search

smallcorgi / Faster-RCNN_TF

Faster-RCNN in Tensorflow
MIT License
2.34k stars 1.12k forks source link

NameError: name 'roi_pooling_op' is not defined #306

Open Tower0823 opened 6 years ago

Tower0823 commented 6 years ago

python3.6 + tensorflow 1.2 + cuda-8.0 How can I modify the make.sh to fit my settings? Need help :(

During handling of the above exception, another exception occurred:

Traceback (most recent call last): File "./tools/train_net.py", line 112, in max_iters=args.max_iters) File "/share/home/hh/towerli/dock_lstm/tools/../lib/fast_rcnn/train.py", line 277, in train_net sw.train_model(sess, max_iters) File "/share/home/hh/towerli/dock_lstm/tools/../lib/fast_rcnn/train.py", line 150, in train_model train_op = tf.train.MomentumOptimizer(lr, momentum).minimize(loss, global_step=global_step) File "/share/home/hh/anaconda3/lib/python3.6/site-packages/tensorflow/python/training/optimizer.py", line 315, in minimize grad_loss=grad_loss) File "/share/home/hh/anaconda3/lib/python3.6/site-packages/tensorflow/python/training/optimizer.py", line 386, in compute_gradients colocate_gradients_with_ops=colocate_gradients_with_ops) File "/share/home/hh/anaconda3/lib/python3.6/site-packages/tensorflow/python/ops/gradients_impl.py", line 540, in gradients grad_scope, op, func_call, lambda: grad_fn(op, out_grads)) File "/share/home/hh/anaconda3/lib/python3.6/site-packages/tensorflow/python/ops/gradients_impl.py", line 346, in _MaybeCompile return grad_fn() # Exit early File "/share/home/hh/anaconda3/lib/python3.6/site-packages/tensorflow/python/ops/gradients_impl.py", line 540, in grad_scope, op, func_call, lambda: grad_fn(op, out_grads)) File "/share/home/hh/towerli/dock_lstm/tools/../lib/roi_pooling_layer/roi_pooling_op_grad.py", line 41, in _roi_pool_grad data_grad = roi_pooling_op.roi_pool_grad(data, rois, argmax, grad, pooled_height, pooled_width, spatial_scale) NameError: name 'roi_pooling_op' is not defined

surajitsaikia27 commented 6 years ago

This repository is python2.7 compatible. Did you try converting it to python3 before compiling?

YCIrving commented 6 years ago

That maybe a problem of import which is different between Python2 and 3. Try to use script 2to3 in python to convert the whole project before debugging if you want to run it with python3.

Here's an example in the begining of _lib/roi_pooling_layer/roi_pooling_opgrad.py: original code: # import roi_pooling_op after converting: from . import roi_pooling_op

Anyway, you can refer to my repository which is converted to python3 and works in TF 1.6