search

ethanyanjiali / deep-vision

Learning and replicating famous deep learning models for computer vision tasks
179 stars 52 forks source link

I have trouble (Hourglass training) #13

Open inhwangYang opened 4 years ago

inhwangYang commented 4 years ago

i run train.py i have trouble

tensorflow.python.framework.errors_impl.InvalidArgumentError: 3 root error(s) found. (0) Invalid argument: Key: image/object/parts/y. Data types don't match. Expected type: int64, Actual type: float [[{{node ParseSingleExample/ParseExample/ParseExampleV2}}]] [[MultiDeviceIteratorGetNextFromShard]] [[RemoteCall]] [[IteratorGetNextAsOptional_1]] [[while/body/_19/while/OptionalHasValue/_77]] (1) Invalid argument: Key: image/object/parts/y. Data types don't match. Expected type: int64, Actual type: float [[{{node ParseSingleExample/ParseExample/ParseExampleV2}}]] [[MultiDeviceIteratorGetNextFromShard]] [[RemoteCall]] [[IteratorGetNextAsOptional]] (2) Invalid argument: Key: image/object/parts/y. Data types don't match. Expected type: int64, Actual type: float [[{{node ParseSingleExample/ParseExample/ParseExampleV2}}]] [[MultiDeviceIteratorGetNextFromShard]] [[RemoteCall]] [[IteratorGetNextAsOptional_1]] 0 successful operations. 0 derived errors ignored. [Op:__inference_distributed_train_epoch_290294]

Function call stack: distributed_train_epoch -> distributed_train_epoch -> distributed_train_epoch

which one is problem? help me

ethanyanjiali commented 3 years ago

how did you create your TFRecords? specifically, is image/object/parts/y field in your TFExample a float number of a integer number?