Sign upLogin

tensorflow/models

View on GitHub
Star

Showing 11,634 of 11,634 total issues

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

    self.assertNotAllClose(
        tf_keras.backend.eval(test_layer._query_dense.kernel),
        tf_keras.backend.eval(test_layer._output_dense.kernel),
Severity: Major
Found in official/nlp/modeling/layers/block_sparse_attention_test.py and 2 other locations - About 40 mins to fix
official/nlp/modeling/layers/block_sparse_attention_test.py on lines 272..274
official/nlp/modeling/layers/block_sparse_attention_test.py on lines 276..278

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 34.

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

Severity: Major
Found in official/projects/mosaic/qat/serving/export_tflite.py and 2 other locations - About 40 mins to fix
official/projects/qat/vision/serving/export_tflite.py on lines 0..23
official/projects/yolo/serving/export_tflite.py on lines 0..23

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 34.

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

    stateful_reader2 = contrib_training.SequenceQueueingStateSaver(
Severity: Major
Found in research/lstm_object_detection/models/lstm_ssd_interleaved_mobilenet_v2_feature_extractor_test.py and 2 other locations - About 40 mins to fix
research/lstm_object_detection/models/lstm_ssd_interleaved_mobilenet_v2_feature_extractor_test.py on lines 276..276
research/lstm_object_detection/models/lstm_ssd_mobilenet_v1_feature_extractor_test.py on lines 154..154

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 34.

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

      (nmsed_boxes, nmsed_scores, nmsed_classes, nmsed_masks,
       nmsed_additional_fields, num_detections
      ) = post_processing.batch_multiclass_non_max_suppression(
          boxes, scores, score_thresh, iou_thresh,
Severity: Major
Found in research/object_detection/core/batch_multiclass_nms_test.py and 2 other locations - About 40 mins to fix
research/object_detection/core/batch_multiclass_nms_test.py on lines 557..560
research/object_detection/core/batch_multiclass_nms_test.py on lines 613..616

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 34.

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

    self.assertNotAllClose(
        tf_keras.backend.eval(test_layer._query_dense.kernel),
        tf_keras.backend.eval(test_layer._value_dense.kernel),
Severity: Major
Found in official/nlp/modeling/layers/block_sparse_attention_test.py and 2 other locations - About 40 mins to fix
official/nlp/modeling/layers/block_sparse_attention_test.py on lines 272..274
official/nlp/modeling/layers/block_sparse_attention_test.py on lines 280..282

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 34.

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

Severity: Major
Found in orbit/__init__.py and 2 other locations - About 40 mins to fix
official/recommendation/uplift/metrics/__init__.py on lines 0..26
orbit/utils/__init__.py on lines 0..30

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 34.

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

    class_targets = tf.random.uniform([batch_size, num_boxes],
                                      minval=0,
                                      maxval=num_classes + 1,
                                      dtype=tf.int32)
Severity: Major
Found in official/vision/losses/maskrcnn_losses_test.py and 2 other locations - About 40 mins to fix
official/vision/losses/maskrcnn_losses_test.py on lines 115..118
official/vision/losses/maskrcnn_losses_test.py on lines 136..139

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 34.

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

    select_class_targets = tf.random.uniform([batch_size, num_masks],
                                             minval=0,
                                             maxval=num_classes + 1,
                                             dtype=tf.int32)
Severity: Major
Found in official/vision/losses/maskrcnn_losses_test.py and 2 other locations - About 40 mins to fix
official/vision/losses/maskrcnn_losses_test.py on lines 75..78
official/vision/losses/maskrcnn_losses_test.py on lines 115..118

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 34.

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

np.testing.assert_allclose(
    [np.mean(embedding_batch), np.std(embedding_batch)],
    [expected_embedding_mean, expected_embedding_std],
Severity: Major
Found in research/audioset/vggish/vggish_smoke_test.py and 2 other locations - About 40 mins to fix
research/audioset/vggish/vggish_export_tfhub.py on lines 96..98
research/audioset/vggish/vggish_smoke_test.py on lines 110..112

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 34.

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

      critic_train_op = slim.learning.create_train_op(
          critic_loss,
          critic_opt,
          global_step=n_updates,
          update_ops=None,
Severity: Major
Found in research/efficient-hrl/train.py and 2 other locations - About 40 mins to fix
research/efficient-hrl/train.py on lines 498..505
research/efficient-hrl/train.py on lines 560..567

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 34.

