Sign upLogin

tensorflow/models

View on GitHub
Star
official/projects/detr/modeling/transformer.py

Summary

Maintainability
F
1 wk
Test Coverage

File transformer.py has 765 lines of code (exceeds 250 allowed). Consider refactoring.
Open

# Copyright 2024 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
Severity: Major
Found in official/projects/detr/modeling/transformer.py - About 1 day to fix

    Function call has a Cognitive Complexity of 19 (exceeds 5 allowed). Consider refactoring.
    Open

      def call(self, inputs):
    """Transformer self-attention encoder block call.

    Args:
      inputs: a single tensor or a list of tensors. `input tensor` as the single
    Severity: Minor
    Found in official/projects/detr/modeling/transformer.py - About 2 hrs to fix

    Cognitive Complexity

    Cognitive Complexity is a measure of how difficult a unit of code is to intuitively understand. Unlike Cyclomatic Complexity, which determines how difficult your code will be to test, Cognitive Complexity tells you how difficult your code will be to read and comprehend.

    A method's cognitive complexity is based on a few simple rules:

    • Code is not considered more complex when it uses shorthand that the language provides for collapsing multiple statements into one
    • Code is considered more complex for each "break in the linear flow of the code"
    • Code is considered more complex when "flow breaking structures are nested"

    Further reading

    Function __init__ has 20 arguments (exceeds 4 allowed). Consider refactoring.
    Open

      def __init__(self,
    Severity: Major
    Found in official/projects/detr/modeling/transformer.py - About 2 hrs to fix

      Function __init__ has 18 arguments (exceeds 4 allowed). Consider refactoring.
      Open

        def __init__(self,
      Severity: Major
      Found in official/projects/detr/modeling/transformer.py - About 2 hrs to fix

        Function __init__ has 11 arguments (exceeds 4 allowed). Consider refactoring.
        Open

          def __init__(self,
        Severity: Major
        Found in official/projects/detr/modeling/transformer.py - About 1 hr to fix

          Function __init__ has 11 arguments (exceeds 4 allowed). Consider refactoring.
          Open

            def __init__(self,
          Severity: Major
          Found in official/projects/detr/modeling/transformer.py - About 1 hr to fix

            Function call has 9 arguments (exceeds 4 allowed). Consider refactoring.
            Open

              def call(self,
            Severity: Major
            Found in official/projects/detr/modeling/transformer.py - About 1 hr to fix

              Function call has a Cognitive Complexity of 9 (exceeds 5 allowed). Consider refactoring.
              Open

                def call(self, inputs, cache=None, decode_loop_step=None):
    input_tensor, memory, attention_mask, self_attention_mask, input_pos_embed, memory_pos_embed = inputs
    source_tensor = input_tensor
    if self._norm_first:
      input_tensor = self.self_attention_layer_norm(input_tensor)
              Severity: Minor
              Found in official/projects/detr/modeling/transformer.py - About 55 mins to fix

              Cognitive Complexity

              Cognitive Complexity is a measure of how difficult a unit of code is to intuitively understand. Unlike Cyclomatic Complexity, which determines how difficult your code will be to test, Cognitive Complexity tells you how difficult your code will be to read and comprehend.

              A method's cognitive complexity is based on a few simple rules:

              • Code is not considered more complex when it uses shorthand that the language provides for collapsing multiple statements into one
              • Code is considered more complex for each "break in the linear flow of the code"
              • Code is considered more complex when "flow breaking structures are nested"

              Further reading

              Function call has a Cognitive Complexity of 8 (exceeds 5 allowed). Consider refactoring.
              Open

                def call(self,
           target,
           memory,
           self_attention_mask=None,
           cross_attention_mask=None,
              Severity: Minor
              Found in official/projects/detr/modeling/transformer.py - About 45 mins to fix

              Cognitive Complexity

              Cognitive Complexity is a measure of how difficult a unit of code is to intuitively understand. Unlike Cyclomatic Complexity, which determines how difficult your code will be to test, Cognitive Complexity tells you how difficult your code will be to read and comprehend.

