adapt code

setup.py

@@ -71,17 +71,17 @@ extras["sklearn"] = ["scikit-learn"]
 # keras2onnx and onnxconverter-common version is specific through a commit until 1.7.0 lands on pypi
 extras["tf"] = [
     "tensorflow",
-    "onnxconverter-common",
-    "keras2onnx"
-    # "onnxconverter-common @ git+git://github.com/microsoft/onnxconverter-common.git@f64ca15989b6dc95a1f3507ff6e4c395ba12dff5#egg=onnxconverter-common",
-    # "keras2onnx @ git+git://github.com/onnx/keras-onnx.git@cbdc75cb950b16db7f0a67be96a278f8d2953b48#egg=keras2onnx"
+    # "onnxconverter-common",
+    # "keras2onnx"
+    "onnxconverter-common @ git+git://github.com/microsoft/onnxconverter-common.git@f64ca15989b6dc95a1f3507ff6e4c395ba12dff5#egg=onnxconverter-common",
+    "keras2onnx @ git+git://github.com/onnx/keras-onnx.git@cbdc75cb950b16db7f0a67be96a278f8d2953b48#egg=keras2onnx"
 ]
 extras["tf-cpu"] = [
     "tensorflow-cpu",
-    "onnxconverter-common",
-    "keras2onnx"
-    # "onnxconverter-common @ git+git://github.com/microsoft/onnxconverter-common.git@f64ca15989b6dc95a1f3507ff6e4c395ba12dff5#egg=onnxconverter-common",
-    # "keras2onnx @ git+git://github.com/onnx/keras-onnx.git@cbdc75cb950b16db7f0a67be96a278f8d2953b48#egg=keras2onnx"
+    # "onnxconverter-common",
+    # "keras2onnx"
+    "onnxconverter-common @ git+git://github.com/microsoft/onnxconverter-common.git@f64ca15989b6dc95a1f3507ff6e4c395ba12dff5#egg=onnxconverter-common",
+    "keras2onnx @ git+git://github.com/onnx/keras-onnx.git@cbdc75cb950b16db7f0a67be96a278f8d2953b48#egg=keras2onnx"
 ]
 extras["torch"] = ["torch"]
 
@@ -93,8 +93,8 @@ extras["testing"] = ["pytest", "pytest-xdist", "timeout-decorator", "psutil"]
 extras["docs"] = ["recommonmark", "sphinx", "sphinx-markdown-tables", "sphinx-rtd-theme==0.4.3", "sphinx-copybutton"]
 extras["quality"] = [
     "black",
-    "isort",
-    # "isort @ git+git://github.com/timothycrosley/isort.git@e63ae06ec7d70b06df9e528357650281a3d3ec22#egg=isort",
+    # "isort",
+    "isort @ git+git://github.com/timothycrosley/isort.git@e63ae06ec7d70b06df9e528357650281a3d3ec22#egg=isort",
     "flake8",
 ]
 extras["dev"] = extras["testing"] + extras["quality"] + ["mecab-python3<1", "scikit-learn", "tensorflow", "torch"]

src/transformers/generation_tf_utils.py

@@ -347,6 +347,10 @@ class TFGenerationMixin:
             encoder_outputs = None
             cur_len = shape_list(input_ids)[-1]
 
+        assert (
+            cur_len < max_length
+        ), f"The context has {cur_len} number of tokens, but `max_length` is only {max_length}. Please make sure that `max_length` is bigger than the number of tokens, by setting either `generate(max_length=...,...)` or `config.max_length = ...`"
+
         if num_beams > 1:
             output = self._generate_beam_search(
                 input_ids,

src/transformers/generation_utils.py

@@ -428,6 +428,10 @@ class GenerationMixin:
             encoder_outputs = None
             cur_len = input_ids.shape[-1]
 
+        assert (
+            cur_len < max_length
+        ), f"The context has {cur_len} number of tokens, but `max_length` is only {max_length}. Please make sure that `max_length` is bigger than the number of tokens, by setting either `generate(max_length=...,...)` or `config.max_length = ...`"
+
         if num_beams > 1:
             output = self._generate_beam_search(
                 input_ids,

src/transformers/modeling_reformer.py

@@ -373,7 +373,7 @@ class LSHSelfAttention(nn.Module, EfficientAttentionMixin):
             # use cached buckets for backprop only
             if buckets is None:
                 # hash query key vectors into buckets
-                buckets = self._hash_vectors(query_key_vectors, num_hashes)
+                buckets = self._hash_vectors(query_key_vectors, num_hashes, attention_mask)
 
             assert (
                 int(buckets.shape[-1]) == num_hashes * sequence_length
@@ -460,7 +460,7 @@ class LSHSelfAttention(nn.Module, EfficientAttentionMixin):
 
         return LSHSelfAttentionOutput(hidden_states=out_vectors, attention_probs=attention_probs, buckets=buckets)
 
