cripac-dig / dgsr Goto Github PK

Here is the command I used and the execution result:
%run new_main.py --data=Games
--gpu=0
--epoch=20
--hidden_size=50
--batchSize=50
--user_long=orgat
--user_short=last
--item_long=orgat
--item_short=last
--user_update=rnn
--item_update=rnn
--lr=0.001
--l2=0.0001
--layer_num=2
--item_max_length=50
--user_max_length=50
--attn_drop=0.3
--feat_drop=0.3
--record \

RuntimeError Traceback (most recent call last)
/content/drive/MyDrive/TCSS556/DGSR-master/new_main.py in ()
112 for user, batch_graph, label, last_item in train_data:
113 iter += 1
--> 114 score = model(batch_graph.to(device), user.to(device), last_item.to(device), is_training=True)
115 loss = loss_func(score, label.to(device))
116 optimizer.zero_grad()

8 frames
/content/drive/MyDrive/TCSS556/DGSR-master/DGSR.py in user_reduce_func(self, nodes)
272 return {'user_h': h[0]}
273 else:
--> 274 return {'user_h': self.agg_gate_u(torch.cat(h,-1))}
275
276 def graph_user(bg, user_index, user_embedding):

RuntimeError: Tensors must have same number of dimensions: got 2 and 1

您好，想问一下No such file or directory: 'Beauty_neg'，Beauty_neg文件怎么生成，运行generate_neg文件没有生成对应的文件

老师您好，我搜的最新版本的dgl是0.6.1，没有找到dgl 0.7.2版本，请问您是在哪下载的

Hello, I note that in new_data.py, when it start to create dynamic sub-graph, it skip the first timestamp(i.e. if j == 0: continue). I dont't know why it shoud be skipped. I would appreciate that if you can answer my question!!!

Beauty只能跑到NDGG@10:0.3264，movie只能跑到NDGG@10:0.51

老师您好，我在执行new_data.py时，出现了以下问题：
terminate called after throwing an instance of 'std::bad_alloc'
what(): std::bad_alloc
Using backend: pytorch
terminate called after throwing an instance of 'std::bad_alloc'
what(): std::bad_alloc
terminate called after throwing an instance of 'std::bad_alloc'
what(): std::bad_alloc
terminate called after throwing an instance of 'std::bad_alloc'
what(): std::bad_alloc
exception calling callback for <Future at 0x7f068bb4c610 state=finished raised TerminatedWorkerError>
Traceback (most recent call last):
File "/opt/conda/lib/python3.7/site-packages/joblib/externals/loky/_base.py", line 625, in _invoke_callbacks
callback(self)
File "/opt/conda/lib/python3.7/site-packages/joblib/parallel.py", line 347, in call
self.parallel.dispatch_next()
File "/opt/conda/lib/python3.7/site-packages/joblib/parallel.py", line 780, in dispatch_next
if not self.dispatch_one_batch(self._original_iterator):
File "/opt/conda/lib/python3.7/site-packages/joblib/parallel.py", line 847, in dispatch_one_batch
self._dispatch(tasks)
File "/opt/conda/lib/python3.7/site-packages/joblib/parallel.py", line 765, in _dispatch
job = self._backend.apply_async(batch, callback=cb)
File "/opt/conda/lib/python3.7/site-packages/joblib/_parallel_backends.py", line 529, in apply_async
future = self._workers.submit(SafeFunction(func))
File "/opt/conda/lib/python3.7/site-packages/joblib/externals/loky/reusable_executor.py", line 178, in submit
fn, *args, **kwargs)
File "/opt/conda/lib/python3.7/site-packages/joblib/externals/loky/process_executor.py", line 1102, in submit
raise self._flags.broken
joblib.externals.loky.process_executor.TerminatedWorkerError: A worker process managed by the executor was unexpectedly terminated. This could be caused by a segmentation fault while calling the function or by an excessive memory usage causing the Operating System to kill the worker.

The exit codes of the workers are {SIGABRT(-6)}
Traceback (most recent call last):
File "new_data.py", line 161, in
all_num = generate_data(data, graph, opt.item_max_length, opt.user_max_length, train_path, test_path, val_path, job=opt.job, k_hop=opt.job)
File "new_data.py", line 129, in generate_data
a = Parallel(n_jobs=job)(delayed(lambda u: generate_user(u, data, graph, item_max_length, user_max_length, train_path, test_path, k_hop, val_path))(u) for u in user)
File "/opt/conda/lib/python3.7/site-packages/joblib/parallel.py", line 1042, in call
self.retrieve()
File "/opt/conda/lib/python3.7/site-packages/joblib/parallel.py", line 921, in retrieve
self._output.extend(job.get(timeout=self.timeout))
File "/opt/conda/lib/python3.7/site-packages/joblib/_parallel_backends.py", line 540, in wrap_future_result
return future.result(timeout=timeout)
File "/opt/conda/lib/python3.7/concurrent/futures/_base.py", line 435, in result
return self.__get_result()
File "/opt/conda/lib/python3.7/concurrent/futures/_base.py", line 384, in __get_result
raise self._exception
File "/opt/conda/lib/python3.7/site-packages/joblib/externals/loky/_base.py", line 625, in _invoke_callbacks
callback(self)
File "/opt/conda/lib/python3.7/site-packages/joblib/parallel.py", line 347, in call
self.parallel.dispatch_next()
File "/opt/conda/lib/python3.7/site-packages/joblib/parallel.py", line 780, in dispatch_next
if not self.dispatch_one_batch(self._original_iterator):
File "/opt/conda/lib/python3.7/site-packages/joblib/parallel.py", line 847, in dispatch_one_batch
self._dispatch(tasks)
File "/opt/conda/lib/python3.7/site-packages/joblib/parallel.py", line 765, in _dispatch
job = self._backend.apply_async(batch, callback=cb)
File "/opt/conda/lib/python3.7/site-packages/joblib/_parallel_backends.py", line 529, in apply_async
future = self._workers.submit(SafeFunction(func))
File "/opt/conda/lib/python3.7/site-packages/joblib/externals/loky/reusable_executor.py", line 178, in submit
fn, *args, **kwargs)
File "/opt/conda/lib/python3.7/site-packages/joblib/externals/loky/process_executor.py", line 1102, in submit
raise self._flags.broken
joblib.externals.loky.process_executor.TerminatedWorkerError: A worker process managed by the executor was unexpectedly terminated. This could be caused by a segmentation fault while calling the function or by an excessive memory usage causing the Operating System to kill the worker.

