bugs/021/github.json

{
  "owner": "tensorflow",
  "repo": "models",
  "number": 10947,
  "html_url": "https://github.com/tensorflow/models/issues/10947",
  "is_pull_request": false,
  "state": "closed",
  "state_reason": "completed",
  "title": " ValueError: List argument 'values' to 'ConcatV2' Op with length 0",
  "author": "JunHyungKang",
  "created_at": "2023-03-07T01:08:44Z",
  "updated_at": "2023-04-12T01:52:38Z",
  "closed_at": "2023-04-12T01:52:35Z",
  "labels": [
    "stat:awaiting response",
    "type:bug",
    "models:official",
    "stale"
  ],
  "milestone": null,
  "comments_count": 12,
  "reactions": {
    "total_count": 0,
    "+1": 0,
    "-1": 0,
    "laugh": 0,
    "hooray": 0,
    "confused": 0,
    "heart": 0,
    "rocket": 0,
    "eyes": 0
  },
  "resolution_days": 36.03,
  "fix": {
    "closing_commit": null,
    "linked_prs": [],
    "best_guess_fix_commit": null,
    "has_fix": false
  },
  "fetched_at": "2026-07-28T12:21:46.611597+00:00",
  "comments": [
    {
      "author": "laxmareddyp",
      "created_at": "2023-03-13T17:11:21Z",
      "body": "Hi @JunHyungKang ,\r\n\r\nSorry for delay in response, In order to expedite the trouble-shooting process, please provide a code snippet/colab code to reproduce the issue reported here. \r\nThanks."
    },
    {
      "author": "JunHyungKang",
      "created_at": "2023-03-16T04:18:17Z",
      "body": "@laxmareddyp \r\nHere is my config and detail error. \r\n\r\n# config \r\n```\r\n{'runtime': {'all_reduce_alg': None,\r\n             'batchnorm_spatial_persistent': False,\r\n             'dataset_num_private_threads': None,\r\n             'default_shard_dim': -1,\r\n             'distribution_strategy': 'mirrored',\r\n             'enable_xla': False,\r\n             'gpu_thread_mode': None,\r\n             'loss_scale': None,\r\n             'mixed_precision_dtype': None,\r\n             'num_cores_per_replica': 1,\r\n             'num_gpus': 2,\r\n             'num_packs': 1,\r\n             'per_gpu_thread_count': 0,\r\n             'run_eagerly': False,\r\n             'task_index': -1,\r\n             'tpu': None,\r\n             'tpu_enable_xla_dynamic_padder': None,\r\n             'worker_hosts': None},\r\n 'task': {'allow_image_summary': False,\r\n          'annotation_file': None,\r\n          'differential_privacy_config': None,\r\n          'export_config': {'cast_detection_classes_to_float': False,\r\n                            'cast_num_detections_to_float': False,\r\n                            'output_intermediate_features': False,\r\n                            'output_normalized_coordinates': False},\r\n          'freeze_backbone': False,\r\n          'init_checkpoint': None,\r\n          'init_checkpoint_modules': 'all',\r\n          'losses': {'box_loss_weight': 50,\r\n                     'focal_loss_alpha': 0.25,\r\n                     'focal_loss_gamma': 1.5,\r\n                     'huber_loss_delta': 0.1,\r\n                     'l2_weight_decay': 0.0001,\r\n                     'loss_weight': 1.0},\r\n          'max_num_eval_detections': 100,\r\n          'model': {'anchor': {'anchor_size': 4.0,\r\n                               'aspect_ratios': [0.5, 1.0, 2.0],\r\n                               'num_scales': 3},\r\n                    'backbone': {'resnet': {'bn_trainable': True,\r\n                                            'depth_multiplier': 1.0,\r\n                                            'model_id': 50,\r\n                                            'replace_stem_max_pool': False,\r\n                                            'resnetd_shortcut': False,\r\n                                            'scale_stem': True,\r\n                                            'se_ratio': 0.0,\r\n                                            'stem_type': 'v0',\r\n                                            'stochastic_depth_drop_rate': 0.0},\r\n                                 'type': 'resnet'},\r\n                    'decoder': {'fpn': {'fusion_type': 'sum',\r\n                                        'num_filters': 256,\r\n                                        'use_keras_layer': False,\r\n                                        