-    def _hash_vectors(self, vectors, num_hashes):
+    def _hash_vectors(self, vectors, num_hashes, attention_mask):
         batch_size = vectors.shape[0]
 
         # See https://arxiv.org/pdf/1509.02897.pdf
@@ -487,14 +487,20 @@ class LSHSelfAttention(nn.Module, EfficientAttentionMixin):
 
         if self.hash_seed is not None:
             # for determinism
-            torch.manual_seed(self.hash_seed)
+            rotations_shape = (vectors.shape[-1], num_hashes, rotation_size // 2)
+            #            torch.manual_seed(self.hash_seed)
+            np.random.seed(self.hash_seed)
+            random_rotations = torch.tensor(
+                np.random.normal(size=rotations_shape), dtype=vectors.dtype, device=vectors.device,
+            )
+            rotated_vectors = torch.einsum("bmtd,dhr->bmhtr", vectors, random_rotations)
+        else:
+            rotations_shape = (self.num_attention_heads, vectors.shape[-1], num_hashes, rotation_size // 2)
+            # create a random self.attention_head_size x num_hashes x num_buckets/2
+            random_rotations = torch.randn(rotations_shape, device=vectors.device, dtype=vectors.dtype)
 
-        rotations_shape = (self.num_attention_heads, vectors.shape[-1], num_hashes, rotation_size // 2)
-        # create a random self.attention_head_size x num_hashes x num_buckets/2
-        random_rotations = torch.randn(rotations_shape, device=vectors.device, dtype=vectors.dtype)
-
-        # Output dim: Batch_Size x Num_Attn_Heads x Num_Hashes x Seq_Len x Num_Buckets/2
-        rotated_vectors = torch.einsum("bmtd,mdhr->bmhtr", vectors, random_rotations)
+            # Output dim: Batch_Size x Num_Attn_Heads x Num_Hashes x Seq_Len x Num_Buckets/2
+            rotated_vectors = torch.einsum("bmtd,mdhr->bmhtr", vectors, random_rotations)
 
         if isinstance(self.num_buckets, int) or len(self.num_buckets) == 1:
             rotated_vectors = torch.cat([rotated_vectors, -rotated_vectors], dim=-1)
@@ -514,6 +520,15 @@ class LSHSelfAttention(nn.Module, EfficientAttentionMixin):
 
                 cur_product = cur_product * bucket_factor
 
+        if attention_mask is not None:
+            # add an extra bucket for padding tokens only
+            num_buckets = num_buckets + 1
+            # assign padding tokens extra bucket
+            buckets_mask = attention_mask.to(torch.uint8)[:, None, None, :].expand(buckets.shape)
+            buckets = torch.where(
+                buckets_mask, buckets, torch.tensor(num_buckets - 1, dtype=torch.long, device=buckets.device)
+            )
+
         # buckets is now (Batch_size x Num_Attn_Heads x Num_Hashes x Seq_Len).
         # Next we add offsets so that bucket numbers from different hashing rounds don't overlap.
         offsets = torch.arange(num_hashes, device=vectors.device)
@@ -614,7 +629,9 @@ class LSHSelfAttention(nn.Module, EfficientAttentionMixin):
             self_mask_value = self.self_mask_value_float32
             mask_value = self.mask_value_float32
 
-        mask = self._compute_attn_mask(query_bucket_idx, key_value_bucket_idx, attention_mask, sequence_length)
+        mask = self._compute_attn_mask(
+            query_bucket_idx, key_value_bucket_idx, attention_mask, query_key_dots.shape, sequence_length
+        )
 
         if mask is not None:
             query_key_dots = torch.where(mask, query_key_dots, mask_value)
@@ -669,7 +686,7 @@ class LSHSelfAttention(nn.Module, EfficientAttentionMixin):
 
         return out_vectors, logits, attention_probs
 
-    def _compute_attn_mask(self, query_indices, key_indices, attention_mask, sequence_length):
+    def _compute_attn_mask(self, query_indices, key_indices, attention_mask, query_key_dot_shape, sequence_length):
         mask = None
 
