++/4.8.2/x86_64-redhat-linux/bits/gthr-default.h:747:(.text+0x773): additional relocation overflows omitted from the output collect2: error: ld returned 1 exit status make: *** [lib/libmxnet.so] Error 1 make: *** Waiting for unfinished jobs /usr/lib/gcc/x86_64-redhat-linux/4.8.5/../../../../lib64/crt1.o:(.eh_frame+0x20): relocation truncated to fit: R_X86_64_PC32 against `.text' /usr/lib/gcc/x86_64-redhat-linux/4.8.5/../../../../lib64/crti.o: In function `_init': (.init+0x7): relocation truncated to fit: R_X86_64_GOTPCREL against undefined symbol `__gmon_start__' /home/yang/software/incubator-mxnet/3rdparty/ps-lite/build/libps.a(customer.o): In function `ps::Customer::NumResponse(int)':
[GitHub] szha opened a new issue #11726: test_operator_gpu.test_bilinear_sampler_with_type has fixed seed that can mask flakiness
szha opened a new issue #11726: test_operator_gpu.test_bilinear_sampler_with_type has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11726 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11724: test_operator.test_sequence_last has fixed seed that can mask flakiness
szha opened a new issue #11724: test_operator.test_sequence_last has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11724 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11725: test_operator.test_unary_math_operators has fixed seed that can mask flakiness
szha opened a new issue #11725: test_operator.test_unary_math_operators has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11725 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11723: test_operator.test_roipooling has fixed seed that can mask flakiness
szha opened a new issue #11723: test_operator.test_roipooling has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11723 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11721: test_operator.test_laop_4 has fixed seed that can mask flakiness
szha opened a new issue #11721: test_operator.test_laop_4 has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11721 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11722: test_operator.test_rnnrelu_bidirectional has fixed seed that can mask flakiness
szha opened a new issue #11722: test_operator.test_rnnrelu_bidirectional has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11722 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11720: test_operator.test_laop_3 has fixed seed that can mask flakiness
szha opened a new issue #11720: test_operator.test_laop_3 has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11720 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11718: test_operator.test_laop has fixed seed that can mask flakiness
szha opened a new issue #11718: test_operator.test_laop has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11718 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11719: test_operator.test_laop_2 has fixed seed that can mask flakiness
szha opened a new issue #11719: test_operator.test_laop_2 has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11719 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11716: test_operator.test_elementwise_sum has fixed seed that can mask flakiness
szha opened a new issue #11716: test_operator.test_elementwise_sum has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11716 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11717: test_operator.test_l2_normalization has fixed seed that can mask flakiness
szha opened a new issue #11717: test_operator.test_l2_normalization has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11717 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11715: test_operator.test_dropout has fixed seed that can mask flakiness
szha opened a new issue #11715: test_operator.test_dropout has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11715 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] haojin2 commented on issue #10387: Flaky test(scala): test_arange
haojin2 commented on issue #10387: Flaky test(scala): test_arange URL: https://github.com/apache/incubator-mxnet/issues/10387#issuecomment-404729617 Plus, seems like a duplicate issue for #8834 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11714: test_operator.test_dot has fixed seed that can mask flakiness
szha opened a new issue #11714: test_operator.test_dot has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11714 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11713: test_operator.test_deformable_psroipooling has fixed seed that can mask flakiness
szha opened a new issue #11713: test_operator.test_deformable_psroipooling has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11713 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11712: test_operator.test_binary_math_operators has fixed seed that can mask flakiness
szha opened a new issue #11712: test_operator.test_binary_math_operators has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11712 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[incubator-mxnet] branch master updated: [MXNET-323] Improve performance of broadcast ops backward pass (#11252)
This is an automated email from the ASF dual-hosted git repository.
