1
運行TensorBox複檢實施
我想我的機器上訓練TensorBox複檢實現（https://github.com/Russell91/TensorBox/）與一個GPU（NVIDIA公司的GeForce GTX 750 Ti）的。 當我運行train.py腳本：錯誤而在TensorFlow
python train.py --hypes hypes/overfeat_rezoom.json --gpu 0 --logdir output
我得到以下錯誤：
Traceback (most recent call last):
File "train.py", line 543, in <module>
main()
File "train.py", line 540, in main
train(H, test_images=[])
File "train.py", line 453, in train
smooth_op, global_step, learning_rate, encoder_net) = build(H, q)
File "train.py", line 346, in build
boxes_loss[phase]) = build_forward_backward(H, x, encoder_net, phase, boxes, flags)
File "train.py", line 245, in build_forward_backward
pred_confidences, pred_confs_deltas, pred_boxes_deltas) = build_forward(H, x, googlenet, phase, reuse)
File "train.py", line 167, in build_forward
lstm_outputs = build_lstm_inner(H, lstm_input)
File "train.py", line 50, in build_lstm_inner
state = tf.zeros([batch_size, lstm.state_size])
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/ops/array_ops.py", line 1136, in zeros
shape = ops.convert_to_tensor(shape, dtype=dtypes.int32, name="shape")
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/ops.py", line 628, in convert_to_tensor
ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/constant_op.py", line 180, in _constant_tensor_conversion_function
return constant(v, dtype=dtype, name=name)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/constant_op.py", line 163, in constant
tensor_util.make_tensor_proto(value, dtype=dtype, shape=shape))
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/tensor_util.py", line 354, in make_tensor_proto
nparray = np.array(values, dtype=np_dt)
ValueError: setting an array element with a sequence.
如果我運行的機器上相同的代碼沒有GPU，只是CPU上不會發生這種情況。
什麼可以此錯誤的原因是什麼？有什麼方法可以調試它嗎？