'use_separable_conv': False},\r\n                                'type': 'fpn'},\r\n                    'detection_generator': {'apply_nms': True,\r\n                                            'max_num_detections': 100,\r\n                                            'nms_iou_threshold': 0.5,\r\n                                            'nms_version': 'v2',\r\n                                            'pre_nms_score_threshold': 0.05,\r\n                                            'pre_nms_top_k': 5000,\r\n                                            'return_decoded': None,\r\n                                            'soft_nms_sigma': None,\r\n                                            'tflite_post_processing': {'max_classes_per_detection': 5,\r\n                                                                       'max_detections': 200,\r\n                                                                       'nms_iou_threshold': 0.5,\r\n                                                                       'nms_score_threshold': 0.1,\r\n                                                                       'normalize_anchor_coordinates': False,\r\n                                                                       'use_regular_nms': False},\r\n                                            'use_class_agnostic_nms': False,\r\n                                            'use_cpu_nms': False},\r\n                    'head': {'attribute_heads': [],\r\n                             'num_convs': 4,\r\n                             'num_filters': 256,\r\n                             'share_classification_heads': False,\r\n                             'use_separable_conv': False},\r\n                    'input_size': [1024, 1024, 3],\r\n                    'max_level': 7,\r\n                    'min_level': 3,\r\n                    'norm_activation': {'activation': 'relu',\r\n                                        'norm_epsilon': 0.001,\r\n                                        'norm_momentum': 0.99,\r\n                                        'use_sync_bn': True},\r\n                    'num_classes': 1},\r\n          'name': None,\r\n          'per_category_metrics': False,\r\n          'train_data': {'apply_tf_data_service_before_batching': False,\r\n                         'block_length': 1,\r\n                         'cache': False,\r\n                         'cycle_length': None,\r\n                         'decoder': {'simple_decoder': {'attribute_names': [],\r\n                                                        'mask_binarize_threshold': None,\r\n                                                        'regenerate_source_id': False},\r\n                                     'type': 'simple_decoder'},\r\n                         'deterministic': None,\r\n                         'drop_remainder': True,\r\n                         'dtype': 'float32',\r\n                         'enable_shared_tf_data_service_between_parallel_trainers': False,\r\n                         'enable_tf_data_service': False,\r\n                         'file_type': 'tfrecord',\r\n                         'global_batch_size': 2,\r\n                         'input_path': '/mnt/sda1/exc_cctv/1st/tfrecords/train/*',\r\n                         'is_training': True,\r\n                         'parser': {'aug_policy': None,\r\n                                    'aug_rand_hflip': False,\r\n                                    'aug_scale_max': 1.2,\r\n                                    'aug_scale_min': 0.8,\r\n                                    'aug_type': None,\r\n                                    'match_threshold': 0.5,\r\n                                    'max_num_instances': 100,\r\n                                    'num_channels': 3,\r\n                                    'skip_crowd_during_training': True,\r\n                                    'unmatched_threshold': 0.5},\r\n                         'prefetch_buffer_size': None,\r\n                         'seed': None,\r\n                         'sharding': True,\r\n                         'shuffle_buffer_size': 