         # Causal mask
@@ -680,32 +697,20 @@ class LSHSelfAttention(nn.Module, EfficientAttentionMixin):
         # IMPORTANT: official trax code does not use a mask for LSH Atttention. Not sure why.
         if attention_mask is not None:
             # if chunked attention, the attention mask has to correspond to LSH order
+            attention_mask = attention_mask.to(torch.uint8)[:, None, :]
             if sequence_length > self.chunk_length:
-                attention_mask = attention_mask.to(torch.uint8)[:, None, None, :]
                 # expand attn_mask to fit with key_value_bucket_idx shape
+                attention_mask = attention_mask[:, None, :]
                 attention_mask = attention_mask.expand(query_indices.shape[:-1] + (-1,))
-                key_attn_mask = torch.gather(attention_mask, -1, key_indices)
-                query_attn_mask = torch.gather(attention_mask, -1, query_indices)
-                # expand to query_key_dots shape: duplicate along query axis since key sorting is the same for each query position in chunk
-                attn_mask = query_attn_mask.unsqueeze(-1) * key_attn_mask.unsqueeze(-2)
+                attention_mask = torch.gather(attention_mask, -1, key_indices)
 
-                # free memory
-                del query_attn_mask, key_attn_mask
-            else:
-                # usual attention mask creation
-                attention_mask = attention_mask.to(torch.uint8)[:, None, :]
-                attn_mask = (attention_mask.unsqueeze(-1) * attention_mask.unsqueeze(-2)).expand(
-                    query_indices.shape + attention_mask.shape[-1:]
-                )
-
-            # free memory
-            del attention_mask
+            attention_mask = attention_mask.unsqueeze(-2).expand(query_key_dot_shape)
 
             # multiply by casaul mask if necessary
             if mask is not None:
-                mask = mask * attn_mask
+                mask = mask * attention_mask
             else:
-                mask = attn_mask
+                mask = attention_mask
 
         return mask
 
@@ -931,9 +936,7 @@ class LocalSelfAttention(nn.Module, EfficientAttentionMixin):
 
             if self.chunk_length < sequence_length:
                 attention_mask = self._split_seq_length_dim_to(attention_mask, -1, self.chunk_length, 1)
-                attention_mask_key = self._look_adjacent(attention_mask, self.num_chunks_before, self.num_chunks_after)
-            else:
-                attention_mask_key = attention_mask
+                attention_mask = self._look_adjacent(attention_mask, self.num_chunks_before, self.num_chunks_after)
 
         # Causal mask
         if self.is_decoder is True:
@@ -942,12 +945,12 @@ class LocalSelfAttention(nn.Module, EfficientAttentionMixin):
         # Attention mask
         if attention_mask is not None:
             # create attn_mask
-            attn_mask = (attention_mask.unsqueeze(-1) * attention_mask_key.unsqueeze(-2)).expand(query_key_dots_shape)
+            attention_mask = attention_mask.unsqueeze(-2).expand(query_key_dots_shape)
             # multiply by casaul mask if necessary
             if mask is not None:
-                mask = mask * attn_mask
+                mask = mask * attention_mask
             else:
-                mask = attn_mask
+                mask = attention_mask
         return mask
 
 
@@ -1796,9 +1799,9 @@ class ReformerModelWithLMHead(ReformerPreTrainedModel):
 class ReformerForMaskedLM(ReformerPreTrainedModel):
     def __init__(self, config):
         super().__init__(config)
-        assert (
-            not config.is_decoder
-        ), "If you want to use `ReformerForMaskedLM` make sure `config.is_decoder=False` for bi-directional self-attention."
+        #        assert (
+        #            not config.is_decoder
+        #        ), "If you want to use `ReformerForMaskedLM` make sure `config.is_decoder=False` for bi-directional self-attention."
         self.reformer = ReformerModel(config)
         self.lm_head = ReformerOnlyLMHead(config)
 

src/transformers/modeling_tf_t5.py

@@ -478,7 +478,7 @@ class TFT5Block(tf.keras.layers.Layer):
         return outputs  # hidden-states, present_key_value_states, (self-attention weights), (self-attention position bias), (cross-attention weights), (cross-attention position bias)
 
 
-class _NoLayerEmbedTokens(object):
+class _NoLayerEmbedTokens:
     """
      this class wraps a the TFSharedEmbeddingTokens layer into a python 'no-keras-layer'
      class to avoid problem with weight restoring. Also it makes sure that the layer is
@@ -655,7 +655,7 @@ class TFT5MainLayer(tf.keras.layers.Layer):
         # Since we are adding it to the raw scores before the softmax, this is
         # effectively the same as removing these entirely.
 
