search

balancap / SSD-Tensorflow

Single Shot MultiBox Detector in TensorFlow
4.11k stars 1.89k forks source link

ValueError: No variables to optimize. #311

Closed engineer1109 closed 5 years ago

engineer1109 commented 5 years ago

Traceback (most recent call last): File "train_ssd_network.py", line 390, in tf.app.run() File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/platform/app.py", line 126, in run _sys.exit(main(argv)) File "train_ssd_network.py", line 345, in main var_list=variables_to_train) File "/media/wjl/back/SSD-Tensorflow/deployment/model_deploy.py", line 301, in optimize_clones optimizer, clone, num_clones, regularization_losses, kwargs) File "/media/wjl/back/SSD-Tensorflow/deployment/model_deploy.py", line 265, in _optimize_clone clone_grad = optimizer.compute_gradients(sum_loss, kwargs) File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/optimizer.py", line 520, in compute_gradients raise ValueError("No variables to optimize.") ValueError: No variables to optimize.

Training from VGG error.

engineer1109 commented 5 years ago

It is my fault. I forget to replace all 300 to 512.