10000,\r\n                         'tf_data_service_address': None,\r\n                         'tf_data_service_job_name': None,\r\n                         'tfds_as_supervised': False,\r\n                         'tfds_data_dir': '',\r\n                         'tfds_name': '',\r\n                         'tfds_skip_decoding_feature': '',\r\n                         'tfds_split': '',\r\n                         'trainer_id': None,\r\n                         'weights': None},\r\n          'use_coco_metrics': True,\r\n          'use_wod_metrics': False,\r\n          'validation_data': {'apply_tf_data_service_before_batching': False,\r\n                              'block_length': 1,\r\n                              'cache': False,\r\n                              'cycle_length': None,\r\n                              'decoder': {'simple_decoder': {'attribute_names': [],\r\n                                                             'mask_binarize_threshold': None,\r\n                                                             'regenerate_source_id': False},\r\n                                          'type': 'simple_decoder'},\r\n                              'deterministic': None,\r\n                              'drop_remainder': True,\r\n                              'dtype': 'float32',\r\n                              'enable_shared_tf_data_service_between_parallel_trainers': False,\r\n                              'enable_tf_data_service': False,\r\n                              'file_type': 'tfrecord',\r\n                              'global_batch_size': 2,\r\n                              'input_path': '/mnt/sda1/exc_cctv/1st/tfrecords/val/*',\r\n                              'is_training': False,\r\n                              'parser': {'aug_policy': None,\r\n                                         'aug_rand_hflip': False,\r\n                                         'aug_scale_max': 1.0,\r\n                                         'aug_scale_min': 1.0,\r\n                                         'aug_type': None,\r\n                                         'match_threshold': 0.5,\r\n                                         'max_num_instances': 100,\r\n                                         'num_channels': 3,\r\n                                         'skip_crowd_during_training': True,\r\n                                         'unmatched_threshold': 0.5},\r\n                              'prefetch_buffer_size': None,\r\n                              'seed': None,\r\n                              'sharding': True,\r\n                              'shuffle_buffer_size': 10000,\r\n                              'tf_data_service_address': None,\r\n                              'tf_data_service_job_name': None,\r\n                              'tfds_as_supervised': False,\r\n                              'tfds_data_dir': '',\r\n                              'tfds_name': '',\r\n                              'tfds_skip_decoding_feature': '',\r\n                              'tfds_split': '',\r\n                              'trainer_id': None,\r\n                              'weights': None}},\r\n 'trainer': {'allow_tpu_summary': False,\r\n             'best_checkpoint_eval_metric': '',\r\n             'best_checkpoint_export_subdir': 'best',\r\n             'best_checkpoint_metric_comp': 'higher',\r\n             'checkpoint_interval': 9744,\r\n             'continuous_eval_timeout': 3600,\r\n             'eval_tf_function': True,\r\n             'eval_tf_while_loop': False,\r\n             'loss_upper_bound': 1000000.0,\r\n             'max_to_keep': 5,\r\n             'optimizer_config': {'ema': None,\r\n                                  'learning_rate': {'stepwise': {'boundaries': [555408,\r\n                                                                                652848],\r\n                                                                 'name': 'PiecewiseConstantDecay',\r\n                                                                 'offset': 0,\r\n                                                                 'values': [0.0025,\r\n                                                                            