[WIP] Sparse Tensor #5800

eric-haibin-lin · 2017-04-12T18:36:41Z

Please don't merge in. This PR includes some changes to enable imperative execution for sparse ndarrays for a simple elemwise_add operator on cpu.

nnvm changes at dmlc/nnvm#110

@piiswrong @mli @madjam @reminisce

piiswrong · 2017-04-12T19:31:35Z

include/mxnet/ndarray.h

    Storage::Handle shandle;
+    /*! \brief storage handles for aux data (e.g index) */
+    std::vector<Storage::Handle> aux_handles;


vector is probably overkill

CSR needs 2 aux data. I can make it an array of size 2

piiswrong · 2017-04-12T19:31:54Z

include/mxnet/ndarray.h

    /*! \brief variable from engine */
    Engine::VarHandle var;
+    std::vector<Engine::VarHandle> aux_vars;


no need for additional var

piiswrong · 2017-04-12T19:32:28Z

include/mxnet/op_attr_types.h

+ * TODO should probably change const std::vector<NDArray>& outputs to
+                               std::vector<NDArray> *outputs
+ */
+using FComputeND = std::function<void (const nnvm::NodeAttrs& attrs,


we can probably just call it FComputeEx

piiswrong · 2017-04-12T19:33:06Z

python/mxnet/ndarray.py

@@ -560,24 +559,6 @@ def wait_to_read(self):
        check_call(_LIB.MXNDArrayWaitToRead(self.handle))


-    @property
-    def shape(self):


I plan to move all shared properties/method to NDArrayBase.

can SparseNDArray inherit NDArray instead of NDArrayBase?

That means for any new function added to NDArray, we need to either implement it for SparseNDArray or disable it in SparseNDArray. Otherwise it confuses user with an inherited method which doesn't work

any such examples? NDArray interface has been fairly stable.
The problem with SparseNDArray not being a NDArray is all the places where you use isinstance(x, NDArray) need to be changed to isinstance(x, NDArrayBase)

If the interface is stable, that's fine. I just want to make sure user doesn't make the wrong function call from SparseND which is directed to NDArray and doesn't work

reminisce · 2017-04-13T05:43:15Z

include/mxnet/ndarray.h

@@ -67,20 +91,66 @@ class NDArray {
      Mkl_mem_ = std::make_shared<MKLMemHolder>();
 #endif
  }
+  NDArray(NDArray data, std::vector<NDArray> aux_data, Context ctx, NDArrayChunkType chunk_type, const TShape &shape)


Why is aux_data passed by value?

It doesn't matter. NDArray has the pointer to the chunk

The point is to avoid copying entire vector.

reminisce · 2017-04-13T05:49:21Z

include/mxnet/ndarray.h

+        aux_handles.push_back(Storage::Get()->Alloc(aux_bytes, ctx));
+        delay_alloc = false;
+        // Initialize aux_shape and shape
+        this->aux_shapes = aux_shapes;


this->aux_shapes = std::move(aux_shapes)?

To save memory ops?

Same as the above comment, avoid copying the entire vector.

reminisce · 2017-04-18T06:25:48Z

src/operator/tensor/elemwise_binary_op.h

+  mshadow::Stream<xpu> *s = ctx.get_stream<xpu>();
+  auto indices_l = nd_l.aux_data(0).FlatTo1D<xpu, ROW_SPARSE_TYPE>(s);
+  auto indices_r = nd_r.aux_data(0).FlatTo1D<xpu, ROW_SPARSE_TYPE>(s);
+  auto indices_out = output.aux_data(0).FlatTo1D<xpu, ROW_SPARSE_TYPE>(s);


A general comment is that since this implements row sparse format of tensors, can we have meaningful interfaces to get the number of rows and row index array from a NDArray? Currently, these two attributes are retrieved from NDArray::aux_shape(0)[0] and NDArray::aux_data(0).FlatTo1D<xpu, ROW_SPARSE_TYPE>(s), which seems not very natural. I would have to look into the detailed implementation of Chunk::CheckAndAlloc to understand the meaning of these.

Good point. I'll add some interface in NDArray to abstract this out

piiswrong · 2017-05-14T19:42:38Z

please run git rebase upstream/master to clean up diff

merge with 38f7c55 compiles on GPU update check alloc: Checkpoint. Pass elem-sum gpu test bug fix for copyfromto. sparse sgd test pass on gpu inefficient implementation for csr copy update submodule fix lint Simple bind with infer storage type (#32) * Symbol binding for sparse tensor development. (#31) * Initial checkin * Add init functions for simple bind in graph_executor * Add simple_bind c_api * Add simple bind c-api * Assign zeros to in_args, arg_grads, and aux_states * Add simple_bind2 python interface * Fix python interface bugs * Interface changes * Fix * Fix core dump * Add bind_ith_exec c_api * Change simple_bind2 * Fix seg fault * Finish simple_bind * Change _bind_ith_exec * Refactor simple_bind initialization flow for bind * Consolidate bind and simple_bind graph init flow * Fix bug * Clean up * Add comments * Clean up * Clean up * Minor correction * Rename APIs in graph executor * Refactor * Rebase * Delete deprecated functions * Move more front-end work to backend * Bug fix * Fix failed tests * Minor fix * Fix lint * Fix lint * Revert unnecessary changes * Revert * Revert * Clean up * Fix lint Conflicts: python/mxnet/symbol.py src/executor/graph_executor.cc * Add inferstorage to graph executor * re-enable tests for sparse embedding with simple_bind * type switch fix in sparse embedding" ; change `default` to `default_storage` for cast storage op (#33) * change default to default_storage * disable cpp test build temporarily attempt to fix windows build error, and fix lint (#34) update nnvm submodule (#37) Scipy build (#38) * update nnvm submodule * add scipy pip install for dockerfile Python3 unit tests (#39) * change xrange to range for python3 compatiblity" * remove more xrange from tests replace long with int for python3 (#40) fix the rest of TShape constructor errors (#41) fix lint (#42) fix wrong usage of mshadow::Shape1" (#43) implementation for Csr slice on cpu (#36) * CPU implementation for CSR remove seg_len from csr slice add some docs for slice csr change indptr, values, etc to be private member bug fix in sparse embedding update nnvm submoduel fix lint update unit test for sparse nd" * add const for SliceCsrIndPtr kernel Fix sparse dot according to the new RSP definition (#35) * Fix csr dot dns * Fix sparse dot * Add fallback and test cases for dot(csr, dns)=dns * Add int type switch * Fix * Fix * Fix update mshadow submodule (#44) Fix dns to rsp (#46) fix lint (#47) add runtime storage fallback detection" (#48) * add runtime storage fallback detection" * replace cast storage ex with cast storage impl Fm example (#45) * update csr slice logic to avoid confusion. add more exmaples. * add hint to module.update * more testcases(fallback) for sparse_nd * add to_csr() and to_rsp() method. More unit test (fallback now) * add fm test. fix lint * register sparse sgd under Optim.SGD * update dmlc-core submoduel * change indptr to _indptr temporarily. add const ref to fname fix lint fix lint; (#51) Guard gpu cast storage (#50) * Clean up * Fix typo Rearrange unit test files (#52) fix lint. add scipy for python_test. fix scipy.sparse import error. fix truediv for python3 fix travis test (#54) * remove pyc files * add verbose for travis nosetests cleanup some testing code and enums (#57) * update Makefile * refactor test_sparse_operator * change `default_storage` back to `default` * remove unused cpp tests port libsvm parser to mxnet as libsvm iter (#55) * copied csv iter to libsvm iter test libsvm iter draft handle round batch == false for csr batch loader code refactoring add get stype, shape interface to iiter separate class for sparse iter add missing file fix mem corruption' rename variables add comments also read label from libsvm add test. update docs. update submodule Conflicts: python/mxnet/sparse_ndarray.py * update submodule * fix lint * update test * revert naming change add benchmark scritp for dot (#59) * add benchmark scritp for dot add gpu option for bench add get_data funciton for benchmark print t_sparse, too; add comment change nnz to dnesity add backward * add comment update fm test (#62) introduce CSRNDarray and rowsparseNDarray to python frontend api (#58) * introduce CSRNDarray and rowsparseNDarray to python frontend api * temporarily disable fm_module test fix lint (#64) fix typo. disable libsvm io test (#65) Improve dot (#61) * Init checkin * Fix * Adjust dot parallelization methods * Set num_omp_threads for benchmark from command line * Fix omp thread number * Clean up * Add scipy as dot baseline * Fix format sparse_retain op (#66) * Initial checkin * Fix bugs * Add unit test for sparse_retain * Add example and modify test add storage cast for outputs that have non-default storage (#67) fix gpu build (#69) Fix test_sparse_retain python3 issue (#68) revert nnvm version

support sgd(rsp, rsp) support dot(csr, rsp) when rsp is full add ref to const ndarray params support sparse embedding with rsp weight' fix lint modify embedding backward to produce dense grad remove invalid_rid for rsp->dns remove previous embedding op changes pass sparse embedding test add STORAGE_TYPE_ASSIGN_CHECK remove backward storage infer

Conflicts: python/mxnet/test_utils.py tests/python/unittest/test_sparse_operator.py update mshadow submodule fix extra quotes in test script change indptr type to int64 better err message for rsp"

eric-haibin-lin

Added a few TODOs for the next commit

eric-haibin-lin · 2017-06-22T01:33:45Z

include/mxnet/c_api.h

+ * \return 0 when success, -1 when failure happens
+ */
+MXNET_DLL int MXNDArrayCreateSparseEx(int storage_type,
+                    const mx_uint *shape,


eric-haibin-lin · 2017-06-22T01:34:16Z

include/mxnet/c_api.h

+/*!
+ * \brief infer storage type of unknown input types given the known one.
+ */
+MXNET_DLL int MXSymbolInferStorageType(SymbolHandle sym,


This API could probably be removed

eric-haibin-lin · 2017-06-22T01:34:45Z

include/mxnet/ndarray.h

+namespace ndarray {
+template<typename from_xpu, typename to_xpu>
+void Copy(const TBlob &from, TBlob *to, Context from_ctx, Context to_ctx, RunContext ctx);
+};


No need for this forward declaration

eric-haibin-lin · 2017-06-22T01:36:03Z

include/mxnet/op_attr_types.h

+                                     const OpContext& ctx,
+                                     const std::vector<NDArray>& inputs,
+                                     const std::vector<OpReqType>& req,
+                                     const std::vector<NDArray>& outputs)>;


indentation

eric-haibin-lin · 2017-06-22T01:36:49Z

python/mxnet/kvstore.py

@@ -48,7 +48,7 @@ def updater_handle(key, lhs_handle, rhs_handle, _):

 class KVStore(object):
    """A key-value store for synchronization of values, over multiple devices."""
-    def __init__(self, handle):
+    def __init__(self, handle, name2idx=None):


eric-haibin-lin · 2017-06-22T01:40:48Z

python/mxnet/test_utils.py

@@ -657,14 +715,29 @@ def check_symbolic_backward(sym, location, out_grads, expected, rtol=1e-5, atol=
    if isinstance(expected, (list, tuple)):
        expected = {k:v for k, v in zip(sym.list_arguments(), expected)}
    args_grad_npy = {k:_rng.normal(size=v.shape) for k, v in expected.items()}
-    args_grad_data = {k: mx.nd.array(v, ctx=ctx) for k, v in args_grad_npy.items()}
+    # args_grad_data should be casted to storage type if hinted
+    # TODO(haibin) this is a temporary solution for testing. remove later


verify if it's still needed

eric-haibin-lin · 2017-06-22T01:40:59Z

src/c_api/c_api.cc

@@ -154,6 +154,39 @@ int MXNDArrayCreateEx(const mx_uint *shape,
  API_END();
 }

+int MXNDArrayCreateSparseEx(int storage_type,
+                    const mx_uint *shape,


indentation

eric-haibin-lin · 2017-06-22T01:42:29Z

src/executor/attach_op_execs_pass.cc

 };