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

      for key, value in box_predictions.items():
        box_predictions[key] = tf.concat(value, axis=1)
Severity: Major
Found in research/object_detection/predictors/convolutional_box_predictor_tf1_test.py and 2 other locations - About 40 mins to fix
research/object_detection/predictors/convolutional_keras_box_predictor_tf2_test.py on lines 926..927
research/object_detection/model_lib_v2.py on lines 828..829

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 34.

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

      self.r_w_bias = (
          self.add_weight(
              "r_w_bias",
              shape=[self.n_head, self.d_head],
              dtype=self.tf_float,
Severity: Major
Found in official/legacy/xlnet/xlnet_modeling.py and 2 other locations - About 40 mins to fix
official/legacy/xlnet/xlnet_modeling.py on lines 248..252
official/legacy/xlnet/xlnet_modeling.py on lines 484..489

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 34.

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

        center_matched_gt_boxes = center_match.gather_based_on_match(
            groundtruth_boxes.get(),
            unmatched_value=tf.zeros(4),
            ignored_value=tf.zeros(4))
Severity: Major
Found in official/vision/utils/object_detection/target_assigner.py and 2 other locations - About 40 mins to fix
official/vision/utils/object_detection/target_assigner.py on lines 201..204
official/vision/utils/object_detection/target_assigner.py on lines 495..498

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 34.

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

    self.end_logits_proj_layer0 = tf_keras.layers.Dense(
        units=self.hidden_size,
        kernel_initializer=self.initializer,
        activation=tf.nn.tanh,
Severity: Major
Found in official/legacy/xlnet/xlnet_modeling.py and 2 other locations - About 40 mins to fix
official/legacy/xlnet/xlnet_modeling.py on lines 1026..1030
official/legacy/xlnet/xlnet_modeling.py on lines 1199..1202

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 34.

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

    class_targets = tf.random.uniform([batch_size, num_boxes],
                                      minval=0,
                                      maxval=num_classes + 1,
                                      dtype=tf.int32)
Severity: Major
Found in official/vision/losses/maskrcnn_losses_test.py and 2 other locations - About 40 mins to fix
official/vision/losses/maskrcnn_losses_test.py on lines 75..78
official/vision/losses/maskrcnn_losses_test.py on lines 136..139

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 34.

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

    matched_gt_boxes = match.gather_based_on_match(
        groundtruth_boxes.get(),
        unmatched_value=tf.zeros(4),
        ignored_value=tf.zeros(4))
Severity: Major
Found in official/vision/utils/object_detection/target_assigner.py and 2 other locations - About 40 mins to fix
official/vision/utils/object_detection/target_assigner.py on lines 201..204
official/vision/utils/object_detection/target_assigner.py on lines 461..464

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 34.

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

np.testing.assert_allclose(
    [np.mean(postprocessed_batch), np.std(postprocessed_batch)],
    [expected_postprocessed_mean, expected_postprocessed_std],
Severity: Major
Found in research/audioset/vggish/vggish_smoke_test.py and 2 other locations - About 40 mins to fix
research/audioset/vggish/vggish_export_tfhub.py on lines 96..98
research/audioset/vggish/vggish_smoke_test.py on lines 103..105

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 34.

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

      (nmsed_boxes, nmsed_scores, nmsed_classes, nmsed_masks,
       nmsed_additional_fields, num_detections
      ) = post_processing.batch_multiclass_non_max_suppression(
          boxes, scores, score_thresh, iou_thresh,
Severity: Major
Found in research/object_detection/core/batch_multiclass_nms_test.py and 2 other locations - About 40 mins to fix
research/object_detection/core/batch_multiclass_nms_test.py on lines 475..478
research/object_detection/core/batch_multiclass_nms_test.py on lines 613..616

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 34.

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

    matched_anchors_mask = tf.tile(
        tf.expand_dims(matched_anchors_mask, 1),
        [1, tf.shape(matched_reg_targets)[1]])
Severity: Major
Found in official/vision/utils/object_detection/target_assigner.py and 2 other locations - About 40 mins to fix
official/legacy/detection/dataloader/anchor.py on lines 417..418
official/vision/utils/object_detection/target_assigner.py on lines 517..519

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 34.

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 len(examples) % 1000 == 0:
          logging.info("Loading dev example %d", len(examples))
Severity: Major
Found in official/legacy/xlnet/preprocess_classification_data.py and 2 other locations - About 40 mins to fix
official/legacy/bert/run_squad_helper.py on lines 210..211
official/legacy/xlnet/run_squad.py on lines 175..176

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 34.

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

Severity
Category
Status
Source
Language