hitlic
commented
4 years ago I'm not sure what the problem is based on your information. I don't have a tf1 environment now, but I had implemented a tf2 version of DGCNN, which I hope will be useful to you. https://github.com/hitlic/SEAL-tf2/blob/master/dgcnn_models.py
hello, i have met one problem,can you help me? i want to use the node feature and construction ,eg the Synthie.And when i run the code,it shows :
2020-08-31 11:33:51,451 - loading data class_num 4 conv1d_channels [15, 30] conv1d_kernel_size [0, 5] dense_dim 128 feature_dim 16 gcnn_dims [30, 30, 30, 1] k 30 keep_prob 1 learning_rate 0.0001 node_feature_dim 15 node_label_dim 1
2020-08-31 11:33:52,089 - 训练数据量 360，测试数据量 40 F:\Anaconda3\envs\tensor112\lib\site-packages\tensorflow\python\ops\gradients_impl.py:112: UserWarning: Converting sparse IndexedSlices to a dense Tensor of unknown shape. This may consume a large amount of memory. "Converting sparse IndexedSlices to a dense Tensor of unknown shape. " 2020-08-31 11:33:52.699053: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 Traceback (most recent call last): File "D:/work/code/DGCNN-tensorflow1.12.0-master/DGCNN-tensorflow-master/train.py", line 108, in
loss, acc, auc = loop_dataset(model, params, train_set, sess, batch_size)
File "D:/work/code/DGCNN-tensorflow1.12.0-master/DGCNN-tensorflow-master/train.py", line 55, in loopdataset
predicts, loss, acc, = sess.run(to_run, feed_dict=feed_dict)
File "F:\Anaconda3\envs\tensor112\lib\site-packages\tensorflow\python\client\session.py", line 929, in run
run_metadata_ptr)
File "F:\Anaconda3\envs\tensor112\lib\site-packages\tensorflow\python\client\session.py", line 1128, in _run
str(subfeed_t.get_shape())))
ValueError: Cannot feed value of shape (4750, 15) for Tensor 'place_hoders/batch_node_features:0', which has shape '(?, 16)'
Process finished with exit code 1
i add the params.set("feature_dim", 15) in step0 ,but it stlii no use. thx