 // backward executor
 class BackwardOpExecutor : public OpExecutor {
 public:
-  void Run(RunContext rctx) override {
+  void Run(RunContext rctx, bool is_gpu) override {
+    // TODO(haibin) support storage fallback for BackwardOpExecutor
    op_ctx.run_ctx = rctx;


support cast storage for BackwardOpExecutor

eric-haibin-lin · 2017-06-22T01:43:47Z

src/kvstore/comm.h

@@ -64,11 +68,18 @@ class CommCPU : public Comm {
  CommCPU() {
    nthread_reduction_ = dmlc::GetEnv("MXNET_KVSTORE_REDUCTION_NTHREADS", 4);
    bigarray_bound_ = dmlc::GetEnv("MXNET_KVSTORE_BIGARRAY_BOUND", 1000 * 1000);
+    // TODO(junwu) delete the following data member, now for benchmark only
+    is_serial_push_ = dmlc::GetEnv("MXNET_KVSTORE_SERIAL_PUSH", 0);


remove after the benchmark

eric-haibin-lin · 2017-06-22T01:44:19Z

src/ndarray/ndarray.cc

-      op, attrs, &inputs, &outputs);
-    return outputs[0];
-  } else {
+  CHECK_NE(storage_type(), kUndefinedStorage);


remove shared memory implementation

eric-haibin-lin · 2017-06-22T16:25:21Z

A list of frontend changes in this PR

Note: rsp is short for row_sparse

Frontend changes:
NDArrayBase
-> NDArray
-> SparseNDArray
-> CSRNDArray / RowSparseNDArray
All NDArray(NDArrayHandle(handle)) is replaced by _ndarray_cls(NDArrayHandle(handle)) to construct NDArray/CSRNDArray/RowSparseNDArray based on storage type
Added to_rsp, to_csr and to_dense method for NDArray storage type conversion
Implemented zeros, load, save for SparseNDArray, functions are moved to ndarray_utils
SparseNDArray only supports [:] for assignment
support __getitem__ with rows for CSRNDArray, but not for RowSparseNDArray
__iadd__, __imul__, __idiv___, __isub__ is not supported for SparseNDArray. __add__, __mul__, etc are inherited from NDArray
at, reshape, broadcast_to, T not supported for SparseNDArray yet
user is able to retrieve aux (dense) NDArray via indices, values and indptr properties (indptr is only available for CSRNArray)
Variable has one more attr called __storage_type__
List of operators implemented (CPU only - many of them shares templated implementation but don't have great performance on GPU):
elemwise_add (row_sparse)
cast_storage (csr<->dense, rsp<->dense, forward only)
dot (csr dot dense, csr.T dot dense, csr dot row_sparse, csr.T dot row_sparse)
sparse_embedding (does the same thing as dot, expect weight is auto-allocated)
sparse_retain (see https://github.com/dmlc/mxnet/pull/5800/files#diff-1b545c728261fd4bf9d82d1340ea7a63R275)
sgd_update & sgd_mom_update (only updating weights with non-zeros rows in gradient, when either weight or gradient is row_sparse)
LibSVMIter reads libsvm format input and output CSRNDArray
NDArrayIter can accept CSRNDArray as input
kvstore is able to initialize row_sparse weight

TODO list

kvstore API needs to adopt changes in support str key type in kvstore #6765
remove infer_storage_type python interface
Add test for multi-dimensional RowSparse NDArray
remove shared memory _slice for SparseNDArray, always perform slice and assign
merge unary & binary operators from Chris's branch
bug fix for NDArrayIter and LibSVMIter (NDArrayIter doesn't handle CSRNDArray input well eric-haibin-lin/mxnet#110 libsvm iter bug eric-haibin-lin/mxnet#98)
Plus the ones I mentioned in the previous review
refactor ReduceSumCPUExSerial and ReduceSumCPUExparallel

eric-haibin-lin · 2017-06-22T22:43:14Z

* Refactor cast_storage * Add cast_storage cc and cu files * Remove redundant comments * Replace std::accumulate with ParallelAccumulate * Clean up * Fix windows build

* update kvstore unit test Conflicts: tests/python/unittest/test_kvstore.py update model/module.py Conflicts: python/mxnet/model.py python/mxnet/module/module.py fix lint resolve conflict remove int keys in kvstore update cast to str function * fix failed dist_sync_kv test * bug fix in comm to ensure merged gradient is of the right type bug fix in comm * row sparse dist kvstore draft (push only) row_sparse pull * add ndarray row sparse shared mem constructor * code refactoring * add test for row_sparse weight bug fix for kv server slicing add async support rsolve race condition in kvstore * resolve error after reb ase

* _to_rsp * _to_csr. raise NotImplementedError * todense

* squash merge with 38f7c55 compiles on GPU update check alloc: Checkpoint. Pass elem-sum gpu test bug fix for copyfromto. sparse sgd test pass on gpu inefficient implementation for csr copy update submodule fix lint Simple bind with infer storage type (#32) * Symbol binding for sparse tensor development. (#31) * Initial checkin * Add init functions for simple bind in graph_executor * Add simple_bind c_api * Add simple bind c-api * Assign zeros to in_args, arg_grads, and aux_states * Add simple_bind2 python interface * Fix python interface bugs * Interface changes * Fix * Fix core dump * Add bind_ith_exec c_api * Change simple_bind2 * Fix seg fault * Finish simple_bind * Change _bind_ith_exec * Refactor simple_bind initialization flow for bind * Consolidate bind and simple_bind graph init flow * Fix bug * Clean up * Add comments * Clean up * Clean up * Minor correction * Rename APIs in graph executor * Refactor * Rebase * Delete deprecated functions * Move more front-end work to backend * Bug fix * Fix failed tests * Minor fix * Fix lint * Fix lint * Revert unnecessary changes * Revert * Revert * Clean up * Fix lint Conflicts: python/mxnet/symbol.py src/executor/graph_executor.cc * Add inferstorage to graph executor * re-enable tests for sparse embedding with simple_bind * type switch fix in sparse embedding" ; change `default` to `default_storage` for cast storage op (#33) * change default to default_storage * disable cpp test build temporarily attempt to fix windows build error, and fix lint (#34) update nnvm submodule (#37) Scipy build (#38) * update nnvm submodule * add scipy pip install for dockerfile Python3 unit tests (#39) * change xrange to range for python3 compatiblity" * remove more xrange from tests replace long with int for python3 (#40) fix the rest of TShape constructor errors (#41) fix lint (#42) fix wrong usage of mshadow::Shape1" (#43) implementation for Csr slice on cpu (#36) * CPU implementation for CSR remove seg_len from csr slice add some docs for slice csr change indptr, values, etc to be private member bug fix in sparse embedding update nnvm submoduel fix lint update unit test for sparse nd" * add const for SliceCsrIndPtr kernel Fix sparse dot according to the new RSP definition (#35) * Fix csr dot dns * Fix sparse dot * Add fallback and test cases for dot(csr, dns)=dns * Add int type switch * Fix * Fix * Fix update mshadow submodule (#44) Fix dns to rsp (#46) fix lint (#47) add runtime storage fallback detection" (#48) * add runtime storage fallback detection" * replace cast storage ex with cast storage impl Fm example (#45) * update csr slice logic to avoid confusion. add more exmaples. * add hint to module.update * more testcases(fallback) for sparse_nd * add to_csr() and to_rsp() method. More unit test (fallback now) * add fm test. fix lint * register sparse sgd under Optim.SGD * update dmlc-core submoduel * change indptr to _indptr temporarily. add const ref to fname fix lint fix lint; (#51) Guard gpu cast storage (#50) * Clean up * Fix typo Rearrange unit test files (#52) fix lint. add scipy for python_test. fix scipy.sparse import error. fix truediv for python3 fix travis test (#54) * remove pyc files * add verbose for travis nosetests cleanup some testing code and enums (#57) * update Makefile * refactor test_sparse_operator * change `default_storage` back to `default` * remove unused cpp tests port libsvm parser to mxnet as libsvm iter (#55) * copied csv iter to libsvm iter test libsvm iter draft handle round batch == false for csr batch loader code refactoring add get stype, shape interface to iiter separate class for sparse iter add missing file fix mem corruption' rename variables add comments also read label from libsvm add test. update docs. update submodule Conflicts: python/mxnet/sparse_ndarray.py * update submodule * fix lint * update test * revert naming change add benchmark scritp for dot (#59) * add benchmark scritp for dot add gpu option for bench add get_data funciton for benchmark print t_sparse, too; add comment change nnz to dnesity add backward * add comment update fm test (#62) introduce CSRNDarray and rowsparseNDarray to python frontend api (#58) * introduce CSRNDarray and rowsparseNDarray to python frontend api * temporarily disable fm_module test fix lint (#64) fix typo. disable libsvm io test (#65) Improve dot (#61) * Init checkin * Fix * Adjust dot parallelization methods * Set num_omp_threads for benchmark from command line * Fix omp thread number * Clean up * Add scipy as dot baseline * Fix format sparse_retain op (#66) * Initial checkin * Fix bugs * Add unit test for sparse_retain * Add example and modify test add storage cast for outputs that have non-default storage (#67) fix gpu build (#69) Fix test_sparse_retain python3 issue (#68) revert nnvm version * draft for sgd rsp rsp (#75) support sgd(rsp, rsp) support dot(csr, rsp) when rsp is full add ref to const ndarray params support sparse embedding with rsp weight' fix lint modify embedding backward to produce dense grad remove invalid_rid for rsp->dns remove previous embedding op changes pass sparse embedding test add STORAGE_TYPE_ASSIGN_CHECK remove backward storage infer * fix lint (#78) * fix lint (#79) * serial elemwise sum impl (#80) update module kvstore interface add other missing params and functions revert some interface changes revert some more changes reomve explicit casting for gradients on kvstore update Comm interface update fm example Conflicts: python/mxnet/model.py python/mxnet/ndarray.py * bug fix for initializing module with row_sparse weight (#81) * bug fix for initializing module with row_sparse weight * update log message * Sparse ndarray serialization and deserialization (#77) * Initial checkin * Add unit tests * Fix lint * Fix lint (#84) * Sgd with row_sparse weight, dns gradient (#83) * sgd rsp dns draft * support sgd_mom(rsp, dns, rsp) * update doc * remove cast storage for kv updater * code refactoring * update mshadow version (#88) * csr slice bug fix (#90) * benchmark dot code refactor (#87) * q^x6x add some code in benchmark * refactor * minor fixes * fix * lint fix * Add unit test (#91) * add unittest * minor fix * remove commented lines * change test func name * add test rsp * kvstore push row sparse (#93) * Add multi-thread cpu elemwise sum for rsps * Minor fix * Add flag to switch between serial and multi-thread kvstore push * Fix lint in sparse_ndarray.py * Revert "Fix lint in sparse_ndarray.py" This reverts commit d7225ec. * Fix ndarray init in copy(ctx) * Add env var to control the flow of serial/parallel reduce * Refactor * Fix copy ndarray bug * Fix lint * Refactor * Fix windows openmp build failure (#94) * update mshadow submoduel (#95) * Revert "update mshadow submoduel (#95)" (#96) This reverts commit 1a129e4. * Refactor sparse tensor code (#99) * Initial checkin test_sparse_ndarray passes * Fix test failure * Clean up * Clean up * Move init backend op to ndarray_utils * Fix lint * Eliminate circular dependency on headers * More refactor * Fix gpu build and consolidate Slice for dense and sparse * Clean up * More refactor * Clean up * Fix gpu build * Fix comment * fix pylint (#100) * Fix refactor sparse gpu test (#104) * Fix gpu build * Fix * Fix gpu test failure * change idx types from int32 to int64 (#101) Conflicts: python/mxnet/test_utils.py tests/python/unittest/test_sparse_operator.py update mshadow submodule fix extra quotes in test script change indptr type to int64 better err message for rsp" * revert LOG(DEBUG) change (#105) * fix undefined zeros in optimizer.py (#106) * move init dns zeros to init_op.h for kvstore to use (#107) * Refactor cast storage (#109) * Refactor cast_storage * Add cast_storage cc and cu files * Remove redundant comments * Replace std::accumulate with ParallelAccumulate * Clean up * Fix windows build * Rowsparse kv (#111) * update kvstore unit test Conflicts: tests/python/unittest/test_kvstore.py update model/module.py Conflicts: python/mxnet/model.py python/mxnet/module/module.py fix lint resolve conflict remove int keys in kvstore update cast to str function * fix failed dist_sync_kv test * bug fix in comm to ensure merged gradient is of the right type bug fix in comm * row sparse dist kvstore draft (push only) row_sparse pull * add ndarray row sparse shared mem constructor * code refactoring * add test for row_sparse weight bug fix for kv server slicing add async support rsolve race condition in kvstore * resolve error after reb ase * fix lint (#113) * rename some python funciton (#114) * _to_rsp * _to_csr. raise NotImplementedError * todense * fix lint (#115) enable libsvm uniit test (apache#6839) remove shared mem slice for csr add csr ndarray iter test make osx nose test verbose disable libsvm iter test Move InferAttr to mxnet from nnvm (apache#6830) * Move InferAttr to mxnet from nnvm Replace nnvm infer attr functions in c_api Initial checkin Clean up Remove nnvm namespace for FInferShape, FInferType, and FInferStorageType Add new interface for InferStorageType Revert "Remove nnvm namespace for FInferShape, FInferType, and FInferStorageType" This reverts commit 8aedf05. Fix and clean up Fix lint Add nnvm changes Change infer function interface to accept only rvalue reference of graph Clean up Flush commits to show up in PR Add error handling for storage type inference failure Update nnvm * Fix pylint Change idx type switch for aux data (apache#6860) * Change idx type switch for aux data * Add mshadow commit Sparse dot enhancement (apache#6842) * Initial checkin Initial checkin Fix sparse dot test Fix unitest and add fallback for sparse dot * Add benchmark code * Revert "Add benchmark code" This reverts commit be009fe. * Fix bug * Fix storage shape * Remove unnecessary test code * Use idx type switch Implement dot(csr, rsp)=dns and dot(csr.T, rsp)=rsp and refactor (apache#6902) * Initial checkin Add dot(csr.T, rsp)=rsp2 Add infer storage for dot(csr, rsp)=dns and dot(csr.T, rsp)=rsp2 * Fix comments * Replace std::lower_bound with own impl for gpu use too * Add time profiling * Revert "Add time profiling" This reverts commit 8f5bb98. * Move dot and batch_dot to a single file * Move dot gpu impl to a .cuh file * More refactor * Fix include error LibsvmIter fix (apache#6898) * fix bug in libsvm iter which causes mem corruption * add test for news dataset * fix wrong path in test * fix import error for urllib * update url * replace bz command with bz module Optimized gpu dot kernels (apache#6937) * pulled update to mshadow * mshadow update * added optimized gpu kernels for dot(csr,dns)=dns and dot(csr.T,dns)=dns, and unit test * added __syncwarp to vector kernel and reduced number of writes to shared memory Refactor sparse tensor code (apache#6955) * Save stype in frontend to avoid c-api call for stype * Change storage_type to stype * Revert "Change storage_type to stype" This reverts commit 90db7d1. * Revert "Revert "Change storage_type to stype"" This reverts commit 0932838. Move ndarray.py, sparse_ndarray.py, ndarray_utils.py, and _ndarray_internal to ndarrary folder More refactor Move elementwise sum for rsp to ndarray_function.cc Remove unnecessary import in ndarray module Fix pylint Remove redundant code Remove _stype from slots Fix cpp-package build error caused by the change to imperative invoke interface Use relative import Remove print line Rename _ndarray_internal.py to _internal.py * Relaunch test... minor bug fix in warp synchronous code (apache#7029)