jxie pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git
The following commit(s) were added to refs/heads/master by this push:
new 32d298b [MXNET-323] Improve performance of broadcast ops backward
pass (#11252)
32d298b is described below
commit 32d298bd35a1b8513c51fbe505c2ee878480cb0f
Author: Anirudh Subramanian
AuthorDate: Thu Jul 12 22:27:30 2018 -0700
[MXNET-323] Improve performance of broadcast ops backward pass (#11252)
* Fix cached broadcast
* Fix
* Use seq_reduce_compute logic for stable sum
* Fix lint
* Add declarations
* Add elemwise binary broadcast op cuh file
* Add license for elemwise_binary_broadcast_op-inl.cuh
* Fix broadcast
* Fix indentation
* Use cpu and gpu instead of xpu
---
src/operator/tensor/broadcast_reduce-inl.cuh | 4 ++
src/operator/tensor/broadcast_reduce-inl.h | 48 +++-
.../tensor/elemwise_binary_broadcast_op-inl.cuh| 64 ++
src/operator/tensor/elemwise_binary_broadcast_op.h | 35 +++-
4 files changed, 135 insertions(+), 16 deletions(-)
diff --git a/src/operator/tensor/broadcast_reduce-inl.cuh
b/src/operator/tensor/broadcast_reduce-inl.cuh
index 5c9b45f..be3d1f9 100644
--- a/src/operator/tensor/broadcast_reduce-inl.cuh
+++ b/src/operator/tensor/broadcast_reduce-inl.cuh
@@ -613,6 +613,10 @@ void Reduce(Stream *s, const TBlob& small, const
OpReqType req,
ReduceImpl(stream, small, req, big, workspace,
config);
}
+template
+void ReduceWithExtraMem(Stream* s, const TBlob& small, const OpReqType
req,
+const Tensor& workspace, const TBlob&
big) {};
+
template
void Reduce(Stream *s, const TBlob& small, const OpReqType req,
const Tensor& workspace, const TBlob& big,
diff --git a/src/operator/tensor/broadcast_reduce-inl.h
b/src/operator/tensor/broadcast_reduce-inl.h
index 713e3f1..39c68bd 100644
--- a/src/operator/tensor/broadcast_reduce-inl.h
+++ b/src/operator/tensor/broadcast_reduce-inl.h
@@ -31,6 +31,7 @@
#include
#include
#include "../mshadow_op.h"
+#include "../operator_common.h"
namespace mxnet {
namespace op {
@@ -205,16 +206,57 @@ void seq_reduce_compute(const int N, const int M, const
bool addto,
}
}
-template
-void Reduce(Stream *s, const TBlob& small, const OpReqType req,
+template
+void seq_reduce_compute_extra_mem(const int N, const int M, const bool addto,
+ const DType* big, DType* small,
+ const Shape bshape,
+ const Shape sshape,
+ const Shape rshape,
+ const Shape rstride,
+ const index_t* ws_dptr) {
+ #pragma omp parallel for
num_threads(engine::OpenMP::Get()->GetRecommendedOMPThreadCount())
+ for (int idx = 0; idx < N; ++idx) {
+Shape coord = unravel(idx, sshape);
+int j = ravel(coord, bshape);
+DType val, residual;
+Reducer::SetInitValue(val, residual);
+for (int k = 0; k < M; ++k) {
+ Reducer::Reduce(val, OP::Map(big[j + ws_dptr[k]]), residual);
+}
+assign([idx], addto, val);
+ }
+}
+
+template
+void Reduce(Stream* s, const TBlob& small, const OpReqType req,
const Tensor& workspace, const TBlob& big) {
if (req == kNullOp) return;
Shape rshape, rstride;
diff(small.shape_.get(), big.shape_.get(), , );
int N = small.shape_.Size(), M = rshape.Size();
seq_reduce_compute(
+N, M, req == kAddTo, big.dptr(), small.dptr(),
+big.shape_.get(), small.shape_.get(), rshape, rstride);
+}
+
+template
+void ReduceWithExtraMem(Stream* s, const TBlob& small, const OpReqType
req,
+const Tensor& workspace, const TBlob&
big) {
+ using namespace mxnet_op;
+ if (req == kNullOp) return;
+ Shape rshape, rstride;
+ diff(small.shape_.get(), big.shape_.get(), , );
+ index_t* ws_dptr = reinterpret_cast(workspace.dptr_);
+ int N = small.shape_.Size(), M = rshape.Size();
+ #pragma omp parallel for
num_threads(engine::OpenMP::Get()->GetRecommendedOMPThreadCount())
+ for (int k = 0; k < M; k++) {
+Shape coord = unravel(k, rshape);
+ws_dptr[k] = dot(coord, rstride);
+ }
+
+ seq_reduce_compute_extra_mem(
N, M, req == kAddTo, big.dptr(), small.dptr(),
big.shape_.get(),
-small.shape_.get(), rshape, rstride);
+small.shape_.get(), rshape, rstride, ws_dptr);
}
template
diff --git a/src/operator/tensor/elemwise_binary_broadcast_op-inl.cuh
b/src/operator/tensor/elemwise_binary_broadcast_op-inl.cuh
new file mode 100644
index 000..1998576
--- /dev/null
+++ b/src/operator/tensor/elemwise_binary_broadcast_op-inl.cuh
@@ -0,0 +1,64 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor
[GitHub] szha opened a new issue #11711: test_operator.test_bilinear_sampler has fixed seed that can mask flakiness
szha opened a new issue #11711: test_operator.test_bilinear_sampler has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11711 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] piiswrong closed pull request #11252: [MXNET-323] Improve performance of broadcast ops backward pass
piiswrong closed pull request #11252: [MXNET-323] Improve performance of
broadcast ops backward pass
URL: https://github.com/apache/incubator-mxnet/pull/11252
This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:
As this is a foreign pull request (from a fork), the diff is supplied
below (as it won't show otherwise due to GitHub magic):
diff --git a/src/operator/tensor/broadcast_reduce-inl.cuh
b/src/operator/tensor/broadcast_reduce-inl.cuh
index b6bb39a1984..f1a55315da7 100644
--- a/src/operator/tensor/broadcast_reduce-inl.cuh
+++ b/src/operator/tensor/broadcast_reduce-inl.cuh
@@ -602,6 +602,10 @@ void Reduce(Stream *s, const TBlob& small, const
OpReqType req,
ReduceImpl(stream, small, req, big, workspace,
config);
}
+template
+void ReduceWithExtraMem(Stream* s, const TBlob& small, const OpReqType
req,
+const Tensor& workspace, const TBlob&
big) {};
+
template
void Reduce(Stream *s, const TBlob& small, const OpReqType req,
const Tensor& workspace, const TBlob& big,
diff --git a/src/operator/tensor/broadcast_reduce-inl.h
b/src/operator/tensor/broadcast_reduce-inl.h
index 76ec92a9e72..c969a219e64 100644
--- a/src/operator/tensor/broadcast_reduce-inl.h
+++ b/src/operator/tensor/broadcast_reduce-inl.h
@@ -31,6 +31,7 @@
#include
#include
#include "../mshadow_op.h"
+#include "../operator_common.h"
namespace mxnet {
namespace op {
@@ -204,16 +205,57 @@ void seq_reduce_compute(const int N, const int M, const
bool addto,
}
}
-template
-void Reduce(Stream *s, const TBlob& small, const OpReqType req,
+template
+void seq_reduce_compute_extra_mem(const int N, const int M, const bool addto,
+ const DType* big, DType* small,
+ const Shape bshape,
+ const Shape sshape,
+ const Shape rshape,
+ const Shape rstride,
+ const index_t* ws_dptr) {
+ #pragma omp parallel for
num_threads(engine::OpenMP::Get()->GetRecommendedOMPThreadCount())
+ for (int idx = 0; idx < N; ++idx) {
+Shape coord = unravel(idx, sshape);
+int j = ravel(coord, bshape);
+DType val, residual;
+Reducer::SetInitValue(val, residual);
+for (int k = 0; k < M; ++k) {
+ Reducer::Reduce(val, OP::Map(big[j + ws_dptr[k]]), residual);
+}
+assign([idx], addto, val);
+ }
+}
+
+template
+void Reduce(Stream* s, const TBlob& small, const OpReqType req,
const Tensor& workspace, const TBlob& big) {
if (req == kNullOp) return;
Shape rshape, rstride;
diff(small.shape_.get(), big.shape_.get(), , );
int N = small.shape_.Size(), M = rshape.Size();
seq_reduce_compute(
+N, M, req == kAddTo, big.dptr(), small.dptr(),
+big.shape_.get(), small.shape_.get(), rshape, rstride);
+}
+
+template
+void ReduceWithExtraMem(Stream* s, const TBlob& small, const OpReqType
req,
+const Tensor& workspace, const TBlob&
big) {
+ using namespace mxnet_op;
+ if (req == kNullOp) return;
+ Shape rshape, rstride;
+ diff(small.shape_.get(), big.shape_.get(), , );
+ index_t* ws_dptr = reinterpret_cast(workspace.dptr_);
+ int N = small.shape_.Size(), M = rshape.Size();
+ #pragma omp parallel for
num_threads(engine::OpenMP::Get()->GetRecommendedOMPThreadCount())
+ for (int k = 0; k < M; k++) {
+Shape coord = unravel(k, rshape);
+ws_dptr[k] = dot(coord, rstride);
+ }
+
+ seq_reduce_compute_extra_mem(
N, M, req == kAddTo, big.dptr(), small.dptr(),
big.shape_.get(),
-small.shape_.get(), rshape, rstride);
+small.shape_.get(), rshape, rstride, ws_dptr);
}
template
diff --git a/src/operator/tensor/elemwise_binary_broadcast_op-inl.cuh
b/src/operator/tensor/elemwise_binary_broadcast_op-inl.cuh
new file mode 100644
index 000..1998576172b
--- /dev/null
+++ b/src/operator/tensor/elemwise_binary_broadcast_op-inl.cuh
@@ -0,0 +1,64 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing,
+ * software distributed under the License is distributed on an
+ * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+ * KIND, either express or implied. See the License for the
+ * specific language governing permissions and limitations
+ * under the License.