-        # T5 has a mask that can compare sequence ids, we can simulate this here with this transposistion
+        # T5 has a mask that can compare sequence ids, we can simulate this here with this transposition
         # Cf. https://github.com/tensorflow/mesh/blob/8d2465e9bc93129b913b5ccc6a59aa97abd96ec6/mesh_tensorflow/transformer/transformer_layers.py#L270
         # extended_attention_mask = tf.math.equal(extended_attention_mask,
         #                                         tf.transpose(extended_attention_mask, perm=(-1, -2)))
@@ -682,16 +682,8 @@ class TFT5MainLayer(tf.keras.layers.Layer):
         else:
             encoder_extended_attention_mask = None
 
-        # Prepare head mask if needed
-        # 1.0 in head_mask indicate we keep the head
-        # attention_probs has shape bsz x n_heads x N x N
-        # input head_mask has shape [num_heads] or [num_hidden_layers x num_heads]
-        # and head_mask is converted to shape [num_hidden_layers x batch x num_heads x seq_length x seq_length]
-        if head_mask is not None:
-            raise NotImplementedError
-        else:
-            head_mask = [None] * self.num_hidden_layers
-            # head_mask = tf.constant([0] * self.num_hidden_layers)
+        assert head_mask is None, "Head mask not supported"
+        head_mask = [None] * self.num_hidden_layers
 
         present_key_value_states = ()
         all_hidden_states = ()
@@ -1054,8 +1046,6 @@ class TFT5ForConditionalGeneration(TFT5PreTrainedModel):
         r"""
     Returns:
         :obj:`tuple(tf.Tensor)` comprising various elements depending on the configuration (:class:`~transformers.T5Config`) and inputs:
-        loss (:obj:`tf.Tensor` of shape :obj:`(1,)`, `optional`, returned when :obj:`lm_label` is provided):
-            Classification loss (cross entropy).
         prediction_scores (:obj:`tf.Tensor` of shape :obj:`(batch_size, sequence_length, config.vocab_size)`)
             Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).
         decoder_past_key_value_states (:obj:`tuple(tuple(tf.Tensor))` of length :obj:`config.n_layers` with each tuple having 4 tensors of shape :obj:`(batch_size, num_heads, sequence_length, embed_size_per_head)`, `optional`, returned when ``use_cache=True``):

tests/test_modeling_tf_t5.py

@@ -17,6 +17,7 @@
 import unittest
 
 from transformers import T5Config, is_tf_available
+from transformers.file_utils import cached_property
 from transformers.testing_utils import require_tf, slow
 
 from .test_configuration_common import ConfigTester
@@ -255,9 +256,13 @@ class TFT5ModelTest(TFModelTesterMixin, unittest.TestCase):
 
 @require_tf
 class TFT5ModelIntegrationTests(unittest.TestCase):
+    @cached_property
+    def model(self):
+        return TFT5ForConditionalGeneration.from_pretrained("t5-base")
+
     @slow
     def test_summarization(self):
-        model = TFT5ForConditionalGeneration.from_pretrained("t5-base")
+        model = self.model
         tok = T5Tokenizer.from_pretrained("t5-base")
 