* [WIP] Sparse Tensor (apache#5800) * squash merge with 38f7c55 compiles on GPU update check alloc: Checkpoint. Pass elem-sum gpu test bug fix for copyfromto. sparse sgd test pass on gpu inefficient implementation for csr copy update submodule fix lint Simple bind with infer storage type (#32) * Symbol binding for sparse tensor development. (#31) * Initial checkin * Add init functions for simple bind in graph_executor * Add simple_bind c_api * Add simple bind c-api * Assign zeros to in_args, arg_grads, and aux_states * Add simple_bind2 python interface * Fix python interface bugs * Interface changes * Fix * Fix core dump * Add bind_ith_exec c_api * Change simple_bind2 * Fix seg fault * Finish simple_bind * Change _bind_ith_exec * Refactor simple_bind initialization flow for bind * Consolidate bind and simple_bind graph init flow * Fix bug * Clean up * Add comments * Clean up * Clean up * Minor correction * Rename APIs in graph executor * Refactor * Rebase * Delete deprecated functions * Move more front-end work to backend * Bug fix * Fix failed tests * Minor fix * Fix lint * Fix lint * Revert unnecessary changes * Revert * Revert * Clean up * Fix lint Conflicts: python/mxnet/symbol.py src/executor/graph_executor.cc * Add inferstorage to graph executor * re-enable tests for sparse embedding with simple_bind * type switch fix in sparse embedding" ; change `default` to `default_storage` for cast storage op (#33) * change default to default_storage * disable cpp test build temporarily attempt to fix windows build error, and fix lint (#34) update nnvm submodule (#37) Scipy build (#38) * update nnvm submodule * add scipy pip install for dockerfile Python3 unit tests (#39) * change xrange to range for python3 compatiblity" * remove more xrange from tests replace long with int for python3 (#40) fix the rest of TShape constructor errors (#41) fix lint (#42) fix wrong usage of mshadow::Shape1" (#43) implementation for Csr slice on cpu (#36) * CPU implementation for CSR remove seg_len from csr slice add some docs for slice csr change indptr, values, etc to be private member bug fix in sparse embedding update nnvm submoduel fix lint update unit test for sparse nd" * add const for SliceCsrIndPtr kernel Fix sparse dot according to the new RSP definition (#35) * Fix csr dot dns * Fix sparse dot * Add fallback and test cases for dot(csr, dns)=dns * Add int type switch * Fix * Fix * Fix update mshadow submodule (#44) Fix dns to rsp (#46) fix lint (#47) add runtime storage fallback detection" (#48) * add runtime storage fallback detection" * replace cast storage ex with cast storage impl Fm example (#45) * update csr slice logic to avoid confusion. add more exmaples. * add hint to module.update * more testcases(fallback) for sparse_nd * add to_csr() and to_rsp() method. More unit test (fallback now) * add fm test. fix lint * register sparse sgd under Optim.SGD * update dmlc-core submoduel * change indptr to _indptr temporarily. add const ref to fname fix lint fix lint; (#51) Guard gpu cast storage (#50) * Clean up * Fix typo Rearrange unit test files (#52) fix lint. add scipy for python_test. fix scipy.sparse import error. fix truediv for python3 fix travis test (#54) * remove pyc files * add verbose for travis nosetests cleanup some testing code and enums (#57) * update Makefile * refactor test_sparse_operator * change `default_storage` back to `default` * remove unused cpp tests port libsvm parser to mxnet as libsvm iter (#55) * copied csv iter to libsvm iter test libsvm iter draft handle round batch == false for csr batch loader code refactoring add get stype, shape interface to iiter separate class for sparse iter add missing file fix mem corruption' rename variables add comments also read label from libsvm add test. update docs. update submodule Conflicts: python/mxnet/sparse_ndarray.py * update submodule * fix lint * update test * revert naming change add benchmark scritp for dot (#59) * add benchmark scritp for dot add gpu option for bench add get_data funciton for benchmark print t_sparse, too; add comment change nnz to dnesity add backward * add comment update fm test (#62) introduce CSRNDarray and rowsparseNDarray to python frontend api (#58) * introduce CSRNDarray and rowsparseNDarray to python frontend api * temporarily disable fm_module test fix lint (#64) fix typo. disable libsvm io test (#65) Improve dot (#61) * Init checkin * Fix * Adjust dot parallelization methods * Set num_omp_threads for benchmark from command line * Fix omp thread number * Clean up * Add scipy as dot baseline * Fix format sparse_retain op (#66) * Initial checkin * Fix bugs * Add unit test for sparse_retain * Add example and modify test add storage cast for outputs that have non-default storage (#67) fix gpu build (#69) Fix test_sparse_retain python3 issue (#68) revert nnvm version * draft for sgd rsp rsp (#75) support sgd(rsp, rsp) support dot(csr, rsp) when rsp is full add ref to const ndarray params support sparse embedding with rsp weight' fix lint modify embedding backward to produce dense grad remove invalid_rid for rsp->dns remove previous embedding op changes pass sparse embedding test add STORAGE_TYPE_ASSIGN_CHECK remove backward storage infer * fix lint (#78) * fix lint (#79) * serial elemwise sum impl (#80) update module kvstore interface add other missing params and functions revert some interface changes revert some more changes reomve explicit casting for gradients on kvstore update Comm interface update fm example Conflicts: python/mxnet/model.py python/mxnet/ndarray.py * bug fix for initializing module with row_sparse weight (#81) * bug fix for initializing module with row_sparse weight * update log message * Sparse ndarray serialization and deserialization (#77) * Initial checkin * Add unit tests * Fix lint * Fix lint (#84) * Sgd with row_sparse weight, dns gradient (#83) * sgd rsp dns draft * support sgd_mom(rsp, dns, rsp) * update doc * remove cast storage for kv updater * code refactoring * update mshadow version (#88) * csr slice bug fix (#90) * benchmark dot code refactor (#87) * q^x6x add some code in benchmark * refactor * minor fixes * fix * lint fix * Add unit test (#91) * add unittest * minor fix * remove commented lines * change test func name * add test rsp * kvstore push row sparse (#93) * Add multi-thread cpu elemwise sum for rsps * Minor fix * Add flag to switch between serial and multi-thread kvstore push * Fix lint in sparse_ndarray.py * Revert "Fix lint in sparse_ndarray.py" This reverts commit d7225ec. * Fix ndarray init in copy(ctx) * Add env var to control the flow of serial/parallel reduce * Refactor * Fix copy ndarray bug * Fix lint * Refactor * Fix windows openmp build failure (#94) * update mshadow submoduel (#95) * Revert "update mshadow submoduel (#95)" (#96) This reverts commit 1a129e4. * Refactor sparse tensor code (#99) * Initial checkin test_sparse_ndarray passes * Fix test failure * Clean up * Clean up * Move init backend op to ndarray_utils * Fix lint * Eliminate circular dependency on headers * More refactor * Fix gpu build and consolidate Slice for dense and sparse * Clean up * More refactor * Clean up * Fix gpu build * Fix comment * fix pylint (#100) * Fix refactor sparse gpu test (#104) * Fix gpu build * Fix * Fix gpu test failure * change idx types from int32 to int64 (#101) Conflicts: python/mxnet/test_utils.py tests/python/unittest/test_sparse_operator.py update mshadow submodule fix extra quotes in test script change indptr type to int64 better err message for rsp" * revert LOG(DEBUG) change (#105) * fix undefined zeros in optimizer.py (#106) * move init dns zeros to init_op.h for kvstore to use (#107) * Refactor cast storage (#109) * Refactor cast_storage * Add cast_storage cc and cu files * Remove redundant comments * Replace std::accumulate with ParallelAccumulate * Clean up * Fix windows build * Rowsparse kv (#111) * update kvstore unit test Conflicts: tests/python/unittest/test_kvstore.py update model/module.py Conflicts: python/mxnet/model.py python/mxnet/module/module.py fix lint resolve conflict remove int keys in kvstore update cast to str function * fix failed dist_sync_kv test * bug fix in comm to ensure merged gradient is of the right type bug fix in comm * row sparse dist kvstore draft (push only) row_sparse pull * add ndarray row sparse shared mem constructor * code refactoring * add test for row_sparse weight bug fix for kv server slicing add async support rsolve race condition in kvstore * resolve error after reb ase * fix lint (#113) * rename some python funciton (#114) * _to_rsp * _to_csr. raise NotImplementedError * todense * fix lint (#115) enable libsvm uniit test (apache#6839) remove shared mem slice for csr add csr ndarray iter test make osx nose test verbose disable libsvm iter test Move InferAttr to mxnet from nnvm (apache#6830) * Move InferAttr to mxnet from nnvm Replace nnvm infer attr functions in c_api Initial checkin Clean up Remove nnvm namespace for FInferShape, FInferType, and FInferStorageType Add new interface for InferStorageType Revert "Remove nnvm namespace for FInferShape, FInferType, and FInferStorageType" This reverts commit 8aedf05. Fix and clean up Fix lint Add nnvm changes Change infer function interface to accept only rvalue reference of graph Clean up Flush commits to show up in PR Add error handling for storage type inference failure Update nnvm * Fix pylint Change idx type switch for aux data (apache#6860) * Change idx type switch for aux data * Add mshadow commit Sparse dot enhancement (apache#6842) * Initial checkin Initial checkin Fix sparse dot test Fix unitest and add fallback for sparse dot * Add benchmark code * Revert "Add benchmark code" This reverts commit be009fe. * Fix bug * Fix storage shape * Remove unnecessary test code * Use idx type switch Implement dot(csr, rsp)=dns and dot(csr.T, rsp)=rsp and refactor (apache#6902) * Initial checkin Add dot(csr.T, rsp)=rsp2 Add infer storage for dot(csr, rsp)=dns and dot(csr.T, rsp)=rsp2 * Fix comments * Replace std::lower_bound with own impl for gpu use too * Add time profiling * Revert "Add time profiling" This reverts commit 8f5bb98. * Move dot and batch_dot to a single file * Move dot gpu impl to a .cuh file * More refactor * Fix include error LibsvmIter fix (apache#6898) * fix bug in libsvm iter which causes mem corruption * add test for news dataset * fix wrong path in test * fix import error for urllib * update url * replace bz command with bz module Optimized gpu dot kernels (apache#6937) * pulled update to mshadow * mshadow update * added optimized gpu kernels for dot(csr,dns)=dns and dot(csr.T,dns)=dns, and unit test * added __syncwarp to vector kernel and reduced number of writes to shared memory Refactor sparse tensor code (apache#6955) * Save stype in frontend to avoid c-api call for stype * Change storage_type to stype * Revert "Change storage_type to stype" This reverts commit 90db7d1. * Revert "Revert "Change storage_type to stype"" This reverts commit 0932838. Move ndarray.py, sparse_ndarray.py, ndarray_utils.py, and _ndarray_internal to ndarrary folder More refactor Move elementwise sum for rsp to ndarray_function.cc Remove unnecessary import in ndarray module Fix pylint Remove redundant code Remove _stype from slots Fix cpp-package build error caused by the change to imperative invoke interface Use relative import Remove print line Rename _ndarray_internal.py to _internal.py * Relaunch test... minor bug fix in warp synchronous code (apache#7029) * move storage type vector from nnvm to mxnet (apache#7054) * move storage type vector from nnvm to mxnet * update nnvm * update nnvm * Improve copy sparse tensors (apache#7003) * Use cast_storage when copying ndarrays of different stypes on same context * Relaunch test * fix failed tests. add back 64bit support for dot fix lint * bug fix for IdentityComputeRsp * fix lint fix lint fix lint * add data partition for libsvm iter (apache#7027) * remove sparse embedding (apache#7165) * fix ndarray namespace * remove untested gpu operators (apache#7172) * skip sparse dot gpu tset. add sparse_nd_zeros gpu test * remove sparse_retain gpu Conflicts: tests/python/gpu/test_operator_gpu.py * Fix ndarray aux data issue (apache#7098) * Fix getting sparse ndarray data/aux_data issues * Add tests for func csr and row_sparse * Make get/set data/aux_data thread safe * Fix a bug * Fix typo and comment * More comments * Correct comment Conflicts: tests/python/gpu/test_operator_gpu.py * Support K-dimensional row-sparse tensor (apache#7179) * remove check for k dimensional rowsparse tensor * change var name for rsp sgd operator * add checks for sparse dot * bug fix for kdim rowsparse cast storage cpu * update IdentityLikeRhsComputeEx interface * remove set_storage_shape from ndarray. support elemwise_add with kdim row_sparse tensor * use get_with_shape instead of reshape * update according to comments Conflicts: src/operator/tensor/elemwise_unary_op.h * Improve sparse ndarray error message (apache#7181) * add test for broadcast_to * add comments Conflicts: python/mxnet/base.py * construct row_sparse ndarray for dist-async fix bug in rsp add rsp sync push race condition for push fix bug in rsp pull. refactor test cleanup comments refactor dist server fix lint fix storage shape issue with the new ndarray constructor data sharding draft; fix lint. add comment add support for zeros gradients use std::upper_bound/lower_bound remove special init function for rowsparse dist kvstore temporary support for inplace operators for sparse add test. fix return type store kRowSparseNDArray in kv server remove fcomp_ex sgd with dns weight and rsp gradient bug fix in sparse retain sparse pull c_api revise rowsparse pull api use engine to compute unique to ensure thread safety add rowsparse pull to dist-kv fix lint add example for rsp_pull remove name2idx; add sparse_pull_dict param to module fix unit test and c rowid conversion support str key type in kvstore (apache#6765) * update kvstore unit test * update model/module.py * fix lint * remove int keys in kvstore * update cast to str function * remove _cast_to_str_keys * fix lint * always cast to str Conflicts: include/mxnet/c_api.h include/mxnet/kvstore.h python/mxnet/kvstore.py python/mxnet/model.py python/mxnet/module/module.py src/c_api/c_api.cc src/kvstore/kvstore_local.h tests/python/unittest/test_kvstore.py update module API for other submodules update stypes in kvstore after refactoring change type of size from size_t to int64_t add sparse linear regression example remove sparse_pull_dict from module fix init_optim for seq_module. update sparse example resolve conflict for binary add rsp rsp Conflicts: python/mxnet/kvstore.py tests/python/unittest/test_kvstore.py * fix DotCsrRspRspImpl error message (apache#7191) * GPU implementation of cast_storage (dense to csr) (apache#7081) * Added gpu implementation for cast_storage dense to csr, unit tests, and benchmark. Additionally, cast_storage interface change to accommodate the need of temporary storage in cuda kernels. * fixed whitespace * minor unittest update * removed whitespace * add cast storage benchmark params info Conflicts: tests/python/gpu/test_operator_gpu.py * Sparse square sum (apache#7206) * Add square_sum op * Add unit test and fix check_numeric_gradient * Add .cu file and example * Fix lint * Remove gpu registration * Use square_sum in test_module_fm * Modify and Add documentation for mx.nd.zeros (apache#7197) * Modify and Add documentation for mx.nd.zeros * Change context to cpu * Change stype to optional * Change ordering and remove optional for _zeros_sparse_ndarray * Expose kWriteInplace for imperative execution (fcompute_ex and fstatefulcompute_ex) (#133) * expose kWriteInplace to FComputeEx and FStatefulComputeEx * refactor ccode * remove duplicated test * Operator add_n for row sparse ndarrays (apache#7244) * Add add_n op for row-sparse ndarrays and identity FComputeEx * Fix bug in square_sum * Remove test_cast_storage_ex from gpu test since it's not implemented yet * Fix according to the cr Conflicts: src/operator/tensor/elemwise_sum.cc src/operator/tensor/elemwise_unary_op.cc tests/python/gpu/test_operator_gpu.py resolve conflict * GPU implementation of cast_storage (dense to rsp) (apache#7223) * CastStorageDnsRsp GPU Implementation * updating function doc and some variable types and names * adding cuda_get_device_prop() util function * added rand_shape function for n-dimensional tensors * updated cast storage unit test * added dns_to_rsp to cast storage benchmark script * removing redundant unit test * fix lint * minor change in benchmark script * fix lint * correct function description * change storage_type to stype * changed scope of using namespaces * changed variable types from index_t to dim_t * resolve merge conflict in ndarray.load * Improve StatefulOp/FCompute storage fallback (#134) * test for fcomp fallback add storage fallback test and optimize fallback logic rename function, add comments use std size() * add autograd test with sparse inputs * update sparse ndarray api (#139) * support mx.nd.empty for sparse ndarray Change SparseNDArray to BaseSparseNDArray support mx.nd.array with BaseSparseNDArray inputs. Update documentation with explicit subclasses of NDArrays Conflicts: python/mxnet/ndarray/__init__.py python/mxnet/ndarray/ndarray.py python/mxnet/ndarray/sparse_ndarray.py tests/python/unittest/test_sparse_ndarray.py * fix print msg in test * Handle ograd_stype='row_sparse' for square_sum backward (#143) * Add one kernel for square_sum backward pass to take rsp ograd * Add kNullOp and change to use type_assign in infer stype fallback * Sparse retain improvement (#138) * Add one more kernel for sparse retain * Fix compile * Change STORAGE_TYPE_ASSIGN_CHECK to type_assign for fallback * Fix * Add gpu compile * ignoring variables in SimpleBind that is used on python's sparse branch for now. (#135) * add bias term to fm test (#145) * update ndarray.nd, remove `invoke` from excluded members (#137) remove __weakref__ from SparseNDArray add data indice to doc revert dlpack update revert mxdoc changes move methods from BaseSparseNDarray to csrndarray and rwosparse ndarray * support storage fallback with mutable inputs (#147) * include mutatable inputs in storage fallback. refactor executor add fallback test for rms prop and adam fix lint fix lint fix test in optimizer * update according to comments * fix unit tests * fix gpu compilation err * Code changes based on reviews (#144) * code changes according to review comments remove executor debug. add doc to optimizer update sparse sgd test add dtype option to rand_sparse_ndarray * overhauled reqs for sparse operators * patch FCompExFallback with mutable inputs. update test_optimizer with more fallback cases * change executor debug macro to env var * add comment * update doc * change ndarray.aux_shape() to return const reference * remove todense to_rsp to_csr. replace with tostype * replace manual calls to cast_storage with tostype * disable gpu fallback test for optimizer * fix lint * add backward pass for cast_storage. refactor cast_storage test * rand_sparse_ndarray bug fix * fix cast_storage for gpu * disable csr test for fp16 * update row sparse ndarray doc * update doc * small edits according to reviews (#151) * fix lint (#152) * add license to all new files in sparse brnach (#154) * Allocate temp data on the fly for some casting operations (#149) * fix utf8 encoding in sparse ndarray * Extending the GPU dot operator (apache#7226) * Added GPU DotCsrRspDnsImpl declaration and TODOs * cleaning up function doc, variable types, and code-style * minor bug fixes * enable GPU dot(csr,rsp)=dns unit test * extend sparse dot unit test * adding GPU impl of DotCsrRspDns and its kernels * add TODO * changed variable types from index_t to dim_t * fix function description * added DotCsrRspRspImpl and its kernels (baseline, functionality) * added DotCsrDnsRspImpl and its kernels (baseline, functionality); plus code documentation * refactored dot benchmark * optimized DotCsrTransDnsRsp GPU kernel * change of dot impl interface to include OpContext, for temp storage * removing __device__ flag from CPU kernels * minor fixes and changing variable data types * minor fixes based on code reviews Conflicts: benchmark/python/sparse_op.py tests/python/gpu/test_operator_gpu.py tests/python/unittest/test_sparse_operator.py * Add get_synthetic_dataset function to util (#146) * Add get_synthetic_datasets * Move to test_utils * Remove _get_uniform_dataset * Move validation to its own function * Refactor the validation code for csr generation * Make test_powerlaw a nested function * Change SparseNDArray to CSRNDArray * Merge with dtype specific changes in test_utils * temporary fix for batch norm storage fallback (#156) * support random_uniform/normal/gamma with row_sparse output (#155) * add support for initilazer with rowsparse output * add scalar assignment to row_sparse * add setitem test to gpu * Revert "add scalar assignment to row_sparse" This reverts commit 8aef7a5. * Revert "add setitem test to gpu" This reverts commit 3b969ac. * Square sum backward support one more case (#161) * Add documentation for sparse ops (#148) * draft doc for sparse op * add more stype doc for operators * add doc for cast_storage * see also cast_storage. remove base sparse ndarray. fix aux_types comemtn * grammar / spelling fix * A few fixes (#163) * fix batch norm gpu kernel. register random operators on gpu * register sparse random op on gpu, too * Minor fixes sparse ops (#160) * change CPU kernel inline directives, data types, and function doc * update dot dtype switch to use 32 and 64bit floating point only * use type_assign instead of STORAGE_TYPE_ASSIGN_CHECK * added tensor_util-inl.cuh file for common tensor operator GPU kernels * sparse Adam optimizer (#164) * add sparse adam * register gpu op * add comments * cr comments * kvstore.row_sparse_pull for GPU and end-to-end benchmark: CPU vs. multi-GPUs (#150) * Add gpu support for BroadcastRowSparse * Fix bugs * Add benchmark script * Increase output dim size * Update weight on CPU using single GPU for sparse tensors * More fix * Optimize sparse_retain for special case * Change row sparse pull locations * Avoid sparse retain on cpu if possible * Use acc for metric * Fix misc * fix bug in adam update (#167) fix a bug in adam update * change sparse example from regression to classification (#165)