+ */
+
+#ifndef MXNET_OPERATOR_TENSOR_ELEMWISE_BINARY_BROADCAST_OP_CUH_
+#define
[GitHub] haojin2 commented on issue #10274: test_ndarray.test_reduce fails in v1.0.0
haojin2 commented on issue #10274: test_ndarray.test_reduce fails in v1.0.0 URL: https://github.com/apache/incubator-mxnet/issues/10274#issuecomment-404729481 Is this a duplicate issue of #8780? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11710: test_ndarray.test_ndarray_pickle has fixed seed that can mask flakiness
szha opened a new issue #11710: test_ndarray.test_ndarray_pickle has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11710 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11709: test_ndarray.test_ndarray_elementwise has fixed seed that can mask flakiness
szha opened a new issue #11709: test_ndarray.test_ndarray_elementwise has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11709 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11708: test_ndarray.test_cached has fixed seed that can mask flakiness
szha opened a new issue #11708: test_ndarray.test_cached has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11708 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11707: test_nccl.test_nccl_pushpull has fixed seed that can mask flakiness
szha opened a new issue #11707: test_nccl.test_nccl_pushpull has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11707 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] haojin2 commented on issue #11501: Tracking issue for tests with fixed random seed
haojin2 commented on issue #11501: Tracking issue for tests with fixed random seed URL: https://github.com/apache/incubator-mxnet/issues/11501#issuecomment-404729204 @szha So shall I close this issue for now? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11706: test_module.test_monitor has fixed seed that can mask flakiness
szha opened a new issue #11706: test_module.test_monitor has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11706 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11704: test_module.test_factorization_machine_module has fixed seed that can mask flakiness
szha opened a new issue #11704: test_module.test_factorization_machine_module has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11704 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11705: test_module.test_module_set_params has fixed seed that can mask flakiness
szha opened a new issue #11705: test_module.test_module_set_params has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11705 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11703: test_loss.test_triplet_loss has fixed seed that can mask flakiness
szha opened a new issue #11703: test_loss.test_triplet_loss has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11703 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11701: test_loss.test_saveload has fixed seed that can mask flakiness
szha opened a new issue #11701: test_loss.test_saveload has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11701 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11702: test_loss.test_squared_hinge_loss has fixed seed that can mask flakiness
szha opened a new issue #11702: test_loss.test_squared_hinge_loss has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11702 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11699: test_loss.test_l2_loss has fixed seed that can mask flakiness
szha opened a new issue #11699: test_loss.test_l2_loss has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11699 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11698: test_loss.test_l1_loss has fixed seed that can mask flakiness
szha opened a new issue #11698: test_loss.test_l1_loss has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11698 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11700: test_loss.test_sample_weight_loss has fixed seed that can mask flakiness
szha opened a new issue #11700: test_loss.test_sample_weight_loss has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11700 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11697: test_loss.test_kl_loss has fixed seed that can mask flakiness
szha opened a new issue #11697: test_loss.test_kl_loss has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11697 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11695: test_loss.test_hinge_loss has fixed seed that can mask flakiness
szha opened a new issue #11695: test_loss.test_hinge_loss has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11695 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11696: test_loss.test_huber_loss has fixed seed that can mask flakiness
szha opened a new issue #11696: test_loss.test_huber_loss has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11696 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11694: test_loss.test_ctc_loss_train has fixed seed that can mask flakiness
szha opened a new issue #11694: test_loss.test_ctc_loss_train has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11694 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11693: test_loss.test_ctc_loss has fixed seed that can mask flakiness
szha opened a new issue #11693: test_loss.test_ctc_loss has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11693 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11692: test_loss.test_ce_loss has fixed seed that can mask flakiness
szha opened a new issue #11692: test_loss.test_ce_loss has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11692 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] haojin2 commented on issue #8834: Flaky test: Scala:OperatorSuite.arange
haojin2 commented on issue #8834: Flaky test: Scala:OperatorSuite.arange URL: https://github.com/apache/incubator-mxnet/issues/8834#issuecomment-404728554 @marcoabreu If this is no longer happening again would you please close the issue? Thanks! This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11691: test_loss.test_bce_loss has fixed seed that can mask flakiness
szha opened a new issue #11691: test_loss.test_bce_loss has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11691 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11689: test_gluon_data.test_multi_worker has fixed seed that can mask flakiness
szha opened a new issue #11689: test_gluon_data.test_multi_worker has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11689 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] lanking520 commented on issue #8834: Flaky test: Scala:OperatorSuite.arange
lanking520 commented on issue #8834: Flaky test: Scala:OperatorSuite.arange URL: https://github.com/apache/incubator-mxnet/issues/8834#issuecomment-404728275 Yep, @haojin2 kill this one This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11690: test_io.test_ImageRecordIter_seed_augmentation has fixed seed that can mask flakiness
szha opened a new issue #11690: test_io.test_ImageRecordIter_seed_augmentation has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11690 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11688: test_gluon.test_export has fixed seed that can mask flakiness
szha opened a new issue #11688: test_gluon.test_export has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11688 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11687: test_gluon.test_conv has fixed seed that can mask flakiness
szha opened a new issue #11687: test_gluon.test_conv has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11687 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11686: test_executor.test_dot has fixed seed that can mask flakiness
szha opened a new issue #11686: test_executor.test_dot has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11686 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha opened a new issue #11685: test_executor.test_bind has fixed seed that can mask flakiness
szha opened a new issue #11685: test_executor.test_bind has fixed seed that can mask flakiness URL: https://github.com/apache/incubator-mxnet/issues/11685 The unit test in title have been using fixed seed to mask flakiness. Suggested action: 1. Evaluate whether the test is flaky without fixed seed. If not, remove seed. Else move to 2 2. If test is flaky, determine whether it's an actual uncaught edge case. If so, fix the operator. Else move to 3 3. If numerical instability is inevitable, adjust tolerance level appropriately. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] cclauss edited a comment on issue #11669: Fix TabError for Python 3
cclauss edited a comment on issue #11669: Fix TabError for Python 3
URL: https://github.com/apache/incubator-mxnet/pull/11669#issuecomment-404726234
@marcoabreu @ThomasDelteil Can you please review ("Files changed" -->
"Review changes") this change because the goal of this PR is to make a **simple
whitespace-only change** that is easy to review. However, much text has been
written above that will cause reviewers to pause.