              A method's cognitive complexity is based on a few simple rules:

              • Code is not considered more complex when it uses shorthand that the language provides for collapsing multiple statements into one
              • Code is considered more complex for each "break in the linear flow of the code"
              • Code is considered more complex when "flow breaking structures are nested"

              Further reading

              Function build has a Cognitive Complexity of 6 (exceeds 5 allowed). Consider refactoring.
              Open

                def build(self, input_shape):
    if isinstance(input_shape, tf.TensorShape):
      input_tensor_shape = input_shape
    elif isinstance(input_shape, (list, tuple)):
      input_tensor_shape = tf.TensorShape(input_shape[0])
              Severity: Minor
              Found in official/projects/detr/modeling/transformer.py - About 25 mins to fix

              Cognitive Complexity

              Cognitive Complexity is a measure of how difficult a unit of code is to intuitively understand. Unlike Cyclomatic Complexity, which determines how difficult your code will be to test, Cognitive Complexity tells you how difficult your code will be to read and comprehend.

              A method's cognitive complexity is based on a few simple rules:

              • Code is not considered more complex when it uses shorthand that the language provides for collapsing multiple statements into one
              • Code is considered more complex for each "break in the linear flow of the code"
              • Code is considered more complex when "flow breaking structures are nested"

              Further reading

              Similar blocks of code found in 2 locations. Consider refactoring.
              Open

                def build(self, input_shape):
    """Implements build() for the layer."""
    self.decoder_layers = []
    for i in range(self.num_layers):
      self.decoder_layers.append(
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 1 other location - About 1 day to fix
              official/projects/detr/modeling/transformer.py on lines 80..101

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 156.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Similar blocks of code found in 2 locations. Consider refactoring.
              Open

                def build(self, input_shape):
    """Implements build() for the layer."""
    self.encoder_layers = []
    for i in range(self.num_layers):
      self.encoder_layers.append(
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 1 other location - About 1 day to fix
              official/projects/detr/modeling/transformer.py on lines 478..499

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 156.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Similar blocks of code found in 3 locations. Consider refactoring.
              Open

                def __init__(self,
               num_layers=6,
               num_attention_heads=8,
               intermediate_size=2048,
               activation="relu",
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 2 other locations - About 7 hrs to fix
              official/nlp/modeling/models/seq2seq_transformer.py on lines 371..412
              official/projects/detr/modeling/transformer.py on lines 37..78

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 114.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Similar blocks of code found in 3 locations. Consider refactoring.
              Open

                def __init__(self,
               num_layers=6,
               num_attention_heads=8,
               intermediate_size=2048,
               activation="relu",
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 2 other locations - About 7 hrs to fix
              official/nlp/modeling/models/seq2seq_transformer.py on lines 371..412
              official/projects/detr/modeling/transformer.py on lines 436..476

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 114.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Similar blocks of code found in 5 locations. Consider refactoring.
              Open

                def get_config(self):
    config = {
        "num_layers": self.num_layers,
        "num_attention_heads": self.num_attention_heads,
        "intermediate_size": self._intermediate_size,
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 4 other locations - About 5 hrs to fix
              official/legacy/detection/modeling/architecture/nn_blocks.py on lines 132..147
              official/legacy/detection/modeling/architecture/nn_blocks.py on lines 282..297
              official/nlp/modeling/models/seq2seq_transformer.py on lines 435..449
              official/projects/detr/modeling/transformer.py on lines 501..515

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 96.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Similar blocks of code found in 5 locations. Consider refactoring.
              Open

                def get_config(self):
    config = {
        "num_layers": self.num_layers,
        "num_attention_heads": self.num_attention_heads,
        "intermediate_size": self._intermediate_size,
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 4 other locations - About 5 hrs to fix
              official/legacy/detection/modeling/architecture/nn_blocks.py on lines 132..147
              official/legacy/detection/modeling/architecture/nn_blocks.py on lines 282..297
              official/nlp/modeling/models/seq2seq_transformer.py on lines 435..449
              official/projects/detr/modeling/transformer.py on lines 103..117