0.00025,\r\n                                                                            2.5e-05]},\r\n                                                    'type': 'stepwise'},\r\n                                  'optimizer': {'sgd': {'clipnorm': None,\r\n                                                        'clipvalue': None,\r\n                                                        'decay': 0.0,\r\n                                                        'global_clipnorm': None,\r\n                                                        'momentum': 0.9,\r\n                                                        'name': 'SGD',\r\n                                                        'nesterov': False},\r\n                                                'type': 'sgd'},\r\n                                  'warmup': {'linear': {'name': 'linear',\r\n                                                        'warmup_learning_rate': 0.0067,\r\n                                                        'warmup_steps': 500},\r\n                                             'type': 'linear'}},\r\n             'preemption_on_demand_checkpoint': True,\r\n             'recovery_begin_steps': 0,\r\n             'recovery_max_trials': 0,\r\n             'steps_per_loop': 9744,\r\n             'summary_interval': 9744,\r\n             'train_steps': 701568,\r\n             'train_tf_function': True,\r\n             'train_tf_while_loop': True,\r\n             'validation_interval': 9744,\r\n             'validation_steps': -1,\r\n             'validation_summary_subdir': 'validation'}}\r\n```\r\n\r\n\r\n# error \r\n```\r\nI0316 12:00:10.691335 139750881177728 controller.py:502] train | step:  253344 | steps/sec:    3.8 | output: \r\n    {'box_loss': 0.0017456077,\r\n     'cls_loss': 6.524426e-06,\r\n     'learning_rate': 0.0025,\r\n     'model_loss': 0.08728693,\r\n     'total_loss': 0.2591449,\r\n     'training_loss': 0.2591449}\r\ntrain | step:  253344 | steps/sec:    3.8 | output: \r\n    {'box_loss': 0.0017456077,\r\n     'cls_loss': 6.524426e-06,\r\n     'learning_rate': 0.0025,\r\n     'model_loss': 0.08728693,\r\n     'total_loss': 0.2591449,\r\n     'training_loss': 0.2591449}\r\nI0316 12:00:11.730173 139750881177728 controller.py:531] saved checkpoint to /mnt/sda1/exc_cctv/results_retinanet/ckpt-253344.\r\nsaved checkpoint to /mnt/sda1/exc_cctv/results_retinanet/ckpt-253344.\r\nI0316 12:00:11.730983 139750881177728 controller.py:297]  eval | step:  253344 | running complete evaluation...\r\n eval | step:  253344 | running complete evaluation...\r\nINFO:tensorflow:Error reported to Coordinator: Exception encountered when calling layer 'retina_net_model' (type RetinaNetModel).\r\n\r\nin user code:\r\n\r\n    File \"/home/vision/Models/models/official/vision/modeling/retinanet_model.py\", line 169, in call  *\r\n        final_results = self.detection_generator(raw_boxes, raw_scores,\r\n    File \"/home/vision/Models/models/official/vision/modeling/layers/detection_generator.py\", line 1512, in __call__  *\r\n        (nmsed_boxes, nmsed_scores, nmsed_classes, valid_detections) = (\r\n    File \"/home/vision/Models/models/official/vision/modeling/layers/detection_generator.py\", line 588, in _generate_detections_v2  *\r\n        return _generate_detections_v2_class_aware(\r\n    File \"/home/vision/Models/models/official/vision/modeling/layers/detection_generator.py\", line 518, in _generate_detections_v2_class_aware  *\r\n        nmsed_boxes = tf.concat(nmsed_boxes, axis=1)\r\n\r\n    ValueError: List argument 'values' to 'ConcatV2' Op with length 0 shorter than minimum length 2.