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
With regards,
Apache Git Services
[GitHub] szha commented on issue #11501: Tracking issue for tests with fixed random seed
szha commented on issue #11501: Tracking issue for tests with fixed random seed URL: https://github.com/apache/incubator-mxnet/issues/11501#issuecomment-404726406 Thanks for summarizing the issues @haojin2. Fixed seed indeed has the effect of hiding flaky tests so we should treat these as flaky tests until fixed. On the other hand, I think it's probably better to track them as individual items so that people can contribute to the fix one at a time. I will go ahead and split these into individual issues. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] cclauss commented on issue #11669: Fix TabError for Python 3
cclauss commented on issue #11669: Fix TabError for Python 3
URL: https://github.com/apache/incubator-mxnet/pull/11669#issuecomment-404726234
@marcoabreu @ThomasDelteil Can you please review ("Files changed" -->
"Review changes") this change because the goal of this PR was to make a
**simple whitespace-only change** that was easy to review. However, much text
has been written above that will cause reviewers to pause.
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
With regards,
Apache Git Services
[GitHub] haojin2 commented on issue #10387: Flaky test(scala): test_arange
haojin2 commented on issue #10387: Flaky test(scala): test_arange URL: https://github.com/apache/incubator-mxnet/issues/10387#issuecomment-404723360 @marcoabreu Have we observed more occurrence of this? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] haojin2 commented on issue #8834: Flaky test: Scala:OperatorSuite.arange
haojin2 commented on issue #8834: Flaky test: Scala:OperatorSuite.arange URL: https://github.com/apache/incubator-mxnet/issues/8834#issuecomment-404723208 @lanking520 Was this fixed earlier? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] haojin2 commented on issue #11680: Flaky Perl test: test_loss.test_dynamic
haojin2 commented on issue #11680: Flaky Perl test: test_loss.test_dynamic URL: https://github.com/apache/incubator-mxnet/issues/11680#issuecomment-404722419 Thanks a lot for your quick response! This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] szha commented on issue #11684: fix compile error in debug model
szha commented on issue #11684: fix compile error in debug model URL: https://github.com/apache/incubator-mxnet/pull/11684#issuecomment-404719232 @NEWPLAN thanks. What error do you see when compiling in debug mode? What operating system and compiler are you using? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] rahul003 commented on a change in pull request #11633: Fix dist kvstore for trainer and flaky dist kvstore test
rahul003 commented on a change in pull request #11633: Fix dist kvstore for
trainer and flaky dist kvstore test
URL: https://github.com/apache/incubator-mxnet/pull/11633#discussion_r202236977
##
File path: tests/nightly/dist_sync_kvstore.py
##
@@ -398,18 +432,25 @@ def check_trainer_kv_type(stype, grad_stype,
update_on_kv):
parser.add_argument('--no-gpu', dest='gpu', action='store_false')
parser.add_argument('--no-multiprecision', dest='multiprecision',
action='store_false')
opt = parser.parse_args()
-if opt.type == 'gluon':
-test_gluon_trainer()
-if opt.type == 'invalid':
+if opt.type == 'gluon_type_cpu':
+test_gluon_trainer_type()
+exit()
+if opt.type == 'gluon_step_cpu':
+test_gluon_trainer_step()
+exit()
+if opt.type == 'gluon_sparse_step_cpu':
+test_gluon_trainer_sparse_step()
+exit()
+if opt.type == 'invalid_cpu':
test_invalid_operations()
+exit()
if opt.type == 'all' or opt.type == 'init':
test_sync_init(opt.gpu)
+kv = init_kv()
if opt.type == 'all' or opt.type == 'default':
-kv = init_kv()
kv = set_optimizer(use_multiprecision=opt.multiprecision)
test_sync_push_pull(opt.nrepeat)
-# dont run non compressed tests after this as kvstore compression will be
set here
+# don't run non compressed tests after this as kvstore compression will be
set here
if opt.type == 'all' or opt.type == 'compressed':
Review comment:
Because in the current script, it no longer represents all tests.
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
With regards,
Apache Git Services
[GitHub] eric-haibin-lin commented on a change in pull request #11633: Fix dist kvstore for trainer and flaky dist kvstore test
eric-haibin-lin commented on a change in pull request #11633: Fix dist kvstore
for trainer and flaky dist kvstore test
URL: https://github.com/apache/incubator-mxnet/pull/11633#discussion_r202236693
##
File path: tests/nightly/dist_sync_kvstore.py
##
@@ -398,18 +432,25 @@ def check_trainer_kv_type(stype, grad_stype,
update_on_kv):
parser.add_argument('--no-gpu', dest='gpu', action='store_false')
parser.add_argument('--no-multiprecision', dest='multiprecision',
action='store_false')
opt = parser.parse_args()
-if opt.type == 'gluon':
-test_gluon_trainer()
-if opt.type == 'invalid':
+if opt.type == 'gluon_type_cpu':
+test_gluon_trainer_type()
+exit()
+if opt.type == 'gluon_step_cpu':
+test_gluon_trainer_step()
+exit()
+if opt.type == 'gluon_sparse_step_cpu':
+test_gluon_trainer_sparse_step()
+exit()
+if opt.type == 'invalid_cpu':
test_invalid_operations()
+exit()
if opt.type == 'all' or opt.type == 'init':
test_sync_init(opt.gpu)
+kv = init_kv()
if opt.type == 'all' or opt.type == 'default':
-kv = init_kv()
kv = set_optimizer(use_multiprecision=opt.multiprecision)
test_sync_push_pull(opt.nrepeat)
-# dont run non compressed tests after this as kvstore compression will be
set here
+# don't run non compressed tests after this as kvstore compression will be
set here
if opt.type == 'all' or opt.type == 'compressed':
Review comment:
why remove it?