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 96.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Similar blocks of code found in 3 locations. Consider refactoring.
              Open

                    if cache is None:
        output_tensor, _ = self.decoder_layers[layer_idx](transformer_inputs)
      else:
        cache_layer_idx = str(layer_idx)
        output_tensor, cache[cache_layer_idx] = self.decoder_layers[layer_idx](
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 2 other locations - About 4 hrs to fix
              official/nlp/modeling/models/seq2seq_transformer.py on lines 629..635
              official/projects/nhnet/decoder.py on lines 129..135

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 79.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Identical blocks of code found in 3 locations. Consider refactoring.
              Open

                  if isinstance(input_shape, tf.TensorShape):
      input_tensor_shape = input_shape
    elif isinstance(input_shape, (list, tuple)):
      input_tensor_shape = tf.TensorShape(input_shape[0])
    else:
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 2 other locations - About 3 hrs to fix
              official/nlp/modeling/layers/reuse_transformer.py on lines 141..148
              official/nlp/modeling/layers/transformer_scaffold.py on lines 128..135

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 64.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Identical blocks of code found in 2 locations. Consider refactoring.
              Open

                    if self._norm_first:
        source_tensor = input_tensor[:, 0:self._output_range, :]
        input_tensor = self._attention_layer_norm(input_tensor)
        if key_value is not None:
          key_value = self._attention_layer_norm(key_value)
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 1 other location - About 2 hrs to fix
              official/nlp/modeling/layers/reuse_transformer.py on lines 311..315

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 57.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Identical blocks of code found in 2 locations. Consider refactoring.
              Open

                  if self._output_range:
      if self._norm_first:
        source_tensor = input_tensor[:, 0:self._output_range, :]
        input_tensor = self._attention_layer_norm(input_tensor)
        if key_value is not None:
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 1 other location - About 1 hr to fix
              official/nlp/modeling/layers/reuse_transformer.py on lines 310..328

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 49.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Similar blocks of code found in 2 locations. Consider refactoring.
              Open

                  self.output_dense = tf_keras.layers.EinsumDense(
        "abc,cd->abd",
        output_shape=(None, hidden_size),
        bias_axes="d",
        kernel_initializer=tf_utils.clone_initializer(self._kernel_initializer),
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 1 other location - About 1 hr to fix
              official/projects/detr/modeling/transformer.py on lines 691..695

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 41.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Similar blocks of code found in 2 locations. Consider refactoring.
              Open

                  self.self_attention_output_dense = tf_keras.layers.EinsumDense(
        "abc,cd->abd",
        output_shape=(None, hidden_size),
        bias_axes="d",
        kernel_initializer=tf_utils.clone_initializer(self._kernel_initializer),
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 1 other location - About 1 hr to fix
              official/projects/detr/modeling/transformer.py on lines 738..742

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 41.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Identical blocks of code found in 2 locations. Consider refactoring.
              Open

                  common_kwargs = dict(
        bias_initializer=self._bias_initializer,
        kernel_regularizer=self._kernel_regularizer,
        bias_regularizer=self._bias_regularizer,
        activity_regularizer=self._activity_regularizer,
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 1 other location - About 1 hr to fix
              official/projects/detr/modeling/transformer.py on lines 261..267

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 38.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Identical blocks of code found in 2 locations. Consider refactoring.
              Open

                  common_kwargs = dict(
        bias_initializer=self._bias_initializer,
        kernel_regularizer=self._kernel_regularizer,
        bias_regularizer=self._bias_regularizer,
        activity_regularizer=self._activity_regularizer,
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 1 other location - About 1 hr to fix
              official/projects/detr/modeling/transformer.py on lines 675..681

