tests pass

examples/summarization/bertabs/modeling_bertabs.py

@@ -15,11 +15,14 @@
 # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 # IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 # FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
-# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
-# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+
+
 # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 # SOFTWARE.
 import copy
+
+# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
 import math
 
 import numpy as np

src/transformers/modeling_bart.py

@@ -640,9 +640,10 @@ class SelfAttention(nn.Module):
             reshaped = key_padding_mask.unsqueeze(1).unsqueeze(2).to(torch.bool)
             attn_weights = attn_weights.masked_fill(reshaped, float("-inf"))
             attn_weights = attn_weights.view(bsz * self.num_heads, tgt_len, src_len)
-        attn_weights_float = F.softmax(attn_weights, dim=-1, dtype=torch.float32)
-        attn_weights = attn_weights_float.type_as(attn_weights)
+        attn_weights_float = F.softmax(attn_weights, dim=-1)
         attn_probs = F.dropout(attn_weights_float, p=self.dropout, training=self.training,)
+        attn_weights = attn_weights_float.type_as(attn_weights)
+
         assert v is not None
         attn_output = torch.bmm(attn_probs, v)
         assert attn_output.size() == (bsz * self.num_heads, tgt_len, self.head_dim)
@@ -696,8 +697,12 @@ class SelfAttention(nn.Module):
         elif prev_key_padding_mask is not None:
             filler = torch.zeros(batch_size, src_len - prev_key_padding_mask.size(1))
             if prev_key_padding_mask.is_cuda:
-                filler = filler.cuda()
+                filler = filler.to(prev_key_padding_mask.device)
             new_key_padding_mask = torch.cat([prev_key_padding_mask.float(), filler.float()], dim=1)
+            print(new_key_padding_mask.device, new_key_padding_mask.dtype)
+            import ipdb
+
+            ipdb.set_trace()
         elif key_padding_mask is not None:
             filler = torch.zeros(batch_size, src_len - key_padding_mask.size(1))
             if key_padding_mask.is_cuda:

tests/test_modeling_bart.py

@@ -243,15 +243,15 @@ class BartHeadTests(unittest.TestCase):
             decoder_ffn_dim=32,
             max_position_embeddings=48,
         )
-        lm_model = BartForMaskedLM(config)
-        context = torch.Tensor([[71, 82, 18, 33, 46, 91, 2], [68, 34, 26, 58, 30, 2, 1]]).long()
-        summary = torch.Tensor([[82, 71, 82, 18, 2], [58, 68, 2, 1, 1]]).long()
+        lm_model = BartForMaskedLM(config).to(torch_device)
+        context = _long_tensor([[71, 82, 18, 33, 46, 91, 2], [68, 34, 26, 58, 30, 2, 1]])
+        summary = _long_tensor([[82, 71, 82, 18, 2], [58, 68, 2, 1, 1]])
         logits, enc_features = lm_model.forward(input_ids=context, decoder_input_ids=summary)
         expected_shape = (*summary.shape, config.vocab_size)
         self.assertEqual(logits.shape, expected_shape)
 
     def test_generate_beam_search(self):
-        input_ids = torch.Tensor([[71, 82, 2], [68, 34, 2]]).long()
+        input_ids = _long_tensor([[71, 82, 2], [68, 34, 2]])
         config = BartConfig(
             vocab_size=self.vocab_size,
             d_model=24,
@@ -264,7 +264,7 @@ class BartHeadTests(unittest.TestCase):
             max_position_embeddings=48,
             output_past=True,
         )
-        lm_model = BartForMaskedLM(config)
+        lm_model = BartForMaskedLM(config).to(torch_device)
         lm_model.eval()
 
         new_input_ids = lm_model.generate(
@@ -294,6 +294,13 @@ class BartHeadTests(unittest.TestCase):
             bart_toks = tokenizer.encode(ex, return_tensors="pt")
             _assert_tensors_equal(desired_result.long(), bart_toks, prefix=ex)
 
+    @unittest.skipIf(torch_device == "cpu", "Cant do half precision")
+    def test_generate_fp16(self):
+        config, input_ids, batch_size = self._get_config_and_data(output_past=True)
+        attention_mask = input_ids.ne(1)
+        lm_model = BartForMaskedLM(config).eval().to(torch_device).half()
+        lm_model.generate(input_ids, attention_mask)
+
 
 def _assert_tensors_equal(a, b, atol=1e-12, prefix=""):
     """If tensors not close, or a and b arent both tensors, raise a nice Assertion error."""