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
With regards,
Apache Git Services
[GitHub] anirudhacharya commented on issue #10686: Building from source fails
anirudhacharya commented on issue #10686: Building from source fails URL: https://github.com/apache/incubator-mxnet/issues/10686#issuecomment-404716668 @crawforc3 please try these updated instructions and reopen if the issue persists. @nswamy please close this issue. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] NEWPLAN opened a new pull request #11684: fix compile error in debug model
NEWPLAN opened a new pull request #11684: fix compile error in debug model URL: https://github.com/apache/incubator-mxnet/pull/11684 if in debug model, the generate file is large than default size, increase the code size limitation ## Description ## This PR is for fixing compiler error in debug model ## Checklist ## ### Essentials ### Please feel free to remove inapplicable items for your PR. - [ ] The PR title starts with [MXNET-$JIRA_ID], where $JIRA_ID refers to the relevant [JIRA issue](https://issues.apache.org/jira/projects/MXNET/issues) created (except PRs with tiny changes) - [ ] Changes are complete (i.e. I finished coding on this PR) - [ ] All changes have test coverage: - Unit tests are added for small changes to verify correctness (e.g. adding a new operator) - Nightly tests are added for complicated/long-running ones (e.g. changing distributed kvstore) - Build tests will be added for build configuration changes (e.g. adding a new build option with NCCL) - [ ] Code is well-documented: - For user-facing API changes, API doc string has been updated. - For new C++ functions in header files, their functionalities and arguments are documented. - For new examples, README.md is added to explain the what the example does, the source of the dataset, expected performance on test set and reference to the original paper if applicable - Check the API doc at http://mxnet-ci-doc.s3-accelerate.dualstack.amazonaws.com/PR-$PR_ID/$BUILD_ID/index.html - [ ] To the my best knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change ### Changes ### - [ ] Feature1, tests, (and when applicable, API doc) - [ ] Feature2, tests, (and when applicable, API doc) ## Comments ## - If this change is a backward incompatible change, why must this change be made. - Interesting edge cases to note here This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] sergeykolychev commented on issue #11680: Flaky Perl test: test_loss.test_dynamic
sergeykolychev commented on issue #11680: Flaky Perl test: test_loss.test_dynamic URL: https://github.com/apache/incubator-mxnet/issues/11680#issuecomment-404706340 @haojin2 Should be fixed by https://github.com/apache/incubator-mxnet/pull/11642/commits/ef96ad3c11abed719951ba70833cb273c47104fa I plan to merge in parent PR tomorrow. Thanks. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[incubator-mxnet] branch master updated: Update README.md (#11677)
This is an automated email from the ASF dual-hosted git repository. anirudh2290 pushed a commit to branch master in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git The following commit(s) were added to refs/heads/master by this push: new 92366a9 Update README.md (#11677) 92366a9 is described below commit 92366a99b15395583ec2991e722c679da1e281c5 Author: Kellen Sunderland AuthorDate: Fri Jul 13 04:25:18 2018 +0200 Update README.md (#11677) --- contrib/clojure-package/README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/contrib/clojure-package/README.md b/contrib/clojure-package/README.md index 430be4d..5e7356c 100644 --- a/contrib/clojure-package/README.md +++ b/contrib/clojure-package/README.md @@ -193,7 +193,7 @@ This command can be very handy too timestamp, name, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB]` **Supported APIs** -There are 3 high level apis supported in MxNet: (Model/FeedForward), Module, and Gluon. The Module api is supported in the Clojure package because of the existing support for it in the Scala package. The Module api is very similar to the Gluon api and examples of the usage can be found in the examples directory. The Model/FeedForward Api is deprected. +There are 3 high level apis supported in MXNet: (Model/FeedForward), Module, and Gluon. The Module api is supported in the Clojure package because of the existing support for it in the Scala package. The Module api is very similar to the Gluon api and examples of the usage can be found in the examples directory. The Model/FeedForward Api is deprected. Gluon support will come later and may or may not be built on the Scala gluon api (when it lands there)
[GitHub] anirudh2290 closed pull request #11677: Update README.md to fix MXNet capitalization