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 38.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Similar blocks of code found in 7 locations. Consider refactoring.
              Open

                  if hidden_size % self._num_heads != 0:
      raise ValueError(
          "The input size (%d) is not a multiple of the number of attention "
          "heads (%d)" % (hidden_size, self._num_heads))
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 6 other locations - About 50 mins to fix
              official/nlp/modeling/layers/reuse_transformer.py on lines 154..157
              official/nlp/modeling/layers/tn_transformer_expand_condense.py on lines 127..130
              official/nlp/modeling/layers/transformer.py on lines 260..263
              official/nlp/modeling/layers/transformer_scaffold.py on lines 142..145
              official/nlp/modeling/layers/transformer_xl.py on lines 142..145
              official/projects/detr/modeling/transformer.py on lines 670..673

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 36.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Similar blocks of code found in 7 locations. Consider refactoring.
              Open

                  if hidden_size % self.num_attention_heads != 0:
      raise ValueError(
          "The hidden size (%d) is not a multiple of the number of attention "
          "heads (%d)" % (hidden_size, self.num_attention_heads))
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 6 other locations - About 50 mins to fix
              official/nlp/modeling/layers/reuse_transformer.py on lines 154..157
              official/nlp/modeling/layers/tn_transformer_expand_condense.py on lines 127..130
              official/nlp/modeling/layers/transformer.py on lines 260..263
              official/nlp/modeling/layers/transformer_scaffold.py on lines 142..145
              official/nlp/modeling/layers/transformer_xl.py on lines 142..145
              official/projects/detr/modeling/transformer.py on lines 256..259

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 36.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Identical blocks of code found in 5 locations. Consider refactoring.
              Open

                  if attention_initializer:
      self._attention_initializer = tf_keras.initializers.get(
          attention_initializer)
    else:
      self._attention_initializer = tf_utils.clone_initializer(
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 4 other locations - About 45 mins to fix
              official/nlp/modeling/layers/reuse_transformer.py on lines 132..137
              official/nlp/modeling/layers/tn_transformer_expand_condense.py on lines 100..105
              official/nlp/modeling/layers/transformer.py on lines 234..239
              official/projects/detr/modeling/transformer.py on lines 656..661

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 35.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Identical blocks of code found in 5 locations. Consider refactoring.
              Open

                  if attention_initializer:
      self._attention_initializer = tf_keras.initializers.get(
          attention_initializer)
    else:
      self._attention_initializer = tf_utils.clone_initializer(
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 4 other locations - About 45 mins to fix
              official/nlp/modeling/layers/reuse_transformer.py on lines 132..137
              official/nlp/modeling/layers/tn_transformer_expand_condense.py on lines 100..105
              official/nlp/modeling/layers/transformer.py on lines 234..239
              official/projects/detr/modeling/transformer.py on lines 235..240

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 35.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Similar blocks of code found in 10 locations. Consider refactoring.
              Open

                  if self._norm_first:
      attention_output = source_tensor + attention_output
    else:
      attention_output = self._attention_layer_norm(target_tensor +
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 9 other locations - About 45 mins to fix
              official/nlp/modeling/layers/reuse_transformer.py on lines 338..341
              official/nlp/modeling/layers/tn_transformer_expand_condense.py on lines 235..238
              official/nlp/modeling/layers/tn_transformer_expand_condense.py on lines 250..253
              official/nlp/modeling/layers/transformer.py on lines 419..423
              official/nlp/modeling/layers/transformer.py on lines 438..442
              official/nlp/modeling/layers/transformer.py on lines 453..456
              official/projects/detr/modeling/transformer.py on lines 814..818
              official/projects/detr/modeling/transformer.py on lines 830..834
              official/projects/detr/modeling/transformer.py on lines 845..848

