tianbian95 / bigcn Goto Github PK

Hi, I have gone through your code and found that you only optimize the parameter of BUGCN. May I know the reason that you do not optimize the parameter of TUGCN

Could you please release the source code for early rumor detection?Thank you!!!

RuntimeError: scatter_add() expected at most 5 argument(s) but received 6 argument(s). Declaration: scatter_add(Tensor src, Tensor index, int dim=-1, Tensor? out=None, int? dim_size=None) -> (Tensor)
0%| | 0/10 [00:01<?, ?it/s]

Hi. Thank you for your sharing.

In your paper Rumor Detection on Social Media with Bi-Directional Graph Convolutional Networks, it is mentioned that GCN is applied on a graph, which uses source tweet r and its responsive/retweet posts as nodes. But in the dataset Twitter15/16, only source tweet has a label. What are the labels of these responsive/retweet posts? Maybe their labels are consistent with their source tweet?

Many thanks.

There are "index: count" pairs in each row of the preprocessed data file....what do they signify? is it something like the tweet text has been tokenized and based on all the token you have a vocabulary/dictionary with key: count pairs and each row symbolizes the token index: count pair from the vocabulary. Could you please provide more details on how have you preprocessed the dataset?

你好，能否更具体说明一下weibotree.txt每一行4个部分的最后一个部分代表什么？谢谢

I have tried to reduce the batch_size to 2, but I can only run a few of epochs, and I also tried to add th.cuda.empty_cache() in your code.
I run this: python ./model/Weibo/BiGCN_Weibo.py 100

and my envirment is :
GPU : RTX3080.
pytorch 1.4.0
CUDA10.1

sys.argv[2]
*** IndexError: list index out of range

在计算各类准确率的时候有考虑真负（TN），但在考虑综合准确率的时候未考虑真负，导致综合准确率低于各分类准确率的平均

Hi，I run your code with 'python ./model/Twitter/BiGCN_Twitter.py Twitter15 10' to get the average experimental results of 10 iterations of BiGCN model on Twitter15 (running 100 iterations takes too much time).
I got the result 'Total_Test_Accuracy: 0.8582|NR F1: 0.8364|FR F1: 0.8537|TR F1: 0.9099|UR F1: 0.8220', which has a big gap compared with the results reported in your paper '0.886 0.891 0.860 0.930 0.864'.

Hello I'm getting the following error when I run the command 'python ./model/Twitter/BiGCN_Weibo.py 100'
I've tested it on Colab, PyCharm, and by using WindowsShell