anirudh2290 closed pull request #11677: Update README.md to fix MXNet capitalization URL: https://github.com/apache/incubator-mxnet/pull/11677 This is a PR merged from a forked repository. As GitHub hides the original diff on merge, it is displayed below for the sake of provenance: As this is a foreign pull request (from a fork), the diff is supplied below (as it won't show otherwise due to GitHub magic): diff --git a/contrib/clojure-package/README.md b/contrib/clojure-package/README.md index 430be4d6a84..5e7356caf64 100644 --- a/contrib/clojure-package/README.md +++ b/contrib/clojure-package/README.md @@ -193,7 +193,7 @@ This command can be very handy too timestamp, name, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB]` **Supported APIs** -There are 3 high level apis supported in MxNet: (Model/FeedForward), Module, and Gluon. The Module api is supported in the Clojure package because of the existing support for it in the Scala package. The Module api is very similar to the Gluon api and examples of the usage can be found in the examples directory. The Model/FeedForward Api is deprected. +There are 3 high level apis supported in MXNet: (Model/FeedForward), Module, and Gluon. The Module api is supported in the Clojure package because of the existing support for it in the Scala package. The Module api is very similar to the Gluon api and examples of the usage can be found in the examples directory. The Model/FeedForward Api is deprected. Gluon support will come later and may or may not be built on the Scala gluon api (when it lands there) This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] sergeykolychev commented on issue #11680: Flaky Perl test: test_loss.test_dynamic
sergeykolychev commented on issue #11680: Flaky Perl test: test_loss.test_dynamic URL: https://github.com/apache/incubator-mxnet/issues/11680#issuecomment-404704376 @haojin2 I'll take a look ad will fix. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] yukang2017 opened a new issue #11683: time cost in grad clipping
yukang2017 opened a new issue #11683: time cost in grad clipping
URL: https://github.com/apache/incubator-mxnet/issues/11683
I compare the speed of grad clipping between pytorch(0.3.1) and
mxnet-gluon(1.2.0) on a Titan X. Pytorch gets nearly 10 times faster than
mxnet-gluon.
(The network I use is [nasnet](https://arxiv.org/abs/1707.07012).)
mxnet:
```
grads = [i.grad(ctx) for i in model.collect_params().values() if i._grad is
not None]
gluon.utils.clip_global_norm(grads, args.grad_clip)
```
pytorch:
`nn.utils.clip_grad_norm(model.parameters(), args.grad_clip)
`
However, when I test 'clip_gradient' in mxnet.gluon.Trainer and torch.optim,
following thomelane's reply in the
[issue](https://github.com/apache/incubator-mxnet/issues/11508#issuecomment-404675656).
The time cost is similar.
```
mxnet.gluon.Trainer(net.collect_params(), optimizer='sgd',
optimizer_params={'learning_rate': 0.1,
'clip_gradient':5},
kvstore='device')
```
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
With regards,
Apache Git Services
[GitHub] yukang2017 commented on issue #11508: 【Question】Are there any examples for gradients clipping in gluon?
yukang2017 commented on issue #11508: 【Question】Are there any examples for gradients clipping in gluon? URL: https://github.com/apache/incubator-mxnet/issues/11508#issuecomment-404702227 @thomelane I will open a new issue. Thank you! I tested 'clip_gradient' in _mxnet.gluon.Trainer_ and torch.optim. The time cost is similar. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] lynneChan commented on issue #11137: memory cost problem for different label width when using ssd example
lynneChan commented on issue #11137: memory cost problem for different label width when using ssd example URL: https://github.com/apache/incubator-mxnet/issues/11137#issuecomment-404701765 Hi @thomelane, thanks for your replay. I was using the SSD example code to training a face detector, the WIDER dataset have about 12880 images, so the `batch_size` is fine. And the number of classes is 2(face or non-face). The code used `DetRecordIter` in `dataset/iterator.py` to get image batches which called another API `mx.io.ImageDetRecordIter`. This API need a `label_width` parameter. I keeped all the settings, and just change the `label_width` from 602 to 4784, and the GPU memory cost changes from 6991MB to 10367MB. That's pretty strange, it should not increase so much. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] Feywell commented on issue #9186: How to train model with multi machines
Feywell commented on issue #9186: How to train model with multi machines URL: https://github.com/apache/incubator-mxnet/issues/9186#issuecomment-404700463 @thomelane Thank you！ I will read it carefully. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ijkguo commented on issue #8474: How to train variable input shape CNN?