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 35.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Similar blocks of code found in 10 locations. Consider refactoring.
              Open

                  if self._norm_first:
      attention_output = source_self_attention_output + attention_output
    else:
      attention_output = self.encdec_attention_layer_norm(
          self_attention_output + attention_output)
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 9 other locations - About 45 mins to fix
              official/nlp/modeling/layers/reuse_transformer.py on lines 338..341
              official/nlp/modeling/layers/tn_transformer_expand_condense.py on lines 235..238
              official/nlp/modeling/layers/tn_transformer_expand_condense.py on lines 250..253
              official/nlp/modeling/layers/transformer.py on lines 419..423
              official/nlp/modeling/layers/transformer.py on lines 438..442
              official/nlp/modeling/layers/transformer.py on lines 453..456
              official/projects/detr/modeling/transformer.py on lines 402..405
              official/projects/detr/modeling/transformer.py on lines 814..818
              official/projects/detr/modeling/transformer.py on lines 845..848

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 35.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Similar blocks of code found in 10 locations. Consider refactoring.
              Open

                  if self._norm_first:
      self_attention_output = source_tensor + self_attention_output
    else:
      self_attention_output = self.self_attention_layer_norm(
          input_tensor + self_attention_output)
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 9 other locations - About 45 mins to fix
              official/nlp/modeling/layers/reuse_transformer.py on lines 338..341
              official/nlp/modeling/layers/tn_transformer_expand_condense.py on lines 235..238
              official/nlp/modeling/layers/tn_transformer_expand_condense.py on lines 250..253
              official/nlp/modeling/layers/transformer.py on lines 419..423
              official/nlp/modeling/layers/transformer.py on lines 438..442
              official/nlp/modeling/layers/transformer.py on lines 453..456
              official/projects/detr/modeling/transformer.py on lines 402..405
              official/projects/detr/modeling/transformer.py on lines 830..834
              official/projects/detr/modeling/transformer.py on lines 845..848

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 35.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              Similar blocks of code found in 10 locations. Consider refactoring.
              Open

                  if self._norm_first:
      layer_output = source_attention_output + layer_output
    else:
      layer_output = self.output_layer_norm(layer_output + attention_output)
              Severity: Major
              Found in official/projects/detr/modeling/transformer.py and 9 other locations - About 45 mins to fix
              official/nlp/modeling/layers/reuse_transformer.py on lines 338..341
              official/nlp/modeling/layers/tn_transformer_expand_condense.py on lines 235..238
              official/nlp/modeling/layers/tn_transformer_expand_condense.py on lines 250..253
              official/nlp/modeling/layers/transformer.py on lines 419..423
              official/nlp/modeling/layers/transformer.py on lines 438..442
              official/nlp/modeling/layers/transformer.py on lines 453..456
              official/projects/detr/modeling/transformer.py on lines 402..405
              official/projects/detr/modeling/transformer.py on lines 814..818
              official/projects/detr/modeling/transformer.py on lines 830..834

              Duplicated Code

              Duplicated code can lead to software that is hard to understand and difficult to change. The Don't Repeat Yourself (DRY) principle states:

              Every piece of knowledge must have a single, unambiguous, authoritative representation within a system.

              When you violate DRY, bugs and maintenance problems are sure to follow. Duplicated code has a tendency to both continue to replicate and also to diverge (leaving bugs as two similar implementations differ in subtle ways).

              Tuning

              This issue has a mass of 35.

              We set useful threshold defaults for the languages we support but you may want to adjust these settings based on your project guidelines.

              The threshold configuration represents the minimum mass a code block must have to be analyzed for duplication. The lower the threshold, the more fine-grained the comparison.

              If the engine is too easily reporting duplication, try raising the threshold. If you suspect that the engine isn't catching enough duplication, try lowering the threshold. The best setting tends to differ from language to language.

              See codeclimate-duplication's documentation for more information about tuning the mass threshold in your .codeclimate.yml.

              Refactorings

              Further Reading

              There are no issues that match your filters.

              Category
              Status