Open xiongfeiliu opened 6 years ago
Hello bro, did you solve it? got the same issue.
@namangt68 I have faced the same error while training my own dataset for tiny-yolov3. Now, I am able to train it on ubuntu 16.04 with below packages: python3: 3.5.2 tensorflow: 1.6.0 keras: 2.15
@manjuraehmad Yes I also resolved this issue with same configuration.
thanks ! @manjuraehmad
Traceback (most recent call last): File "/media/lxf/D/Project/keras-yolo3-master/train.py", line 192, in
_main()
File "/media/lxf/D/Project/keras-yolo3-master/train.py", line 34, in _main
freeze_body=2, weights_path='model_data/yolo_weights.h5')# make sure you know what you freeze
File "/media/lxf/D/Project/keras-yolo3-master/train.py", line 131, in create_model
([model_body.output,y_true])
File "/home/lxf/anaconda3/envs/tensorflow-1.5.0/lib/python2.7/site-packages/keras/engine/base_layer.py", line 414, in call
self.assert_input_compatibility(inputs)
File "/home/lxf/anaconda3/envs/tensorflow-1.5.0/lib/python2.7/site-packages/keras/engine/base_layer.py", line 285, in assert_input_compatibility
str(inputs) + '. All inputs to the layer '
ValueError: Layer yolo_loss was called with an input that isn't a symbolic tensor. Received type: <type 'list'>. Full input: [[<tf.Tensor 'conv2d_59/BiasAdd:0' shape=(?, ?, ?, 18) dtype=float32>, <tf.Tensor 'conv2d_67/BiasAdd:0' shape=(?, ?, ?, 18) dtype=float32>, <tf.Tensor 'conv2d_75/BiasAdd:0' shape=(?, ?, ?, 18) dtype=float32>], [<tf.Tensor 'input_3:0' shape=(?, 13, 13, 3, 6) dtype=float32>]]. All inputs to the layer should be tensors.
what should i do?