I'm trying to compile your ecrec branch following your instructions, but my compiler reports some bugs, for example:
/x-deeplearning-redundancy/xdl/ps-plus/ps-plus/server/udf/momentum_server_updater.cc: In member function 'virtual ps::Status ps::server::udf::MomentumServerUpdater::SimpleRun(ps::server::UdfContext*, const std::vectorps::server::Slices&, const std::vectorps::Tensor&, const std::vector&, const std::vector&, const std::vector&) const':
/x-deeplearning-redundancy/xdl/ps-plus/ps-plus/server/udf/momentum_server_updater.cc:212:24: error: 'parity_ids' was not declared in this scope
for (auto pids : parity_ids) {
I want to learn more about your paper "Eficient Fault Tolerance for Recommendation Model Training via Erasure Coding" and strictly follow your set-up instructions. If I roughly checkout previous commit id, such as 04326a7, it will compile smoothly though I don't think the error file is perfectly fixed and I haven't moved forward to make sure it functions well. All things running on the docker you provided in your instruction so I have no idea what the problem is. Would you please have a look and double check the instructions? Thank you so much!
I'm trying to compile your ecrec branch following your instructions, but my compiler reports some bugs, for example:
/x-deeplearning-redundancy/xdl/ps-plus/ps-plus/server/udf/momentum_server_updater.cc: In member function 'virtual ps::Status ps::server::udf::MomentumServerUpdater::SimpleRun(ps::server::UdfContext*, const std::vectorps::server::Slices&, const std::vectorps::Tensor&, const std::vector&, const std::vector&, const std::vector&) const':
/x-deeplearning-redundancy/xdl/ps-plus/ps-plus/server/udf/momentum_server_updater.cc:212:24: error: 'parity_ids' was not declared in this scope
for (auto pids : parity_ids) {
I want to learn more about your paper "Eficient Fault Tolerance for Recommendation Model Training via Erasure Coding" and strictly follow your set-up instructions. If I roughly checkout previous commit id, such as 04326a7, it will compile smoothly though I don't think the error file is perfectly fixed and I haven't moved forward to make sure it functions well. All things running on the docker you provided in your instruction so I have no idea what the problem is. Would you please have a look and double check the instructions? Thank you so much!