fix: change to enumerate by sentence

Author: chenwhql

paddle/fluid/operators/sequence_enumerate_op.cc

@@ -72,14 +72,14 @@ The values of the last insufficient part areall filled with the input pad_value.
 Case 1:
   Input:
     X.lod = [[0, 3, 5]]
-    X.data = [1, 2, 3, 4, 5]
+    X.data = [[1], [2], [3], [4], [5]]
     X.dims = [5, 1]
   Attrs:
     win_size = 2
     pad_value = 0
   Output:
     Out.lod = [[0, 3, 5]]
-    Out.data = [[1, 2], [2, 3], [3, 4], [4, 5], [0, 0]]
+    Out.data = [[1, 2], [2, 3], [3, 0], [4, 5], [5, 0]]
     Out.dims = [5, 2]
 
 )DOC");

paddle/fluid/operators/sequence_enumerate_op.cu

@@ -23,15 +23,23 @@ using platform::PADDLE_CUDA_NUM_THREADS;
 using LoDTensor = framework::LoDTensor;
 
 template <typename T>
-__global__ void CalcOutPut(const T* in_data, const int64_t in_len,
-                           const int64_t win_size, const int64_t pad_value,
-                           T* out_data) {
+__global__ void CalcOutPut(const T* in_data, const size_t* in_lod,
+                           const size_t lod_len, const int64_t win_size,
+                           const int64_t pad_value, T* out_data) {
   int index = blockIdx.x * blockDim.x + threadIdx.x;
-  if (index < in_len) {
+  if (index < in_lod[lod_len - 1]) {
+    int end_idx = 0;
+    // Get LoD interval of index
+    for (int i = 1; i < lod_len; ++i) {
+      if (index < in_lod[i]) {
+        end_idx = in_lod[i];
+        break;
+      }
+    }
     for (size_t i = 0; i < win_size; ++i) {
       int word_pos = index + i;
       out_data[index * win_size + i] =
-          word_pos < in_len ? in_data[word_pos] : pad_value;
+          word_pos < end_idx ? in_data[word_pos] : pad_value;
     }
   }
 }
@@ -54,13 +62,16 @@ class SequenceEnumerateOpCUDAKernel : public framework::OpKernel<T> {
 
     /* Generate enumerate sequence set */
     auto stream = context.cuda_device_context().stream();
+    auto lod0 = in_lod[0];
     auto in_len = in->numel();
     auto in_data = in->data<T>();
     auto out_data = out->mutable_data<T>(context.GetPlace());
+    // Copy LoD to GPU
+    const size_t* dev_in_lod_ptr = lod0.CUDAData(context.GetPlace());
     // Calc output tensor
     CalcOutPut<<<(in_len - 1) / PADDLE_CUDA_NUM_THREADS + 1,
                  PADDLE_CUDA_NUM_THREADS, 0, stream>>>(
-        in_data, in_len, win_size, pad_value, out_data);
+        in_data, dev_in_lod_ptr, lod0.size(), win_size, pad_value, out_data);
   }
 };
 

paddle/fluid/operators/sequence_enumerate_op.h

@@ -37,14 +37,16 @@ class SequenceEnumerateKernel : public framework::OpKernel<T> {
         "The actual input data's size mismatched with LoD information.");
 
     // Generate enumerate sequence set
-    auto seq_length = in_dims[0];
+    auto lod0 = in_lod[0];
     auto in_data = in->data<T>();
     auto out_data = out->mutable_data<T>(context.GetPlace());
-    for (int idx = 0; idx < seq_length; ++idx) {
-      for (int word_idx = 0; word_idx < win_size; ++word_idx) {
-        int word_pos = idx + word_idx;
-        out_data[win_size * idx + word_idx] =
-            word_pos < seq_length ? in_data[word_pos] : pad_value;
+    for (size_t i = 0; i < lod0.size() - 1; ++i) {
+      for (size_t idx = lod0[i]; idx < lod0[i + 1]; ++idx) {
+        for (int word_idx = 0; word_idx < win_size; ++word_idx) {
+          size_t word_pos = idx + word_idx;
+          out_data[win_size * idx + word_idx] =
+              word_pos < lod0[i + 1] ? in_data[word_pos] : pad_value;
+        }
       }
     }
   }