ijkguo commented on issue #8474: How to train variable input shape CNN? URL: https://github.com/apache/incubator-mxnet/issues/8474#issuecomment-404698670 Module API had been updated to support variable shape training, c.f. https://github.com/apache/incubator-mxnet/pull/6805 before this issue. MutableModule from rcnn example was introduced before the previous PR to avoid breaking module API. It was unfortunate that the example was not updated timely since the PR. But if still interested, check out https://github.com/apache/incubator-mxnet/blob/master/example/rcnn/train.py#L117. As for a special tutorial on variable shape training, you don't need one. Use `module.fit` or Gluon just as if it was fixed shape will work. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ijkguo commented on issue #11240: rcnn example throws CUDNN_STATUS_BAD_PARAM when running under cudnn 6.0
ijkguo commented on issue #11240: rcnn example throws CUDNN_STATUS_BAD_PARAM when running under cudnn 6.0 URL: https://github.com/apache/incubator-mxnet/issues/11240#issuecomment-404695463 It was probably related to old SoftmaxActivation layer. Now changed to `mx.sym.softmax` in https://github.com/apache/incubator-mxnet/pull/11373. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] zhreshold commented on issue #11643: Added the diag() operator
zhreshold commented on issue #11643: Added the diag() operator URL: https://github.com/apache/incubator-mxnet/pull/11643#issuecomment-404694877 @ifeherva Try locally simulate the CI with the docker build script. Currently we can get nothing from the log. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] solin319 commented on issue #11165: ImageIter much slower than ImageRecordIter
solin319 commented on issue #11165: ImageIter much slower than ImageRecordIter URL: https://github.com/apache/incubator-mxnet/issues/11165#issuecomment-404694859 We move some augmenters call in c++, and the speed can achieve 0.21s per batch, which is 0.35 smentioned above. So we expect community can add an interface can read raw pic with high speed. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] ifeherva commented on issue #11643: Added the diag() operator
ifeherva commented on issue #11643: Added the diag() operator URL: https://github.com/apache/incubator-mxnet/pull/11643#issuecomment-404694302 Any idea why it does not pass CI? This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[incubator-mxnet] branch master updated: xrange(10) --> range(10) for Python 3 (#11673)
This is an automated email from the ASF dual-hosted git repository.
haibin pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git
The following commit(s) were added to refs/heads/master by this push:
new 6b2cbd4 xrange(10) --> range(10) for Python 3 (#11673)
6b2cbd4 is described below
commit 6b2cbd4ce8fd5088bf4d6c8c6357b3e3305378d6
Author: cclauss
AuthorDate: Fri Jul 13 02:57:46 2018 +0200
xrange(10) --> range(10) for Python 3 (#11673)
__xrange()__ was removed in Python 3 in favor of __range()__. For such a
small number as 10 there is no difference between __xrange()__ and __range()__
in Python 2.
---
benchmark/python/sparse/util.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/benchmark/python/sparse/util.py b/benchmark/python/sparse/util.py
index c20b33a..29000bd 100644
--- a/benchmark/python/sparse/util.py
+++ b/benchmark/python/sparse/util.py
@@ -24,7 +24,7 @@ def estimate_density(DATA_PATH, feature_size):
raise Exception("Data is not there!")
density = []
P = 0.01
-for _ in xrange(10):
+for _ in range(10):
num_non_zero = 0
num_sample = 0
with open(DATA_PATH) as f:
[GitHub] eric-haibin-lin closed pull request #11673: xrange(10) --> range(10) for Python 3
eric-haibin-lin closed pull request #11673: xrange(10) --> range(10) for Python
3
URL: https://github.com/apache/incubator-mxnet/pull/11673
This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:
As this is a foreign pull request (from a fork), the diff is supplied
below (as it won't show otherwise due to GitHub magic):
diff --git a/benchmark/python/sparse/util.py b/benchmark/python/sparse/util.py
index c20b33a86d6..29000bdf82e 100644
--- a/benchmark/python/sparse/util.py
+++ b/benchmark/python/sparse/util.py
@@ -24,7 +24,7 @@ def estimate_density(DATA_PATH, feature_size):
raise Exception("Data is not there!")
density = []
P = 0.01
-for _ in xrange(10):
+for _ in range(10):
num_non_zero = 0
num_sample = 0
with open(DATA_PATH) as f:
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
With regards,
Apache Git Services
[GitHub] zhreshold closed issue #9823: RCNN example fails for using latest mxnet
zhreshold closed issue #9823: RCNN example fails for using latest mxnet URL: https://github.com/apache/incubator-mxnet/issues/9823 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] zhreshold closed issue #8694: Error when running "/rcnn/script/additional_deps.sh"
zhreshold closed issue #8694: Error when running "/rcnn/script/additional_deps.sh" URL: https://github.com/apache/incubator-mxnet/issues/8694 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] zhreshold closed issue #10019: ModuleNotFoundError
zhreshold closed issue #10019: ModuleNotFoundError URL: https://github.com/apache/incubator-mxnet/issues/10019 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] zhreshold commented on issue #9823: RCNN example fails for using latest mxnet
zhreshold commented on issue #9823: RCNN example fails for using latest mxnet URL: https://github.com/apache/incubator-mxnet/issues/9823#issuecomment-404693226 Closing this after merging #11373, feel free to ping me to reopen it it's not fixed. This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] zhreshold commented on issue #8694: Error when running "/rcnn/script/additional_deps.sh"
zhreshold commented on issue #8694: Error when running "/rcnn/script/additional_deps.sh" URL: https://github.com/apache/incubator-mxnet/issues/8694#issuecomment-404693156 Closing this after merging #11373 This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] zhreshold commented on issue #10291: some code in example/rcnn/demo.py is wrong.
zhreshold commented on issue #10291: some code in example/rcnn/demo.py is wrong. URL: https://github.com/apache/incubator-mxnet/issues/10291#issuecomment-404693102 Closing this after #11373 is merged This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