\r\n\r\n\r\nCall arguments received by layer 'retina_net_model' (type RetinaNetModel):\r\n  \u2022 images=tf.Tensor(shape=(1, 1024, 1024, 3), dtype=float32)\r\n  \u2022 image_shape=tf.Tensor(shape=(1, 2), dtype=float32)\r\n  \u2022 anchor_boxes={'3': 'tf.Tensor(shape=(1, 128, 128, 36), dtype=float32)', '4': 'tf.Tensor(shape=(1, 64, 64, 36), dtype=float32)', '5': 'tf.Tensor(shape=(1, 32, 32, 36), dtype=float32)', '6': 'tf.Tensor(shape=(1, 16, 16, 36), dtype=float32)', '7': 'tf.Tensor(shape=(1, 8, 8, 36), dtype=float32)'}\r\n  \u2022 output_intermediate_features=False\r\n  \u2022 training=False\r\nTraceback (most recent call last):\r\n  File \"/home/vision/anaconda3/envs/tfm/lib/python3.8/site-packages/tensorflow/python/training/coordinator.py\", line 293, in stop_on_exception\r\n    yield\r\n  File \"/home/vision/anaconda3/envs/tfm/lib/python3.8/site-packages/tensorflow/python/distribute/mirrored_run.py\", line 386, in run\r\n    self.main_result = self.main_fn(*self.main_args, **self.main_kwargs)\r\n  File \"/tmp/__autograph_generated_filebc32f0c3.py\", line 17, in step_fn\r\n    logs = ag__.converted_call(ag__.ld(self).task.validation_step, (ag__.ld(inputs),), dict(model=ag__.ld(self).model, metrics=ag__.ld(self).validation_metrics), fscope_1)\r\n  File \"/home/vision/anaconda3/envs/tfm/lib/python3.8/site-packages/tensorflow/python/autograph/impl/api.py\", line 439, in converted_call\r\n    result = converted_f(*effective_args, **kwargs)\r\n  File \"/tmp/__autograph_generated_filefxbwta_h.py\", line 12, in tf__validation_step\r\n    outputs = ag__.converted_call(ag__.ld(model), (ag__.ld(features),), dict(anchor_boxes=ag__.ld(labels)['anchor_boxes'], image_shape=ag__.ld(labels)['image_info'][:, 1, :], training=False), fscope)\r\n  File \"/home/vision/anaconda3/envs/tfm/lib/python3.8/site-packages/tensorflow/python/autograph/impl/api.py\", line 331, in converted_call\r\n    return _call_unconverted(f, args, kwargs, options, False)\r\n  File \"/home/vision/anaconda3/envs/tfm/lib/python3.8/site-packages/tensorflow/python/autograph/impl/api.py\", line 458, in _call_unconverted\r\n    return f(*args, **kwargs)\r\n  File \"/home/vision/anaconda3/envs/tfm/lib/python3.8/site-packages/keras/utils/traceback_utils.py\", line 70, in error_handler\r\n    raise e.with_traceback(filtered_tb) from None\r\n  File \"/tmp/__autograph_generated_fileiwd58oi2.py\", line 228, in tf__call\r\n    ag__.if_stmt(ag__.ld(training), if_body_11, else_body_11, get_state_12, set_state_12, ('do_return', 'final_results', 'retval_', 'anchor_boxes'), 3)\r\n  File \"/tmp/__autograph_generated_fileiwd58oi2.py\", line 156, in else_body_11\r\n    final_results = ag__.converted_call(ag__.ld(self).detection_generator, (ag__.ld(raw_boxes), ag__.ld(raw_scores), ag__.ld(anchor_boxes), ag__.ld(image_shape), ag__.ld(raw_attributes)), None, fscope)\r\n  File \"/tmp/__autograph_generated_filepm9g43ot.py\", line 213, in tf____call__\r\n    ag__.if_stmt(ag__.and_((lambda : ag__.ld(self)._config_dict['apply_nms']), (lambda : (ag__.ld(self)._config_dict['nms_version'] == 'tflite'))), if_body_8, else_body_8, get_state_8, set_state_8, ('do_return', 'retval_'), 2)\r\n  File \"/tmp/__autograph_generated_filepm9g43ot.py\", line 200, in else_body_8\r\n    ag__.if_stmt(ag__.not_(ag__.ld(self)._config_dict['apply_nms']), if_body_7, else_body_7, get_state_7, set_state_7, ('do_return', 'retval_'), 2)\r\n  File \"/tmp/__autograph_generated_filepm9g43ot.py\", line 185, in else_body_7\r\n    ag__.if_stmt((ag__.ld(self)._config_dict['nms_version'] == 'batched'), if_body_6, else_body_6, get_state_6, set_state_6, ('nmsed_attributes', 'nmsed_boxes', 'nmsed_classes', 'nmsed_scores', 'valid_detections'), 5)\r\n  File \"/tmp/__autograph_generated_filepm9g43ot.py\", line 179, in else_body_6\r\n    ag__.if_stmt((ag__.ld(self)._config_dict['nms_version'] == 'v1'), if_body_5, else_body_5, get_state_5, set_state_5, ('nmsed_attributes', 'nmsed_boxes', 'nmsed_classes', 'nmsed_scores', 'valid_detections'), 5)\r\n  File \"/tmp/__autograph_generated_filepm9g43ot.py\", line 173, in else_body_5\r\n    ag__.if_stmt((ag__.ld(self)._config_dict['nms_version'] == 'v2'), if_body_4, else_body_4, get_state_4, set_state_4, ('nmsed_attributes', 'nmsed_boxes', 'nmsed_classes', 'nmsed_scores', 'valid_detections'), 5)\r\n  File \"/tmp/__autograph_generated_filepm9g43ot.py\", line 141, in if_body_4\r\n    (nmsed_boxes, nmsed_scores, nmsed_classes, valid_detections) = ag__.converted_call(ag__.ld(_generate_detections_v2), (ag__.ld(boxes), ag__.ld(scores)), dict(pre_nms_top_k=ag__.ld(self)._config_dict['pre_nms_top_k'], pre_nms_score_threshold=ag__.ld(self)._config_dict['pre_nms_score_threshold'], nms_iou_threshold=ag__.ld(self)._config_dict['nms_iou_threshold'], max_num_detections=ag__.ld(self)._config_dict['max_num_detections'], use_class_agnostic_nms=ag__.ld(self)._config_dict['use_class_agnostic_nms']), fscope)\r\n  