python/paddle/fluid/layers/nn.py

@@ -5534,14 +5534,14 @@ def sequence_enumerate(input, win_size, pad_value, name=None):
     Case 1:
       Input:
         X.lod = [[0, 3, 5]]
-        X.data = [1, 2, 3, 4, 5]
+        X.data = [[1], [2], [3], [4], [5]]
         X.dims = [5, 1]
       Attrs:
         win_size = 2
         pad_value = 0
       Output:
         Out.lod = [[0, 3, 5]]
-        Out.data = [[1, 2], [2, 3], [3, 4], [4, 5], [0, 0]]
+        Out.data = [[1, 2], [2, 3], [3, 0], [4, 5], [5, 0]]
         Out.dims = [5, 2]
 
     Args:
@@ -5567,7 +5567,7 @@ def sequence_enumerate(input, win_size, pad_value, name=None):
         attrs={'win_size': win_size,
                'pad_value': pad_value})
 
-    
+
 def sequence_mask(x, maxlen=None, dtype='int64', name=None):
     """
     **SequenceMask Layer**

python/paddle/fluid/tests/unittests/test_sequence_enumerate_op.py

@@ -19,16 +19,20 @@
 from op_test import OpTest
 
 
-def sequence_enumerate(input_seq, win_size, pad_value):
+def sequence_enumerate(input_seq, in_lod, win_size, pad_value):
+    lod0 = [0]
+    for i in range(0, len(in_lod[0])):
+        lod0.append(lod0[i] + in_lod[0][i])
     out_seq = []
-    for idx in range(0, len(input_seq)):
-        single_seq = []
-        for word_idx in range(win_size):
-            word_pos = idx + word_idx
-            dat = input_seq[word_pos] if word_pos < len(input_seq) \
+    for i in range(0, len(lod0) - 1):
+        for idx in range(lod0[i], lod0[i + 1]):
+            single_seq = []
+            for word_idx in range(win_size):
+                word_pos = idx + word_idx
+                dat = input_seq[word_pos] if word_pos < lod0[i+1] \
                     else pad_value
-            single_seq.append(dat)
-        out_seq.append(single_seq)
+                single_seq.append(dat)
+            out_seq.append(single_seq)
     return out_seq
 
 
@@ -48,7 +52,8 @@ def init_test_case(self):
         self.lod = [[9, 4, 11, 6]]
         self.win_size = 2
         self.pad_value = 0
-        out_seq = sequence_enumerate(self.in_seq, self.win_size, self.pad_value)
+        out_seq = sequence_enumerate(self.in_seq, self.lod, self.win_size,
+                                     self.pad_value)
         self.out_seq = np.array(out_seq).astype("int32")
 
 
@@ -58,7 +63,8 @@ def init_test_case(self):
         self.lod = [[9, 4, 11, 6]]
         self.win_size = 2
         self.pad_value = 0
-        out_seq = sequence_enumerate(self.in_seq, self.win_size, self.pad_value)
+        out_seq = sequence_enumerate(self.in_seq, self.lod, self.win_size,
+                                     self.pad_value)
         self.out_seq = np.array(out_seq).astype("int64")
 
 
@@ -68,7 +74,8 @@ def init_test_case(self):
         self.lod = [[9, 4, 11, 6]]
         self.win_size = 30
         self.pad_value = 0
-        out_seq = sequence_enumerate(self.in_seq, self.win_size, self.pad_value)
+        out_seq = sequence_enumerate(self.in_seq, self.lod, self.win_size,
+                                     self.pad_value)
         self.out_seq = np.array(out_seq).astype("int32")