loading weibo label:
4664
2351 2313
reading Weibo tree
tree no: 4659
loading train set
train no: 3618
loading test set
test no: 909
0%| | 0/227 [00:00<?, ?it/s]loading weibo label:
4664
2351 2313
reading Weibo tree
tree no: 4659
loading train set
0%| | 0/227 [00:00<?, ?it/s]Traceback (most recent call last):
File "", line 1, in
File "C:\Users\User\AppData\Local\Programs\Python\Python38\lib\multiprocessing\spawn.py", line 116, in spawn_main
exitcode = _main(fd, parent_sentinel)
File "C:\Users\User\AppData\Local\Programs\Python\Python38\lib\multiprocessing\spawn.py", line 125, in _main
prepare(preparation_data)
File "C:\Users\User\AppData\Local\Programs\Python\Python38\lib\multiprocessing\spawn.py", line 236, in prepare
_fixup_main_from_path(data['init_main_from_path'])
File "C:\Users\User\AppData\Local\Programs\Python\Python38\lib\multiprocessing\spawn.py", line 287, in _fixup_main_from_path
main_content = runpy.run_path(main_path,
File "C:\Users\User\AppData\Local\Programs\Python\Python38\lib\runpy.py", line 263, in run_path
return _run_module_code(code, init_globals, run_name,
File "C:\Users\User\AppData\Local\Programs\Python\Python38\lib\runpy.py", line 96, in _run_module_code
_run_code(code, mod_globals, init_globals,
File "C:\Users\User\AppData\Local\Programs\Python\Python38\lib\runpy.py", line 86, in _run_code
exec(code, run_globals)
File "D:\BiGCN-master\BiGCN-master\model\Weibo\BiGCN_Weibo.py", line 216, in
train_losses, val_losses, train_accs, val_accs, accs_0, acc1_0, pre1_0, rec1_0, F1_0, acc2_0, pre2_0, rec2_0, F2_0 = train_GCN(treeDic,
File "D:\BiGCN-master\BiGCN-master\model\Weibo\BiGCN_Weibo.py", line 116, in train_GCN
for Batch_data in tqdm_train_loader:
File "D:\BiGCN-master\BiGCN-master\venv\lib\site-packages\tqdm\std.py", line 1097, in iter
for obj in iterable:
train no: 3626
loading test set
test no: 901
File "D:\BiGCN-master\BiGCN-master\venv\lib\site-packages\torch\utils\data\dataloader.py", line 359, in iter
return self._get_iterator()
File "D:\BiGCN-master\BiGCN-master\venv\lib\site-packages\torch\utils\data\dataloader.py", line 305, in _get_iterator
return _MultiProcessingDataLoaderIter(self)
File "D:\BiGCN-master\BiGCN-master\venv\lib\site-packages\torch\utils\data\dataloader.py", line 918, in init
w.start()
File "C:\Users\User\AppData\Local\Programs\Python\Python38\lib\multiprocessing\process.py", line 121, in start
self._popen = self._Popen(self)
File "C:\Users\User\AppData\Local\Programs\Python\Python38\lib\multiprocessing\context.py", line 224, in _Popen
return _default_context.get_context().Process._Popen(process_obj)
File "C:\Users\User\AppData\Local\Programs\Python\Python38\lib\multiprocessing\context.py", line 326, in _Popen
return Popen(process_obj)
File "C:\Users\User\AppData\Local\Programs\Python\Python38\lib\multiprocessing\popen_spawn_win32.py", line 45, in init
prep_data = spawn.get_preparation_data(process_obj._name)
File "C:\Users\User\AppData\Local\Programs\Python\Python38\lib\multiprocessing\spawn.py", line 154, in get_preparation_data
_check_not_importing_main()
File "C:\Users\User\AppData\Local\Programs\Python\Python38\lib\multiprocessing\spawn.py", line 134, in _check_not_importing_main
raise RuntimeError('''
RuntimeError:
An attempt has been made to start a new process before the
current process has finished its bootstrapping phase.

    This probably means that you are not using fork to start your
    child processes and you have forgotten to use the proper idiom
    in the main module:

        if __name__ == '__main__':
            freeze_support()
            ...

    The "freeze_support()" line can be omitted if the program
    is not going to be frozen to produce an executable.

0%| | 0/227 [00:00<?, ?it/s]

While trying to run "getTwittergraph.py" file I get the error:
No such file or directory: '/BiGCN/Process/data/Twitter15graph/731166399389962242.npz'
How can I fix this?

In each iteration, load5foldData is called to get 5 different datasets. Each dataset, consisting of the training set and the testing set, is used to train and test the model.
But those 5 datasets are generated from the same data. Thus, the data in fold0_x_test to test the model may be in fold1_x_train and it's used to train the model again, which means the model could learn from the fold$n$_x_test in each iteration. The whole data is used for training but there is no other separate testing set to evaluate the model, resulting in significantly higher accuracy.

这里是在交叉验证的测试集做早停了吧？这样没问题吗？而且测试集也还在做梯度更新

local variable 'fold0_x_test' referenced before assignment

How to set the time for early rumor detection? How to divide the data before the time point, or how to read the data before the static point?

对于lable_All文件中的第五六七列的数据含义不明白

在getweibograph.py中，函数loadeid（）中参数有id，但是后面却没有id的任何定义。因此在np.savez()中id无法显示，后面就无法保存