InvalidArgumentError for large dataset

[Chen-Cai-OSU]

Hi Tomas,

I am able to use gae on the small dataset(<10k nodes) without any problem but when I tried on large dataset. I am having the following error:

2019-03-15 15:36:48.653709: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA Traceback (most recent call last): File "train.py", line 187, in outs = sess.run([opt.opt_op, opt.cost, opt.accuracy], feed_dict=feed_dict) File "/home/gurukar.1/anaconda2/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 929, in run run_metadata_ptr) File "/home/gurukar.1/anaconda2/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 1152, in _run feed_dict_tensor, options, run_metadata) File "/home/gurukar.1/anaconda2/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 1328, in _do_run run_metadata) File "/home/gurukar.1/anaconda2/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 1348, in _do_call raise type(e)(node_def, op, message) tensorflow.python.framework.errors_impl.InvalidArgumentError: Shape output type is 32-bit but dim 0 is 3187957444 [[node optimizer/gradients/optimizer/logistic_loss/mul_1_grad/Shape_1 (defined at /scratch/line_verse/vag/vgae/gae/optimizer.py:15) = ShapeT=DT_FLOAT, out_type=DT_INT32, _device="/job:localhost/replica:0/task:0/device:CPU:0"]]

I did some research (https://github.com/tensorflow/tensorflow/issues/23107) and it seems that it relates to int32 datatype. I am thinking about change all int32 in optimizer.py from int32 to int64. Do u think it will work? I would like to consult you at first since it takes the quite long time to run on the large dataset.

[ChanakaCooray]

@Chen-Cai-OSU Were you be able to solve this issue?