File \"/tmp/__autograph_generated_file0tsux9fz.py\", line 36, in tf___generate_detections_v2\r\n    ag__.if_stmt(ag__.ld(use_class_agnostic_nms), if_body, else_body, get_state, set_state, ('do_return', 'retval_'), 2)\r\n  File \"/tmp/__autograph_generated_file0tsux9fz.py\", line 32, in else_body\r\n    retval_ = ag__.converted_call(ag__.ld(_generate_detections_v2_class_aware), (), dict(boxes=ag__.ld(boxes), scores=ag__.ld(scores), pre_nms_top_k=ag__.ld(pre_nms_top_k), pre_nms_score_threshold=ag__.ld(pre_nms_score_threshold), nms_iou_threshold=ag__.ld(nms_iou_threshold), max_num_detections=ag__.ld(max_num_detections)), fscope)\r\n  File \"/tmp/__autograph_generated_file76u_22bw.py\", line 60, in tf___generate_detections_v2_class_aware\r\n    nmsed_boxes = ag__.converted_call(ag__.ld(tf).concat, (ag__.ld(nmsed_boxes),), dict(axis=1), fscope)\r\nValueError: Exception encountered when calling layer 'retina_net_model' (type RetinaNetModel).\r\n```"
    },
    {
      "author": "JunHyungKang",
      "created_at": "2023-03-16T05:17:12Z",
      "body": "@laxmareddyp \r\n\r\nsame issue with colab env.\r\n\r\nnotebook: https://colab.research.google.com/drive/1hzqvTr_rv9w0nM6jGniCWe05Les9jbKp?usp=sharing"
    },
    {
      "author": "laxmareddyp",
      "created_at": "2023-03-16T18:22:39Z",
      "body": "Hi @JunHyungKang,\r\n\r\nThanks for providing the colab code. We suggest you to once go through this [tutorial](https://www.tensorflow.org/tfmodels/vision/object_detection) how to configure a object detection pipeline. I see that you have included a function in `retinanet.py` to register it. Instead of that you can load the experiment configuration using  this line `exp_config = exp_factory.get_exp_config('retinanet_resnetfpn_coco')`. Which will load the configuration required for `retinanet_resnetfpn_coco`.\r\n\r\nNow you can access the experiment configuration like an object and change all required variables in the configuration and use it for training your model with custom dataset and custom configuration. Please use `distribution_strategy.scope()` so that it will take care of the distribution strategy.\r\n\r\nPlease check this [gist](https://colab.research.google.com/gist/sineeli/c738f3394194b3d53b3b811f38b3eaa1/10947.ipynb) which gives a glimpse how you can modify the configuration and make use of it to fuller extent. \r\n\r\nThere may be new commits in git clone which are not yet in stable release, we suggest you to you use pip install so that if there are any on going errors/bugs, they will not be any problems to your training. I hope this will help you resolve the issue.\r\n\r\n\r\nThanks"
    },
    {
      "author": "JunHyungKang",
      "created_at": "2023-03-17T00:35:00Z",
      "body": "Hi @laxmareddyp , \r\n\r\nThank you for your response. However, the notebook file you provided is no different from the train.py file that I worked on and ran immediately. I am using the stable commit version of the master branch, and all the items you executed in the notebook are coded to run in the same order in train.py.\r\n\r\nI have attached the results from running the notebook you provided in the same environment for reference. In my opinion, as I mentioned earlier, the NMS code may need to be modified to consider cases where there are no predicted bounding boxes, but I haven't been able to look deeper into it. Please check the attached results.\r\n\r\nhttps://colab.research.google.com/drive/1XJma_dqu4RgWk-dODd03sm5n6WlToxii?usp=sharing"
    },
    {
      "author": "laxmareddyp",
      "created_at": "2023-03-17T16:52:31Z",
      "body": "Hi @JunHyungKang ,\r\n\r\nYes, its no different from train.py just wanted to write without creating a function. Now I understood the problem, basically NMS is getting null boxes after prediction which has to be min of two length for concat, we will look into it internal and come back with proper resolution. Really thanks for reporting the bug.\r\n\r\nThanks."
