search

CharlesShang / TFFRCNN

FastER RCNN built on tensorflow
MIT License
875 stars 418 forks source link

Can't find Tensors in Resnet50 when restoring for the demo. #50

Closed louisquinn closed 7 years ago

louisquinn commented 7 years ago

Getting this error when using saver.restore(). Is there some inconsistency with Resnet50_test.py and this checkpoint file?

NotFoundError (see above for traceback): Tensor name "res5c_branch2c/biases" not found in checkpoint files Resnet_iter_200000.ckpt [[Node: save/RestoreV2_320 = RestoreV2[dtypes=[DT_FLOAT], _device="/job:localhost/replica:0/task:0/cpu:0"](_recv_save/Const_0, save/RestoreV2_320/tensor_names, save/RestoreV2_320/shape_and_slices)]] [[Node: save/RestoreV2_218/_641 = _Recv[client_terminated=false, recv_device="/job:localhost/replica:0/task:0/gpu:0", send_device="/job:localhost/replica:0/task:0/cpu:0", send_device_incarnation=1, tensor_name="edge_1627_save/RestoreV2_218", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/gpu:0"]()]]

louisquinn commented 7 years ago

I worked it out, I was incorrectly assigning biases in the network.py script.