* [WIP] Sparse Tensor (#5800) * squash merge with 38f7c55 compiles on GPU update check alloc: Checkpoint. Pass elem-sum gpu test bug fix for copyfromto. sparse sgd test pass on gpu inefficient implementation for csr copy update submodule fix lint Simple bind with infer storage type (#32) * Symbol binding for sparse tensor development. (#31) * Initial checkin * Add init functions for simple bind in graph_executor * Add simple_bind c_api * Add simple bind c-api * Assign zeros to in_args, arg_grads, and aux_states * Add simple_bind2 python interface * Fix python interface bugs * Interface changes * Fix * Fix core dump * Add bind_ith_exec c_api * Change simple_bind2 * Fix seg fault * Finish simple_bind * Change _bind_ith_exec * Refactor simple_bind initialization flow for bind * Consolidate bind and simple_bind graph init flow * Fix bug * Clean up * Add comments * Clean up * Clean up * Minor correction * Rename APIs in graph executor * Refactor * Rebase * Delete deprecated functions * Move more front-end work to backend * Bug fix * Fix failed tests * Minor fix * Fix lint * Fix lint * Revert unnecessary changes * Revert * Revert * Clean up * Fix lint Conflicts: python/mxnet/symbol.py src/executor/graph_executor.cc * Add inferstorage to graph executor * re-enable tests for sparse embedding with simple_bind * type switch fix in sparse embedding" ; change `default` to `default_storage` for cast storage op (#33) * change default to default_storage * disable cpp test build temporarily attempt to fix windows build error, and fix lint (#34) update nnvm submodule (#37) Scipy build (#38) * update nnvm submodule * add scipy pip install for dockerfile Python3 unit tests (#39) * change xrange to range for python3 compatiblity" * remove more xrange from tests replace long with int for python3 (#40) fix the rest of TShape constructor errors (#41) fix lint (#42) fix wrong usage of mshadow::Shape1" (#43) implementation for Csr slice on cpu (#36) * CPU implementation for CSR remove seg_len from csr slice add some docs for slice csr change indptr, values, etc to be private member bug fix in sparse embedding update nnvm submoduel fix lint update unit test for sparse nd" * add const for SliceCsrIndPtr kernel Fix sparse dot according to the new RSP definition (#35) * Fix csr dot dns * Fix sparse dot * Add fallback and test cases for dot(csr, dns)=dns * Add int type switch * Fix * Fix * Fix update mshadow submodule (#44) Fix dns to rsp (#46) fix lint (#47) add runtime storage fallback detection" (#48) * add runtime storage fallback detection" * replace cast storage ex with cast storage impl Fm example (#45) * update csr slice logic to avoid confusion. add more exmaples. * add hint to module.update * more testcases(fallback) for sparse_nd * add to_csr() and to_rsp() method. More unit test (fallback now) * add fm test. fix lint * register sparse sgd under Optim.SGD * update dmlc-core submoduel * change indptr to _indptr temporarily. add const ref to fname fix lint fix lint; (#51) Guard gpu cast storage (#50) * Clean up * Fix typo Rearrange unit test files (#52) fix lint. add scipy for python_test. fix scipy.sparse import error. fix truediv for python3 fix travis test (#54) * remove pyc files * add verbose for travis nosetests cleanup some testing code and enums (#57) * update Makefile * refactor test_sparse_operator * change `default_storage` back to `default` * remove unused cpp tests port libsvm parser to mxnet as libsvm iter (#55) * copied csv iter to libsvm iter test libsvm iter draft handle round batch == false for csr batch loader code refactoring add get stype, shape interface to iiter separate class for sparse iter add missing file fix mem corruption' rename variables add comments also read label from libsvm add test. update docs. update submodule Conflicts: python/mxnet/sparse_ndarray.py * update submodule * fix lint * update test * revert naming change add benchmark scritp for dot (#59) * add benchmark scritp for dot add gpu option for bench add get_data funciton for benchmark print t_sparse, too; add comment change nnz to dnesity add backward * add comment update fm test (#62) introduce CSRNDarray and rowsparseNDarray to python frontend api (#58) * introduce CSRNDarray and rowsparseNDarray to python frontend api * temporarily disable fm_module test fix lint (#64) fix typo. disable libsvm io test (#65) Improve dot (#61) * Init checkin * Fix * Adjust dot parallelization methods * Set num_omp_threads for benchmark from command line * Fix omp thread number * Clean up * Add scipy as dot baseline * Fix format sparse_retain op (#66) * Initial checkin * Fix bugs * Add unit test for sparse_retain * Add example and modify test add storage cast for outputs that have non-default storage (#67) fix gpu build (#69) Fix test_sparse_retain python3 issue (#68) revert nnvm version * draft for sgd rsp rsp (#75) support sgd(rsp, rsp) support dot(csr, rsp) when rsp is full add ref to const ndarray params support sparse embedding with rsp weight' fix lint modify embedding backward to produce dense grad remove invalid_rid for rsp->dns remove previous embedding op changes pass sparse embedding test add STORAGE_TYPE_ASSIGN_CHECK remove backward storage infer * fix lint (#78) * fix lint (#79) * serial elemwise sum impl (#80) update module kvstore interface add other missing params and functions revert some interface changes revert some more changes reomve explicit casting for gradients on kvstore update Comm interface update fm example Conflicts: python/mxnet/model.py python/mxnet/ndarray.py * bug fix for initializing module with row_sparse weight (#81) * bug fix for initializing module with row_sparse weight * update log message * Sparse ndarray serialization and deserialization (#77) * Initial checkin * Add unit tests * Fix lint * Fix lint (#84) * Sgd with row_sparse weight, dns gradient (#83) * sgd rsp dns draft * support sgd_mom(rsp, dns, rsp) * update doc * remove cast storage for kv updater * code refactoring * update mshadow version (#88) * csr slice bug fix (#90) * benchmark dot code refactor (#87) * q^x6x add some code in benchmark * refactor * minor fixes * fix * lint fix * Add unit test (#91) * add unittest * minor fix * remove commented lines * change test func name * add test rsp * kvstore push row sparse (#93) * Add multi-thread cpu elemwise sum for rsps * Minor fix * Add flag to switch between serial and multi-thread kvstore push * Fix lint in sparse_ndarray.py * Revert "Fix lint in sparse_ndarray.py" This reverts commit d7225ec. * Fix ndarray init in copy(ctx) * Add env var to control the flow of serial/parallel reduce * Refactor * Fix copy ndarray bug * Fix lint * Refactor * Fix windows openmp build failure (#94) * update mshadow submoduel (#95) * Revert "update mshadow submoduel (#95)" (#96) This reverts commit 1a129e4. * Refactor sparse tensor code (#99) * Initial checkin test_sparse_ndarray passes * Fix test failure * Clean up * Clean up * Move init backend op to ndarray_utils * Fix lint * Eliminate circular dependency on headers * More refactor * Fix gpu build and consolidate Slice for dense and sparse * Clean up * More refactor * Clean up * Fix gpu build * Fix comment * fix pylint (#100) * Fix refactor sparse gpu test (#104) * Fix gpu build * Fix * Fix gpu test failure * change idx types from int32 to int64 (#101) Conflicts: python/mxnet/test_utils.py tests/python/unittest/test_sparse_operator.py update mshadow submodule fix extra quotes in test script change indptr type to int64 better err message for rsp" * revert LOG(DEBUG) change (#105) * fix undefined zeros in optimizer.py (#106) * move init dns zeros to init_op.h for kvstore to use (#107) * Refactor cast storage (#109) * Refactor cast_storage * Add cast_storage cc and cu files * Remove redundant comments * Replace std::accumulate with ParallelAccumulate * Clean up * Fix windows build * Rowsparse kv (#111) * update kvstore unit test Conflicts: tests/python/unittest/test_kvstore.py update model/module.py Conflicts: python/mxnet/model.py python/mxnet/module/module.py fix lint resolve conflict remove int keys in kvstore update cast to str function * fix failed dist_sync_kv test * bug fix in comm to ensure merged gradient is of the right type bug fix in comm * row sparse dist kvstore draft (push only) row_sparse pull * add ndarray row sparse shared mem constructor * code refactoring * add test for row_sparse weight bug fix for kv server slicing add async support rsolve race condition in kvstore * resolve error after reb ase * fix lint (#113) * rename some python funciton (#114) * _to_rsp * _to_csr. raise NotImplementedError * todense * fix lint (#115) enable libsvm uniit test (#6839) remove shared mem slice for csr add csr ndarray iter test make osx nose test verbose disable libsvm iter test Move InferAttr to mxnet from nnvm (#6830) * Move InferAttr to mxnet from nnvm Replace nnvm infer attr functions in c_api Initial checkin Clean up Remove nnvm namespace for FInferShape, FInferType, and FInferStorageType Add new interface for InferStorageType Revert "Remove nnvm namespace for FInferShape, FInferType, and FInferStorageType" This reverts commit 8aedf05. Fix and clean up Fix lint Add nnvm changes Change infer function interface to accept only rvalue reference of graph Clean up Flush commits to show up in PR Add error handling for storage type inference failure Update nnvm * Fix pylint Change idx type switch for aux data (#6860) * Change idx type switch for aux data * Add mshadow commit Sparse dot enhancement (#6842) * Initial checkin Initial checkin Fix sparse dot test Fix unitest and add fallback for sparse dot * Add benchmark code * Revert "Add benchmark code" This reverts commit be009fe. * Fix bug * Fix storage shape * Remove unnecessary test code * Use idx type switch Implement dot(csr, rsp)=dns and dot(csr.T, rsp)=rsp and refactor (#6902) * Initial checkin Add dot(csr.T, rsp)=rsp2 Add infer storage for dot(csr, rsp)=dns and dot(csr.T, rsp)=rsp2 * Fix comments * Replace std::lower_bound with own impl for gpu use too * Add time profiling * Revert "Add time profiling" This reverts commit 8f5bb98. * Move dot and batch_dot to a single file * Move dot gpu impl to a .cuh file * More refactor * Fix include error LibsvmIter fix (#6898) * fix bug in libsvm iter which causes mem corruption * add test for news dataset * fix wrong path in test * fix import error for urllib * update url * replace bz command with bz module Optimized gpu dot kernels (#6937) * pulled update to mshadow * mshadow update * added optimized gpu kernels for dot(csr,dns)=dns and dot(csr.T,dns)=dns, and unit test * added __syncwarp to vector kernel and reduced number of writes to shared memory Refactor sparse tensor code (#6955) * Save stype in frontend to avoid c-api call for stype * Change storage_type to stype * Revert "Change storage_type to stype" This reverts commit 90db7d1. * Revert "Revert "Change storage_type to stype"" This reverts commit 0932838. Move ndarray.py, sparse_ndarray.py, ndarray_utils.py, and _ndarray_internal to ndarrary folder More refactor Move elementwise sum for rsp to ndarray_function.cc Remove unnecessary import in ndarray module Fix pylint Remove redundant code Remove _stype from slots Fix cpp-package build error caused by the change to imperative invoke interface Use relative import Remove print line Rename _ndarray_internal.py to _internal.py * Relaunch test... minor bug fix in warp synchronous code (#7029) * move storage type vector from nnvm to mxnet (#7054) * move storage type vector from nnvm to mxnet * update nnvm * update nnvm * Improve copy sparse tensors (#7003) * Use cast_storage when copying ndarrays of different stypes on same context * Relaunch test * fix failed tests. add back 64bit support for dot fix lint * bug fix for IdentityComputeRsp * fix lint fix lint fix lint * add data partition for libsvm iter (#7027) * remove sparse embedding (#7165) * fix ndarray namespace * remove untested gpu operators (#7172) * skip sparse dot gpu tset. add sparse_nd_zeros gpu test * remove sparse_retain gpu Conflicts: tests/python/gpu/test_operator_gpu.py * Fix ndarray aux data issue (#7098) * Fix getting sparse ndarray data/aux_data issues * Add tests for func csr and row_sparse * Make get/set data/aux_data thread safe * Fix a bug * Fix typo and comment * More comments * Correct comment Conflicts: tests/python/gpu/test_operator_gpu.py * Support K-dimensional row-sparse tensor (#7179) * remove check for k dimensional rowsparse tensor * change var name for rsp sgd operator * add checks for sparse dot * bug fix for kdim rowsparse cast storage cpu * update IdentityLikeRhsComputeEx interface * remove set_storage_shape from ndarray. support elemwise_add with kdim row_sparse tensor * use get_with_shape instead of reshape * update according to comments Conflicts: src/operator/tensor/elemwise_unary_op.h * Improve sparse ndarray error message (#7181) * add test for broadcast_to * add comments Conflicts: python/mxnet/base.py * construct row_sparse ndarray for dist-async fix bug in rsp add rsp sync push race condition for push fix bug in rsp pull. refactor test cleanup comments refactor dist server fix lint fix storage shape issue with the new ndarray constructor data sharding draft; fix lint. add comment add support for zeros gradients use std::upper_bound/lower_bound remove special init function for rowsparse dist kvstore temporary support for inplace operators for sparse add test. fix return type store kRowSparseNDArray in kv server remove fcomp_ex sgd with dns weight and rsp gradient bug fix in sparse retain sparse pull c_api revise rowsparse pull api use engine to compute unique to ensure thread safety add rowsparse pull to dist-kv fix lint add example for rsp_pull remove name2idx; add sparse_pull_dict param to module fix unit test and c rowid conversion support str key type in kvstore (#6765) * update kvstore unit test * update model/module.py * fix lint * remove int keys in kvstore * update cast to str function * remove _cast_to_str_keys * fix lint * always cast to str Conflicts: include/mxnet/c_api.h include/mxnet/kvstore.h python/mxnet/kvstore.py python/mxnet/model.py python/mxnet/module/module.py src/c_api/c_api.cc src/kvstore/kvstore_local.h tests/python/unittest/test_kvstore.py update module API for other submodules update stypes in kvstore after refactoring change type of size from size_t to int64_t add sparse linear regression example remove sparse_pull_dict from module fix init_optim for seq_module. update sparse example resolve conflict for binary add rsp rsp Conflicts: python/mxnet/kvstore.py tests/python/unittest/test_kvstore.py * fix DotCsrRspRspImpl error message (#7191) * GPU implementation of cast_storage (dense to csr) (#7081) * Added gpu implementation for cast_storage dense to csr, unit tests, and benchmark. Additionally, cast_storage interface change to accommodate the need of temporary storage in cuda kernels. * fixed whitespace * minor unittest update * removed whitespace * add cast storage benchmark params info Conflicts: tests/python/gpu/test_operator_gpu.py * Sparse square sum (#7206) * Add square_sum op * Add unit test and fix check_numeric_gradient * Add .cu file and example * Fix lint * Remove gpu registration * Use square_sum in test_module_fm * Modify and Add documentation for mx.nd.zeros (#7197) * Modify and Add documentation for mx.nd.zeros * Change context to cpu * Change stype to optional * Change ordering and remove optional for _zeros_sparse_ndarray * Expose kWriteInplace for imperative execution (fcompute_ex and fstatefulcompute_ex) (#133) * expose kWriteInplace to FComputeEx and FStatefulComputeEx * refactor ccode * remove duplicated test * Operator add_n for row sparse ndarrays (#7244) * Add add_n op for row-sparse ndarrays and identity FComputeEx * Fix bug in square_sum * Remove test_cast_storage_ex from gpu test since it's not implemented yet * Fix according to the cr Conflicts: src/operator/tensor/elemwise_sum.cc src/operator/tensor/elemwise_unary_op.cc tests/python/gpu/test_operator_gpu.py resolve conflict * GPU implementation of cast_storage (dense to rsp) (#7223) * CastStorageDnsRsp GPU Implementation * updating function doc and some variable types and names * adding cuda_get_device_prop() util function * added rand_shape function for n-dimensional tensors * updated cast storage unit test * added dns_to_rsp to cast storage benchmark script * removing redundant unit test * fix lint * minor change in benchmark script * fix lint * correct function description * change storage_type to stype * changed scope of using namespaces * changed variable types from index_t to dim_t * resolve merge conflict in ndarray.load * Improve StatefulOp/FCompute storage fallback (#134) * test for fcomp fallback add storage fallback test and optimize fallback logic rename function, add comments use std size() * add autograd test with sparse inputs * update sparse ndarray api (#139) * support mx.nd.empty for sparse ndarray Change SparseNDArray to BaseSparseNDArray support mx.nd.array with BaseSparseNDArray inputs. Update documentation with explicit subclasses of NDArrays Conflicts: python/mxnet/ndarray/__init__.py python/mxnet/ndarray/ndarray.py python/mxnet/ndarray/sparse_ndarray.py tests/python/unittest/test_sparse_ndarray.py * fix print msg in test * Handle ograd_stype='row_sparse' for square_sum backward (#143) * Add one kernel for square_sum backward pass to take rsp ograd * Add kNullOp and change to use type_assign in infer stype fallback * Sparse retain improvement (#138) * Add one more kernel for sparse retain * Fix compile * Change STORAGE_TYPE_ASSIGN_CHECK to type_assign for fallback * Fix * Add gpu compile * ignoring variables in SimpleBind that is used on python's sparse branch for now. (#135) * add bias term to fm test (#145) * update ndarray.nd, remove `invoke` from excluded members (#137) remove __weakref__ from SparseNDArray add data indice to doc revert dlpack update revert mxdoc changes move methods from BaseSparseNDarray to csrndarray and rwosparse ndarray * support storage fallback with mutable inputs (#147) * include mutatable inputs in storage fallback. refactor executor add fallback test for rms prop and adam fix lint fix lint fix test in optimizer * update according to comments * fix unit tests * fix gpu compilation err * Code changes based on reviews (#144) * code changes according to review comments remove executor debug. add doc to optimizer update sparse sgd test add dtype option to rand_sparse_ndarray * overhauled reqs for sparse operators * patch FCompExFallback with mutable inputs. update test_optimizer with more fallback cases * change executor debug macro to env var * add comment * update doc * change ndarray.aux_shape() to return const reference * remove todense to_rsp to_csr. replace with tostype * replace manual calls to cast_storage with tostype * disable gpu fallback test for optimizer * fix lint * add backward pass for cast_storage. refactor cast_storage test * rand_sparse_ndarray bug fix * fix cast_storage for gpu * disable csr test for fp16 * update row sparse ndarray doc * update doc * small edits according to reviews (#151) * fix lint (#152) * add license to all new files in sparse brnach (#154) * Allocate temp data on the fly for some casting operations (#149) * fix utf8 encoding in sparse ndarray * Extending the GPU dot operator (#7226) * Added GPU DotCsrRspDnsImpl declaration and TODOs * cleaning up function doc, variable types, and code-style * minor bug fixes * enable GPU dot(csr,rsp)=dns unit test * extend sparse dot unit test * adding GPU impl of DotCsrRspDns and its kernels * add TODO * changed variable types from index_t to dim_t * fix function description * added DotCsrRspRspImpl and its kernels (baseline, functionality) * added DotCsrDnsRspImpl and its kernels (baseline, functionality); plus code documentation * refactored dot benchmark * optimized DotCsrTransDnsRsp GPU kernel * change of dot impl interface to include OpContext, for temp storage * removing __device__ flag from CPU kernels * minor fixes and changing variable data types * minor fixes based on code reviews Conflicts: benchmark/python/sparse_op.py tests/python/gpu/test_operator_gpu.py tests/python/unittest/test_sparse_operator.py * Add get_synthetic_dataset function to util (#146) * Add get_synthetic_datasets * Move to test_utils * Remove _get_uniform_dataset * Move validation to its own function * Refactor the validation code for csr generation * Make test_powerlaw a nested function * Change SparseNDArray to CSRNDArray * Merge with dtype specific changes in test_utils * temporary fix for batch norm storage fallback (#156) * support random_uniform/normal/gamma with row_sparse output (#155) * add support for initilazer with rowsparse output * add scalar assignment to row_sparse * add setitem test to gpu * Revert "add scalar assignment to row_sparse" This reverts commit 8aef7a5. * Revert "add setitem test to gpu" This reverts commit 3b969ac. * Square sum backward support one more case (#161) * Add documentation for sparse ops (#148) * draft doc for sparse op * add more stype doc for operators * add doc for cast_storage * see also cast_storage. remove base sparse ndarray. fix aux_types comemtn * grammar / spelling fix * A few fixes (#163) * fix batch norm gpu kernel. register random operators on gpu * register sparse random op on gpu, too * Minor fixes sparse ops (#160) * change CPU kernel inline directives, data types, and function doc * update dot dtype switch to use 32 and 64bit floating point only * use type_assign instead of STORAGE_TYPE_ASSIGN_CHECK * added tensor_util-inl.cuh file for common tensor operator GPU kernels * sparse Adam optimizer (#164) * add sparse adam * register gpu op * add comments * cr comments * kvstore.row_sparse_pull for GPU and end-to-end benchmark: CPU vs. multi-GPUs (#150) * Add gpu support for BroadcastRowSparse * Fix bugs * Add benchmark script * Increase output dim size * Update weight on CPU using single GPU for sparse tensors * More fix * Optimize sparse_retain for special case * Change row sparse pull locations * Avoid sparse retain on cpu if possible * Use acc for metric * Fix misc * fix bug in adam update (#167) fix a bug in adam update * change sparse example from regression to classification (#165) * fix python import (#166) * Add waitall to sparse_end2end.py (#169) * Add waitall() * Add dummy metric option * Add header license * Dot script changes (#159) * Add get_synthetic_datasets * Move to test_utils * Remove _get_uniform_dataset * Move validation to its own function * Refactor the validation code for csr generation * Make test_powerlaw a nested function * Change SparseNDArray to CSRNDArray * Refactoring changes to dot.py * Fix mxnet test_utils changes * Remove pdb statement * Add distribution parameter * Refactor benchmarking script * Remove unused code * Make style changes and remove unused code * Change typo in comment * Add transpose support * Change typo * 4 decimal points needed for density * Add rsp support for real datasets * Correct variable name mini_file_name * Move wait_to_read outside if * Seperate out scipy and mxnet logic in bench_dot * Fix lhs_trans issue * Move transpose outside measure_cost * Compute transpose inside measure_cost * Remove unused variables * Transpose only if trans_lhs (#171) * fix default val for distribution (#172) * fix lint (#175) * avoid cast_storage in dist-kvstore-server (#174) * avoid cast_storage in dist-kvstore-server * add stream arg to mshadow;;copy * fix copy order * Add sparse namespace to ndarray and symbol (#177) * Register dot, cast_storage, and sparse_retain under mxnet.ndarray.sparse * Add sparse to symbol namespace * Delete commented code * mv sparse_ndarray.py sparse.py * Clean up * Change docstring * changes based on code reviews (#176) * remove scipy dependency * move kvstore checks to backned * add const to lambda * temp fix to ndarray.md (#178) * Fix sparse namespace pylint (#179) * add comments and error msg (#181) * add clarification for csr (#182) * add clarification for csr * cr comments * revert change in test util (#183) * fix amalgamation (#184) * fix lint

