I was trying to implement a object detection on a custom dataset using tensorflow

user_13686418 0
Viewed 609

I am new to tensorflow and I was following this [link]https://tensorflow-object-detection-api-tutorial.readthedocs.io/en/latest/training.html

I ran the python model_main_tf2.py --model_dir=models/my_ssd_resnet50_v1_fpn --pipeline_config_path=models/my_ssd_resnet50_v1_fpn/pipeline.config

and I was getting a error like this and the program stopped.

WARNING:tensorflow:A checkpoint was restored (e.g. tf.train.Checkpoint.restore or tf.keras.Model.load_weights) but not all checkpointed values were used. See above for specific issue s. Use expect_partial() on the load status object, e.g. tf.train.Checkpoint.restore(...).expect_partial(), to silence these warnings, or use assert_consumed() to make the check expli cit. See https://www.tensorflow.org/guide/checkpoint#loading_mechanics for details.

Here are some errors

W0817 10:14:35.677110 16272 util.py:150] Unresolved object in checkpoint: (root).model._feature_extractor._fpn_features_generator.conv_layers.0.1.moving_variance
WARNING:tensorflow:Unresolved object in checkpoint: (root).model._feature_extractor._fpn_features_generator.conv_layers.1.0.kernel
W0817 10:14:35.682097 16272 util.py:150] Unresolved object in checkpoint: (root).model._feature_extractor._fpn_features_generator.conv_layers.1.0.kernel
WARNING:tensorflow:Unresolved object in checkpoint: (root).model._feature_extractor._fpn_features_generator.conv_layers.1.1.axis
W0817 10:14:35.682097 16272 util.py:150] Unresolved object in checkpoint: (root).model._feature_extractor._fpn_features_generator.conv_layers.1.1.axis
WARNING:tensorflow:Unresolved object in checkpoint: (root).model._feature_extractor._fpn_features_generator.conv_layers.1.1.gamma
W0817 10:14:35.683094 16272 util.py:150] Unresolved object in checkpoint: (root).model._feature_extractor._fpn_features_generator.conv_layers.1.1.gamma
WARNING:tensorflow:Unresolved object in checkpoint: (root).model._feature_extractor._fpn_features_generator.conv_layers.1.1.beta
W0817 10:14:35.683094 16272 util.py:150] Unresolved object in checkpoint: (root).model._feature_extractor._fpn_features_generator.conv_layers.1.1.beta
WARNING:tensorflow:Unresolved object in checkpoint: (root).model._feature_extractor._fpn_features_generator.conv_layers.1.1.moving_mean
W0817 10:14:35.683094 16272 util.py:150] Unresolved object in checkpoint: (root).model._feature_extractor._fpn_features_generator.conv_layers.1.1.moving_mean
WARNING:tensorflow:Unresolved object in checkpoint: (root).model._feature_extractor._fpn_features_generator.conv_layers.1.1.moving_variance
W0817 10:14:35.684094 16272 util.py:150] Unresolved object in checkpoint: (root).model._feature_extractor._fpn_features_generator.conv_layers.1.1.moving_variance
WARNING:tensorflow:A checkpoint was restored (e.g. tf.train.Checkpoint.restore or tf.keras.Model.load_weights) but not all checkpointed values were used. See above for specific issue
s. Use expect_partial() on the load status object, e.g. tf.train.Checkpoint.restore(...).expect_partial(), to silence these warnings, or use assert_consumed() to make the check expli
cit. See https://www.tensorflow.org/guide/checkpoint#loading_mechanics for details.
W0817 10:14:35.684094 16272 util.py:158] A checkpoint was restored (e.g. tf.train.Checkpoint.restore or tf.keras.Model.load_weights) but not all checkpointed values were used. See ab
ove for specific issues. Use expect_partial() on the load status object, e.g. tf.train.Checkpoint.restore(...).expect_partial(), to silence these warnings, or use assert_consumed() t
o make the check explicit. See https://www.tensorflow.org/guide/checkpoint#loading_mechanics for details.

some background: I am using windows 10 I installed tensorflow gpu I am using tenforflow 2.3.0

object-detection-api tensorflow2.0 tensorflow python
1 Answers

When I removed all the previously present checkpoints from the models folder inside the workspace, it started to work for me.

user_164200040
Related