         FRANCE_ARTICLE = 'Marseille, France (CNN)The French prosecutor leading an investigation into the crash of Germanwings Flight 9525 insisted Wednesday that he was not aware of any video footage from on board the plane. Marseille prosecutor Brice Robin told CNN that "so far no videos were used in the crash investigation." He added, "A person who has such a video needs to immediately give it to the investigators." Robin\'s comments follow claims by two magazines, German daily Bild and French Paris Match, of a cell phone video showing the harrowing final seconds from on board Germanwings Flight 9525 as it crashed into the French Alps. All 150 on board were killed. Paris Match and Bild reported that the video was recovered from a phone at the wreckage site. The two publications described the supposed video, but did not post it on their websites. The publications said that they watched the video, which was found by a source close to the investigation. "One can hear cries of \'My God\' in several languages," Paris Match reported. "Metallic banging can also be heard more than three times, perhaps of the pilot trying to open the cockpit door with a heavy object.  Towards the end, after a heavy shake, stronger than the others, the screaming intensifies. Then nothing." "It is a very disturbing scene," said Julian Reichelt, editor-in-chief of Bild online. An official with France\'s accident investigation agency, the BEA, said the agency is not aware of any such video. Lt. Col. Jean-Marc Menichini, a French Gendarmerie spokesman in charge of communications on rescue efforts around the Germanwings crash site, told CNN that the reports were "completely wrong" and "unwarranted." Cell phones have been collected at the site, he said, but that they "hadn\'t been exploited yet." Menichini said he believed the cell phones would need to be sent to the Criminal Research Institute in Rosny sous-Bois, near Paris, in order to be analyzed by specialized technicians working hand-in-hand with investigators. But none of the cell phones found so far have been sent to the institute, Menichini said. Asked whether staff involved in the search could have leaked a memory card to the media, Menichini answered with a categorical "no." Reichelt told "Erin Burnett: Outfront" that he had watched the video and stood by the report, saying Bild and Paris Match are "very confident" that the clip is real. He noted that investigators only revealed they\'d recovered cell phones from the crash site after Bild and Paris Match published their reports. "That is something we did not know before. ... Overall we can say many things of the investigation weren\'t revealed by the investigation at the beginning," he said. What was mental state of Germanwings co-pilot? German airline Lufthansa confirmed Tuesday that co-pilot Andreas Lubitz had battled depression years before he took the controls of Germanwings Flight 9525, which he\'s accused of deliberately crashing last week in the French Alps. Lubitz told his Lufthansa flight training school in 2009 that he had a "previous episode of severe depression," the airline said Tuesday. Email correspondence between Lubitz and the school discovered in an internal investigation, Lufthansa said, included medical documents he submitted in connection with resuming his flight training. The announcement indicates that Lufthansa, the parent company of Germanwings, knew of Lubitz\'s battle with depression, allowed him to continue training and ultimately put him in the cockpit. Lufthansa, whose CEO Carsten Spohr previously said Lubitz was 100% fit to fly, described its statement Tuesday as a "swift and seamless clarification" and said it was sharing the information and documents -- including training and medical records -- with public prosecutors. Spohr traveled to the crash site Wednesday, where recovery teams have been working for the past week to recover human remains and plane debris scattered across a steep mountainside. He saw the crisis center set up in Seyne-les-Alpes, laid a wreath in the village of Le Vernet, closer to the crash site, where grieving families have left flowers at a simple stone memorial. Menichini told CNN late Tuesday that no visible human remains were left at the site but recovery teams would keep searching. French President Francois Hollande, speaking Tuesday, said that it should be possible to identify all the victims using DNA analysis by the end of the week, sooner than authorities had previously suggested. In the meantime, the recovery of the victims\' personal belongings will start Wednesday, Menichini said. Among those personal belongings could be more cell phones belonging to the 144 passengers and six crew on board. Check out the latest from our correspondents . The details about Lubitz\'s correspondence with the flight school during his training were among several developments as investigators continued to delve into what caused the crash and Lubitz\'s possible motive for downing the jLine truncated
@@ -308,12 +313,12 @@ class TFT5ModelIntegrationTests(unittest.TestCase):
 
     @slow
     def test_translation_en_to_de(self):
-        model = TFT5ForConditionalGeneration.from_pretrained("t5-base")
         tok = T5Tokenizer.from_pretrained("t5-base")
+        model = self.model
 
         task_specific_config = getattr(model.config, "task_specific_params", {})
         translation_config = task_specific_config.get("translation_en_to_de", {})
-        model.config.update(translation_config)
+        self.model.config.update(translation_config)
 
         original_input = '"Luigi often said to me that he never wanted the brothers to end up in court", she wrote.'
         expected_translation = (
@@ -337,7 +342,7 @@ class TFT5ModelIntegrationTests(unittest.TestCase):
 
     @slow
     def test_translation_en_to_fr(self):
-        model = TFT5ForConditionalGeneration.from_pretrained("t5-base")
+        model = self.model
         tok = T5Tokenizer.from_pretrained("t5-base")
 
         task_specific_config = getattr(model.config, "task_specific_params", {})
@@ -364,7 +369,7 @@ class TFT5ModelIntegrationTests(unittest.TestCase):
 
     @slow
     def test_translation_en_to_ro(self):
-        model = TFT5ForConditionalGeneration.from_pretrained("t5-base")
+        model = self.model
         tok = T5Tokenizer.from_pretrained("t5-base")
 
         task_specific_config = getattr(model.config, "task_specific_params", {})