* [WIP] Sparse Tensor (apache#5800) * squash merge with 38f7c55 compiles on GPU update check alloc: Checkpoint. Pass elem-sum gpu test bug fix for copyfromto. sparse sgd test pass on gpu inefficient implementation for csr copy update submodule fix lint Simple bind with infer storage type (apache#32) * Symbol binding for sparse tensor development. (apache#31) * Initial checkin * Add init functions for simple bind in graph_executor * Add simple_bind c_api * Add simple bind c-api * Assign zeros to in_args, arg_grads, and aux_states * Add simple_bind2 python interface * Fix python interface bugs * Interface changes * Fix * Fix core dump * Add bind_ith_exec c_api * Change simple_bind2 * Fix seg fault * Finish simple_bind * Change _bind_ith_exec * Refactor simple_bind initialization flow for bind * Consolidate bind and simple_bind graph init flow * Fix bug * Clean up * Add comments * Clean up * Clean up * Minor correction * Rename APIs in graph executor * Refactor * Rebase * Delete deprecated functions * Move more front-end work to backend * Bug fix * Fix failed tests * Minor fix * Fix lint * Fix lint * Revert unnecessary changes * Revert * Revert * Clean up * Fix lint Conflicts: python/mxnet/symbol.py src/executor/graph_executor.cc * Add inferstorage to graph executor * re-enable tests for sparse embedding with simple_bind * type switch fix in sparse embedding" ; change `default` to `default_storage` for cast storage op (apache#33) * change default to default_storage * disable cpp test build temporarily attempt to fix windows build error, and fix lint (apache#34) update nnvm submodule (apache#37) Scipy build (apache#38) * update nnvm submodule * add scipy pip install for dockerfile Python3 unit tests (apache#39) * change xrange to range for python3 compatiblity" * remove more xrange from tests replace long with int for python3 (apache#40) fix the rest of TShape constructor errors (apache#41) fix lint (apache#42) fix wrong usage of mshadow::Shape1" (apache#43) implementation for Csr slice on cpu (apache#36) * CPU implementation for CSR remove seg_len from csr slice add some docs for slice csr change indptr, values, etc to be private member bug fix in sparse embedding update nnvm submoduel fix lint update unit test for sparse nd" * add const for SliceCsrIndPtr kernel Fix sparse dot according to the new RSP definition (apache#35) * Fix csr dot dns * Fix sparse dot * Add fallback and test cases for dot(csr, dns)=dns * Add int type switch * Fix * Fix * Fix update mshadow submodule (apache#44) Fix dns to rsp (apache#46) fix lint (apache#47) add runtime storage fallback detection" (apache#48) * add runtime storage fallback detection" * replace cast storage ex with cast storage impl Fm example (apache#45) * update csr slice logic to avoid confusion. add more exmaples. * add hint to module.update * more testcases(fallback) for sparse_nd * add to_csr() and to_rsp() method. More unit test (fallback now) * add fm test. fix lint * register sparse sgd under Optim.SGD * update dmlc-core submoduel * change indptr to _indptr temporarily. add const ref to fname fix lint fix lint; (apache#51) Guard gpu cast storage (apache#50) * Clean up * Fix typo Rearrange unit test files (apache#52) fix lint. add scipy for python_test. fix scipy.sparse import error. fix truediv for python3 fix travis test (apache#54) * remove pyc files * add verbose for travis nosetests cleanup some testing code and enums (apache#57) * update Makefile * refactor test_sparse_operator * change `default_storage` back to `default` * remove unused cpp tests port libsvm parser to mxnet as libsvm iter (apache#55) * copied csv iter to libsvm iter test libsvm iter draft handle round batch == false for csr batch loader code refactoring add get stype, shape interface to iiter separate class for sparse iter add missing file fix mem corruption' rename variables add comments also read label from libsvm add test. update docs. update submodule Conflicts: python/mxnet/sparse_ndarray.py * update submodule * fix lint * update test * revert naming change add benchmark scritp for dot (apache#59) * add benchmark scritp for dot add gpu option for bench add get_data funciton for benchmark print t_sparse, too; add comment change nnz to dnesity add backward * add comment update fm test (apache#62) introduce CSRNDarray and rowsparseNDarray to python frontend api (apache#58) * introduce CSRNDarray and rowsparseNDarray to python frontend api * temporarily disable fm_module test fix lint (apache#64) fix typo. disable libsvm io test (apache#65) Improve dot (apache#61) * Init checkin * Fix * Adjust dot parallelization methods * Set num_omp_threads for benchmark from command line * Fix omp thread number * Clean up * Add scipy as dot baseline * Fix format sparse_retain op (apache#66) * Initial checkin * Fix bugs * Add unit test for sparse_retain * Add example and modify test add storage cast for outputs that have non-default storage (apache#67) fix gpu build (apache#69) Fix test_sparse_retain python3 issue (apache#68) revert nnvm version * draft for sgd rsp rsp (apache#75) support sgd(rsp, rsp) support dot(csr, rsp) when rsp is full add ref to const ndarray params support sparse embedding with rsp weight' fix lint modify embedding backward to produce dense grad remove invalid_rid for rsp->dns remove previous embedding op changes pass sparse embedding test add STORAGE_TYPE_ASSIGN_CHECK remove backward storage infer * fix lint (apache#78) * fix lint (apache#79) * serial elemwise sum impl (apache#80) update module kvstore interface add other missing params and functions revert some interface changes revert some more changes reomve explicit casting for gradients on kvstore update Comm interface update fm example Conflicts: python/mxnet/model.py python/mxnet/ndarray.py * bug fix for initializing module with row_sparse weight (apache#81) * bug fix for initializing module with row_sparse weight * update log message * Sparse ndarray serialization and deserialization (apache#77) * Initial checkin * Add unit tests * Fix lint * Fix lint (apache#84) * Sgd with row_sparse weight, dns gradient (apache#83) * sgd rsp dns draft * support sgd_mom(rsp, dns, rsp) * update doc * remove cast storage for kv updater * code refactoring * update mshadow version (apache#88) * csr slice bug fix (apache#90) * benchmark dot code refactor (apache#87) * q^x6x add some code in benchmark * refactor * minor fixes * fix * lint fix * Add unit test (apache#91) * add unittest * minor fix * remove commented lines * change test func name * add test rsp * kvstore push row sparse (apache#93) * Add multi-thread cpu elemwise sum for rsps * Minor fix * Add flag to switch between serial and multi-thread kvstore push * Fix lint in sparse_ndarray.py * Revert "Fix lint in sparse_ndarray.py" This reverts commit d7225ec. * Fix ndarray init in copy(ctx) * Add env var to control the flow of serial/parallel reduce * Refactor * Fix copy ndarray bug * Fix lint * Refactor * Fix windows openmp build failure (apache#94) * update mshadow submoduel (apache#95) * Revert "update mshadow submoduel (apache#95)" (apache#96) This reverts commit 1a129e4. * Refactor sparse tensor code (apache#99) * Initial checkin test_sparse_ndarray passes * Fix test failure * Clean up * Clean up * Move init backend op to ndarray_utils * Fix lint * Eliminate circular dependency on headers * More refactor * Fix gpu build and consolidate Slice for dense and sparse * Clean up * More refactor * Clean up * Fix gpu build * Fix comment * fix pylint (apache#100) * Fix refactor sparse gpu test (apache#104) * Fix gpu build * Fix * Fix gpu test failure * change idx types from int32 to int64 (apache#101) Conflicts: python/mxnet/test_utils.py tests/python/unittest/test_sparse_operator.py update mshadow submodule fix extra quotes in test script change indptr type to int64 better err message for rsp" * revert LOG(DEBUG) change (apache#105) * fix undefined zeros in optimizer.py (apache#106) * move init dns zeros to init_op.h for kvstore to use (apache#107) * Refactor cast storage (apache#109) * Refactor cast_storage * Add cast_storage cc and cu files * Remove redundant comments * Replace std::accumulate with ParallelAccumulate * Clean up * Fix windows build * Rowsparse kv (apache#111) * update kvstore unit test Conflicts: tests/python/unittest/test_kvstore.py update model/module.py Conflicts: python/mxnet/model.py python/mxnet/module/module.py fix lint resolve conflict remove int keys in kvstore update cast to str function * fix failed dist_sync_kv test * bug fix in comm to ensure merged gradient is of the right type bug fix in comm * row sparse dist kvstore draft (push only) row_sparse pull * add ndarray row sparse shared mem constructor * code refactoring * add test for row_sparse weight bug fix for kv server slicing add async support rsolve race condition in kvstore * resolve error after reb ase * fix lint (apache#113) * rename some python funciton (apache#114) * _to_rsp * _to_csr. raise NotImplementedError * todense * fix lint (apache#115) enable libsvm uniit test (apache#6839) remove shared mem slice for csr add csr ndarray iter test make osx nose test verbose disable libsvm iter test Move InferAttr to mxnet from nnvm (apache#6830) * Move InferAttr to mxnet from nnvm Replace nnvm infer attr functions in c_api Initial checkin Clean up Remove nnvm namespace for FInferShape, FInferType, and FInferStorageType Add new interface for InferStorageType Revert "Remove nnvm namespace for FInferShape, FInferType, and FInferStorageType" This reverts commit 8aedf05. Fix and clean up Fix lint Add nnvm changes Change infer function interface to accept only rvalue reference of graph Clean up Flush commits to show up in PR Add error handling for storage type inference failure Update nnvm * Fix pylint Change idx type switch for aux data (apache#6860) * Change idx type switch for aux data * Add mshadow commit Sparse dot enhancement (apache#6842) * Initial checkin Initial checkin Fix sparse dot test Fix unitest and add fallback for sparse dot * Add benchmark code * Revert "Add benchmark code" This reverts commit be009fe. * Fix bug * Fix storage shape * Remove unnecessary test code * Use idx type switch Implement dot(csr, rsp)=dns and dot(csr.T, rsp)=rsp and refactor (apache#6902) * Initial checkin Add dot(csr.T, rsp)=rsp2 Add infer storage for dot(csr, rsp)=dns and dot(csr.T, rsp)=rsp2 * Fix comments * Replace std::lower_bound with own impl for gpu use too * Add time profiling * Revert "Add time profiling" This reverts commit 8f5bb98. * Move dot and batch_dot to a single file * Move dot gpu impl to a .cuh file * More refactor * Fix include error LibsvmIter fix (apache#6898) * fix bug in libsvm iter which causes mem corruption * add test for news dataset * fix wrong path in test * fix import error for urllib * update url * replace bz command with bz module Optimized gpu dot kernels (apache#6937) * pulled update to mshadow * mshadow update * added optimized gpu kernels for dot(csr,dns)=dns and dot(csr.T,dns)=dns, and unit test * added __syncwarp to vector kernel and reduced number of writes to shared memory Refactor sparse tensor code (apache#6955) * Save stype in frontend to avoid c-api call for stype * Change storage_type to stype * Revert "Change storage_type to stype" This reverts commit 90db7d1. * Revert "Revert "Change storage_type to stype"" This reverts commit 0932838. Move ndarray.py, sparse_ndarray.py, ndarray_utils.py, and _ndarray_internal to ndarrary folder More refactor Move elementwise sum for rsp to ndarray_function.cc Remove unnecessary import in ndarray module Fix pylint Remove redundant code Remove _stype from slots Fix cpp-package build error caused by the change to imperative invoke interface Use relative import Remove print line Rename _ndarray_internal.py to _internal.py * Relaunch test... minor bug fix in warp synchronous code (apache#7029) * move storage type vector from nnvm to mxnet (apache#7054) * move storage type vector from nnvm to mxnet * update nnvm * update nnvm * Improve copy sparse tensors (apache#7003) * Use cast_storage when copying ndarrays of different stypes on same context * Relaunch test * fix failed tests. add back 64bit support for dot fix lint * bug fix for IdentityComputeRsp * fix lint fix lint fix lint * add data partition for libsvm iter (apache#7027) * remove sparse embedding (apache#7165) * fix ndarray namespace * remove untested gpu operators (apache#7172) * skip sparse dot gpu tset. add