    },
    {
      "author": "laxmareddyp",
      "created_at": "2023-03-23T20:07:05Z",
      "body": "Hi @JunHyungKang,\r\n\r\n\r\nCan you please tell me what is the class number that has been declared in the `tfrecords`. Also if possible some `tfrecords` which are dummy and similar to your data would really help us to reproduce the error from our side. I was trying to reproduce with other dataset but not able to do it.\r\n\r\nThanks"
    },
    {
      "author": "JunHyungKang",
      "created_at": "2023-03-24T02:22:14Z",
      "body": "@laxmareddyp \r\nI used one class to generate tfrecords.\r\nplease refer to this [dummy](https://drive.google.com/file/d/1XtI-onOtAf0pXDE9gxYtGTxRHseU4TSJ/view?usp=share_link)."
    },
    {
      "author": "laxmareddyp",
      "created_at": "2023-03-27T17:41:36Z",
      "body": "Hi @JunHyungKang,\r\n\r\nPlease make sure to follow below requirements while creating `tfrecords`.\r\n\r\n'image/encoded' | bytes | Required. The encoded image bytes.\r\n-- | -- | --\r\n'image/source_id' | string | Required. The unique identifier of the image, need to be an number in string.\r\n'image/height' | integer | Optional. The height of the image. If not exisit, inferred from image.\r\n'image/width' | integer | Optional. The width of the image. If not exisit, inferred from image.\r\n'image/object/bbox/xmin' | a list of float | Required. The normalized xmin coordinates of all the instances.\r\n'image/object/bbox/xmax' | a list of float | Required. The normalized xmax coordinates of all the instances.\r\n'image/object/bbox/ymin' | a list of float | Required. The normalized ymin coordinates of all the instances.\r\n'image/object/bbox/ymax' | a list of float | Required. The normalized ymax coordinates of all the instances.\r\n'image/object/class/label' | a list of integer | Required. The class indices of all the instances. Note that 0 is reserved for background.\r\n'image/object/mask' | a list of bytes | Optional. The mask of all the instances in PNG format.\r\n'image/object/area' | a list of float | Optional. The area of all the instances. If not exisit, derived from the bounding boxes.\r\n'image/object/is_crowd' | a list of integer | Optional. The 0/1 integers to denote whether instances are a crowd. The crowd instance get special treatement during the evaluation. 0 (not crowd) by default.\r\n\r\nThe dummy records you provided has a error while evaluation because the source_id has string data with alphbatic characters in it. Please find the [gist](https://colab.sandbox.google.com/gist/sineeli/cd42367449e2eae0687ae3e252df9c3a/10947.ipynb) which I am trying to debug.\r\n\r\nAlso add the classes as 2 because the classes start from 0 in general in the cooc_json. Which can be considered as background if I am not wrong. Please check the below screenshot from [BCCD](https://public.roboflow.com/object-detection/bccd) dataset.\r\n<img width=\"1050\" alt=\"Screenshot 2023-03-27 at 10 30 47 AM\" src=\"https://user-images.githubusercontent.com/113718461/228021266-0731284b-dbf2-4c14-b575-49aa71cb2a30.png\">\r\n\r\nPlease go through this object detection [tutorial](https://www.tensorflow.org/tfmodels/vision/object_detection) if you have not gone through, it uses BCCD dataset and trains the model. Still if the error persists, once you are ready with proper tfrecords format, we are happy to debug and further help you to resolve the issue.\r\n\r\nThanks"
    },
    {
      "author": "github-actions[bot]",
      "created_at": "2023-04-04T01:53:52Z",
      "body": "This issue has been marked stale because it has no recent activity since 7 days. It will be closed if no further activity occurs. Thank you."
    },
    {
      "author": "github-actions[bot]",
      "created_at": "2023-04-12T01:52:34Z",
      "body": "This issue was closed due to lack of activity after being marked stale for past 7 days."
    },
    {
      "author": "google-ml-butler[bot]",
      "created_at": "2023-04-12T01:52:38Z",
      "body": "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/models/issues/10947\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/models/issues/10947\">No</a>\n"
    }
  ]
}