sparse_nd_zeros gpu test * remove sparse_retain gpu Conflicts: tests/python/gpu/test_operator_gpu.py * Fix ndarray aux data issue (apache#7098) * Fix getting sparse ndarray data/aux_data issues * Add tests for func csr and row_sparse * Make get/set data/aux_data thread safe * Fix a bug * Fix typo and comment * More comments * Correct comment Conflicts: tests/python/gpu/test_operator_gpu.py * Support K-dimensional row-sparse tensor (apache#7179) * remove check for k dimensional rowsparse tensor * change var name for rsp sgd operator * add checks for sparse dot * bug fix for kdim rowsparse cast storage cpu * update IdentityLikeRhsComputeEx interface * remove set_storage_shape from ndarray. support elemwise_add with kdim row_sparse tensor * use get_with_shape instead of reshape * update according to comments Conflicts: src/operator/tensor/elemwise_unary_op.h * Improve sparse ndarray error message (apache#7181) * add test for broadcast_to * add comments Conflicts: python/mxnet/base.py * construct row_sparse ndarray for dist-async fix bug in rsp add rsp sync push race condition for push fix bug in rsp pull. refactor test cleanup comments refactor dist server fix lint fix storage shape issue with the new ndarray constructor data sharding draft; fix lint. add comment add support for zeros gradients use std::upper_bound/lower_bound remove special init function for rowsparse dist kvstore temporary support for inplace operators for sparse add test. fix return type store kRowSparseNDArray in kv server remove fcomp_ex sgd with dns weight and rsp gradient bug fix in sparse retain sparse pull c_api revise rowsparse pull api use engine to compute unique to ensure thread safety add rowsparse pull to dist-kv fix lint add example for rsp_pull remove name2idx; add sparse_pull_dict param to module fix unit test and c rowid conversion support str key type in kvstore (apache#6765) * update kvstore unit test * update model/module.py * fix lint * remove int keys in kvstore * update cast to str function * remove _cast_to_str_keys * fix lint * always cast to str Conflicts: include/mxnet/c_api.h include/mxnet/kvstore.h python/mxnet/kvstore.py python/mxnet/model.py python/mxnet/module/module.py src/c_api/c_api.cc src/kvstore/kvstore_local.h tests/python/unittest/test_kvstore.py update module API for other submodules update stypes in kvstore after refactoring change type of size from size_t to int64_t add sparse linear regression example remove sparse_pull_dict from module fix init_optim for seq_module. update sparse example resolve conflict for binary add rsp rsp Conflicts: python/mxnet/kvstore.py tests/python/unittest/test_kvstore.py * fix DotCsrRspRspImpl error message (apache#7191) * GPU implementation of cast_storage (dense to csr) (apache#7081) * Added gpu implementation for cast_storage dense to csr, unit tests, and benchmark. Additionally, cast_storage interface change to accommodate the need of temporary storage in cuda kernels. * fixed whitespace * minor unittest update * removed whitespace * add cast storage benchmark params info Conflicts: tests/python/gpu/test_operator_gpu.py * Sparse square sum (apache#7206) * Add square_sum op * Add unit test and fix check_numeric_gradient * Add .cu file and example * Fix lint * Remove gpu registration * Use square_sum in test_module_fm * Modify and Add documentation for mx.nd.zeros (apache#7197) * Modify and Add documentation for mx.nd.zeros * Change context to cpu * Change stype to optional * Change ordering and remove optional for _zeros_sparse_ndarray * Expose kWriteInplace for imperative execution (fcompute_ex and fstatefulcompute_ex) (apache#133) * expose kWriteInplace to FComputeEx and FStatefulComputeEx * refactor ccode * remove duplicated test * Operator add_n for row sparse ndarrays (apache#7244) * Add add_n op for row-sparse ndarrays and identity FComputeEx * Fix bug in square_sum * Remove test_cast_storage_ex from gpu test since it's not implemented yet * Fix according to the cr Conflicts: src/operator/tensor/elemwise_sum.cc src/operator/tensor/elemwise_unary_op.cc tests/python/gpu/test_operator_gpu.py resolve conflict * GPU implementation of cast_storage (dense to rsp) (apache#7223) * CastStorageDnsRsp GPU Implementation * updating function doc and some variable types and names * adding cuda_get_device_prop() util function * added rand_shape function for n-dimensional tensors * updated cast storage unit test * added dns_to_rsp to cast storage benchmark script * removing redundant unit test * fix lint * minor change in benchmark script * fix lint * correct function description * change storage_type to stype * changed scope of using namespaces * changed variable types from index_t to dim_t * resolve merge conflict in ndarray.load * Improve StatefulOp/FCompute storage fallback (apache#134) * test for fcomp fallback add storage fallback test and optimize fallback logic rename function, add comments use std size() * add autograd test with sparse inputs * update sparse ndarray api (apache#139) * support mx.nd.empty for sparse ndarray Change SparseNDArray to BaseSparseNDArray support mx.nd.array with BaseSparseNDArray inputs. Update documentation with explicit subclasses of NDArrays Conflicts: python/mxnet/ndarray/__init__.py python/mxnet/ndarray/ndarray.py python/mxnet/ndarray/sparse_ndarray.py tests/python/unittest/test_sparse_ndarray.py * fix print msg in test * Handle ograd_stype='row_sparse' for square_sum backward (apache#143) * Add one kernel for square_sum backward pass to take rsp ograd * Add kNullOp and change to use type_assign in infer stype fallback * Sparse retain improvement (apache#138) * Add one more kernel for sparse retain * Fix compile * Change STORAGE_TYPE_ASSIGN_CHECK to type_assign for fallback * Fix * Add gpu compile * ignoring variables in SimpleBind that is used on python's sparse branch for now. (apache#135) * add bias term to fm test (apache#145) * update ndarray.nd, remove `invoke` from excluded members (apache#137) remove __weakref__ from SparseNDArray add data indice to doc revert dlpack update revert mxdoc changes move methods from BaseSparseNDarray to csrndarray and rwosparse ndarray * support storage fallback with mutable inputs (apache#147) * include mutatable inputs in storage fallback. refactor executor add fallback test for rms prop and adam fix lint fix lint fix test in optimizer * update according to comments * fix unit tests * fix gpu compilation err * Code changes based on reviews (apache#144) * code changes according to review comments remove executor debug. add doc to optimizer update sparse sgd test add dtype option to rand_sparse_ndarray * overhauled reqs for sparse operators * patch FCompExFallback with mutable inputs. update test_optimizer with more fallback cases * change executor debug macro to env var * add comment * update doc * change ndarray.aux_shape() to return const reference * remove todense to_rsp to_csr. replace with tostype * replace manual calls to cast_storage with tostype * disable gpu fallback test for optimizer * fix lint * add backward pass for cast_storage. refactor cast_storage test * rand_sparse_ndarray bug fix * fix cast_storage for gpu * disable csr test for fp16 * update row sparse ndarray doc * update doc * small edits according to reviews (apache#151) * fix lint (apache#152) * add license to all new files in sparse brnach (apache#154) * Allocate temp data on the fly for some casting operations (apache#149) * fix utf8 encoding in sparse ndarray * Extending the GPU dot operator (apache#7226) * Added GPU DotCsrRspDnsImpl declaration and TODOs * cleaning up function doc, variable types, and code-style * minor bug fixes * enable GPU dot(csr,rsp)=dns unit test * extend sparse dot unit test * adding GPU impl of DotCsrRspDns and its kernels * add TODO * changed variable types from index_t to dim_t * fix function description * added DotCsrRspRspImpl and its kernels (baseline, functionality) * added DotCsrDnsRspImpl and its kernels (baseline, functionality); plus code documentation * refactored dot benchmark * optimized DotCsrTransDnsRsp GPU kernel * change of dot impl interface to include OpContext, for temp storage * removing __device__ flag from CPU kernels * minor fixes and changing variable data types * minor fixes based on code reviews Conflicts: benchmark/python/sparse_op.py tests/python/gpu/test_operator_gpu.py tests/python/unittest/test_sparse_operator.py * Add get_synthetic_dataset function to util (apache#146) * Add get_synthetic_datasets * Move to test_utils * Remove _get_uniform_dataset * Move validation to its own function * Refactor the validation code for csr generation * Make test_powerlaw a nested function * Change SparseNDArray to CSRNDArray * Merge with dtype specific changes in test_utils * temporary fix for batch norm storage fallback (apache#156) * support random_uniform/normal/gamma with row_sparse output (apache#155) * add support for initilazer with rowsparse output * add scalar assignment to row_sparse * add setitem test to gpu * Revert "add scalar assignment to row_sparse" This reverts commit 8aef7a5. * Revert "add setitem test to gpu" This reverts commit 3b969ac. * Square sum backward support one more case (apache#161) * Add documentation for sparse ops (apache#148) * draft doc for sparse op * add more stype doc for operators * add doc for cast_storage * see also cast_storage. remove base sparse ndarray. fix aux_types comemtn * grammar / spelling fix * A few fixes (apache#163) * fix batch norm gpu kernel. register random operators on gpu * register sparse random op on gpu, too * Minor fixes sparse ops (apache#160) * change CPU kernel inline directives, data types, and function doc * update dot dtype switch to use 32 and 64bit floating point only * use type_assign instead of STORAGE_TYPE_ASSIGN_CHECK * added tensor_util-inl.cuh file for common tensor operator GPU kernels * sparse Adam optimizer (apache#164) * add sparse adam * register gpu op * add comments * cr comments * kvstore.row_sparse_pull for GPU and end-to-end benchmark: CPU vs. multi-GPUs (apache#150) * Add gpu support for BroadcastRowSparse * Fix bugs * Add benchmark script * Increase output dim size * Update weight on CPU using single GPU for sparse tensors * More fix * Optimize sparse_retain for special case * Change row sparse pull locations * Avoid sparse retain on cpu if possible * Use acc for metric * Fix misc * fix bug in adam update (apache#167) fix a bug in adam update * change sparse example from regression to classification (apache#165) * fix python import (apache#166) * Add waitall to sparse_end2end.py (apache#169) * Add waitall() * Add dummy metric option * Add header license * Dot script changes (apache#159) * Add get_synthetic_datasets * Move to test_utils * Remove _get_uniform_dataset * Move validation to its own function * Refactor the validation code for csr generation * Make test_powerlaw a nested function * Change SparseNDArray to CSRNDArray * Refactoring changes to dot.py * Fix mxnet test_utils changes * Remove pdb statement * Add distribution parameter * Refactor benchmarking script * Remove unused code * Make style changes and remove unused code * Change typo in comment * Add transpose support * Change typo * 4 decimal points needed for density * Add rsp support for real datasets * Correct variable name mini_file_name * Move wait_to_read outside if * Seperate out scipy and mxnet logic in bench_dot * Fix lhs_trans issue * Move transpose outside measure_cost * Compute transpose inside measure_cost * Remove unused variables * Transpose only if trans_lhs (apache#171) * fix default val for distribution (apache#172) * fix lint (apache#175) * avoid cast_storage in dist-kvstore-server (apache#174) * avoid cast_storage in dist-kvstore-server * add stream arg to mshadow;;copy * fix copy order * Add sparse namespace to ndarray and symbol (apache#177) * Register dot, cast_storage, and sparse_retain under mxnet.ndarray.sparse * Add sparse to symbol namespace * Delete commented code * mv sparse_ndarray.py sparse.py * Clean up * Change docstring * changes based on code reviews (apache#176) * remove scipy dependency * move kvstore checks to backned * add const to lambda * temp fix to ndarray.md (apache#178) * Fix sparse namespace pylint (apache#179) * add comments and error msg (apache#181) * add clarification for csr (apache#182) * add clarification for csr * cr comments * revert change in test util (apache#183) * fix amalgamation (apache#184) * fix lint

piiswrong reviewed Apr 12, 2017

View reviewed changes

reminisce reviewed Apr 13, 2017

View reviewed changes

reminisce reviewed Apr 18, 2017

View reviewed changes

eric-haibin-lin changed the base branch from master to sparse April 18, 2017 23:53

eric-haibin-lin force-pushed the master branch 2 times, most recently from 969081f to dbb304b Compare April 23, 2017 06:10

eric-haibin-lin force-pushed the master branch from eb44e7d to cb8ca5f Compare May 1, 2017 21:56

eric-haibin-lin mentioned this pull request May 3, 2017

force infer storage on backward pass (#1) dmlc/nnvm#113

Merged

eric-haibin-lin force-pushed the master branch from 5fe9d86 to cb8ca5f Compare May 4, 2017 01:57

eric-haibin-lin force-pushed the master branch 2 times, most recently from 59e7d80 to d6e6d98 Compare May 15, 2017 18:06

piiswrong changed the base branch from sparse to master May 15, 2017 18:45

eric-haibin-lin changed the base branch from master to sparse May 15, 2017 18:48

eric-haibin-lin force-pushed the master branch from 77132fc to 9db62c0 Compare May 24, 2017 20:10

reminisce mentioned this pull request May 26, 2017

Fix cudnn_deconv not checking no_bias #6456

Merged

eric-haibin-lin force-pushed the master branch from 5c7f715 to 8652043 Compare May 27, 2017 05:44

piiswrong changed the base branch from sparse to master May 31, 2017 20:07

eric-haibin-lin changed the base branch from master to sparse June 1, 2017 21:08

piiswrong changed the base branch from sparse to master June 7, 2017 18:04

eric-haibin-lin changed the base branch from master to sparse June 7, 2017 19:14

eric-haibin-lin force-pushed the master branch 3 times, most recently from 8c77488 to ac54762 Compare June 9, 2017 01:33

eric-haibin-lin changed the base branch from sparse to master June 9, 2017 01:34

eric-haibin-lin added 2 commits June 10, 2017 16:12

change idx types from int32 to int64 (#101)

a9612e3

Conflicts: python/mxnet/test_utils.py tests/python/unittest/test_sparse_operator.py update mshadow submodule fix extra quotes in test script change indptr type to int64 better err message for rsp"

eric-haibin-lin closed this Jun 21, 2017

eric-haibin-lin reopened this Jun 21, 2017

piiswrong changed the base branch from sparse to master June 21, 2017 17:47

revert LOG(DEBUG) change (#105)

c134687

eric-haibin-lin changed the base branch from master to sparse June 21, 2017 22:45

fix undefined zeros in optimizer.py (#106)

1914471

eric-haibin-lin commented Jun 22, 2017

View reviewed changes

piiswrong force-pushed the sparse branch from 3ceb6d2 to d75ef8e Compare June 22, 2017 23:01

eric-haibin-lin and others added 6 commits June 22, 2017 16:03

move init dns zeros to init_op.h for kvstore to use (#107)

ad8e74c

Refactor cast storage (#109)

ac554ac

* Refactor cast_storage * Add cast_storage cc and cu files * Remove redundant comments * Replace std::accumulate with ParallelAccumulate * Clean up * Fix windows build

fix lint (#113)

4d4fbc7

rename some python funciton (#114)

f8556cc

* _to_rsp * _to_csr. raise NotImplementedError * todense

fix lint (#115)

cb5deca

piiswrong merged commit 7b2ef68 into apache:sparse Jun 26, 2017

reminisce mentioned this pull request Jul 7, 2017

Refactor sparse tensor code #6955

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[WIP] Sparse Tensor #5800

[WIP] Sparse Tensor #5800

eric-haibin-lin commented Apr 12, 2017 •

edited

Loading

piiswrong Apr 12, 2017

eric-haibin-lin Apr 12, 2017

piiswrong Apr 12, 2017

piiswrong Apr 12, 2017

piiswrong Apr 12, 2017

eric-haibin-lin Apr 12, 2017

piiswrong Apr 12, 2017

eric-haibin-lin Apr 12, 2017

piiswrong Apr 12, 2017

eric-haibin-lin Apr 12, 2017

reminisce Apr 13, 2017

eric-haibin-lin Apr 13, 2017

reminisce Apr 13, 2017

eric-haibin-lin Apr 13, 2017

reminisce Apr 13, 2017

eric-haibin-lin Apr 13, 2017

reminisce Apr 13, 2017

reminisce Apr 18, 2017

eric-haibin-lin Apr 18, 2017

piiswrong commented May 14, 2017

eric-haibin-lin left a comment

eric-haibin-lin Jun 22, 2017

eric-haibin-lin Jun 22, 2017

eric-haibin-lin Jun 22, 2017

eric-haibin-lin Jun 22, 2017

eric-haibin-lin Jun 22, 2017

eric-haibin-lin Jun 22, 2017

eric-haibin-lin Jun 22, 2017

eric-haibin-lin Jun 22, 2017

eric-haibin-lin Jun 22, 2017

eric-haibin-lin Jun 22, 2017

eric-haibin-lin commented Jun 22, 2017 •

edited

Loading

eric-haibin-lin commented Jun 22, 2017 •

edited

Loading

[WIP] Sparse Tensor #5800

[WIP] Sparse Tensor #5800

Conversation

eric-haibin-lin commented Apr 12, 2017 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

piiswrong commented May 14, 2017

eric-haibin-lin left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

eric-haibin-lin commented Jun 22, 2017 • edited Loading

A list of frontend changes in this PR

eric-haibin-lin commented Jun 22, 2017 • edited Loading

eric-haibin-lin commented Apr 12, 2017 •

edited

Loading

eric-haibin-lin commented Jun 22, 2017 •

edited

Loading

eric-haibin-lin commented Jun 22, 2017 •

edited

Loading