The exit codes of the workers are {SIGABRT(-6)}
这似乎是一个内存溢出问题，我尝试创建单个用户的图可以成功，但这样效率太低了，想请问下老师有没有解决的建议？

I received the following error during training and predicting based on the Games dataset.
'''
start training: 2022-04-25 23:07:12.871280
Iter 400, loss 9.3309 2022-04-25 23:17:20.633740
Iter 800, loss 9.1787 2022-04-25 23:27:15.086462
Epoch 0, loss 9.1725 =============================================
start predicting: 2022-04-25 23:27:29.583910

KeyError Traceback (most recent call last)
/content/drive/MyDrive/TCSS556/DGSR-master/new_main.py in ()
145 all_loss = []
146 with torch.no_grad():
--> 147 for user, batch_graph, label, last_item, neg_tar in test_data:
148 iter+=1
149 score, top = model(batch_graph.to(device), user.to(device), last_item.to(device), neg_tar=torch.cat([label.unsqueeze(1), neg_tar],-1).to(device), is_training=False)

3 frames
/usr/local/lib/python3.7/dist-packages/torch/_utils.py in reraise(self)
455 # instantiate since we don't know how to
456 raise RuntimeError(msg) from None
--> 457 raise exception
458
459

KeyError: Caught KeyError in DataLoader worker process 0.
Original Traceback (most recent call last):
File "/usr/local/lib/python3.7/dist-packages/pandas/core/indexes/base.py", line 3361, in get_loc
return self._engine.get_loc(casted_key)
File "pandas/_libs/index.pyx", line 76, in pandas._libs.index.IndexEngine.get_loc
File "pandas/_libs/index_class_helper.pxi", line 105, in pandas._libs.index.Int64Engine._check_type
File "pandas/_libs/index_class_helper.pxi", line 105, in pandas._libs.index.Int64Engine._check_type
KeyError: tensor([0])

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "/usr/local/lib/python3.7/dist-packages/torch/utils/data/_utils/worker.py", line 287, in _worker_loop
data = fetcher.fetch(index)
File "/usr/local/lib/python3.7/dist-packages/torch/utils/data/_utils/fetch.py", line 52, in fetch
return self.collate_fn(data)
File "/content/drive/MyDrive/TCSS556/DGSR-master/new_main.py", line 92, in
test_data = DataLoader(dataset=test_set, batch_size=opt.batchSize, collate_fn=lambda x: collate_test(x, data_neg), pin_memory=True, num_workers=8)
File "/content/drive/MyDrive/TCSS556/DGSR-master/DGSR.py", line 341, in collate_test
return torch.tensor(user).long(), dgl.batch(graph), torch.tensor(label).long(), torch.tensor(last_item).long(), torch.Tensor(neg_generate(user, user_neg)).long()
File "/content/drive/MyDrive/TCSS556/DGSR-master/DGSR.py", line 326, in neg_generate
neg[i] = np.random.choice(data_neg[u], neg_num, replace=False)
File "/usr/local/lib/python3.7/dist-packages/pandas/core/series.py", line 942, in getitem
return self._get_value(key)
File "/usr/local/lib/python3.7/dist-packages/pandas/core/series.py", line 1051, in _get_value
loc = self.index.get_loc(label)
File "/usr/local/lib/python3.7/dist-packages/pandas/core/indexes/base.py", line 3363, in get_loc
raise KeyError(key) from err
KeyError: tensor([0])
'''
Here is all the steps I did to produce this error. I ran the code in google colab.
'''
%run new_data.py --data=Games --job=10 --item_max_length=50 --user_max_length=50 --k_hop=2
%run generate_neg.py
%run new_main.py --data=Games
--gpu=0
--epoch=20
--hidden_size=50
--batchSize=50
--user_long=orgat
--user_short=att
--item_long=orgat
--item_short=att
--user_update=rnn
--item_update=rnn
--lr=0.001
--l2=0.0001
--layer_num=2
--item_max_length=50
--user_max_length=50
--attn_drop=0.3
--feat_drop=0.3
--record \

您好，python 3.8, torch1.7, dgl 0.7.2, 显卡24G，内存160G， 请问，我在跑这个代码中的new_data.py时，用Movie数据集，只取样50条样本的时候能通过，100条样本就会出现下面报错，是什么原因只能跑这么少的样本呢？大家有没有遇上类似的问题，急求解决，谢谢

terminate called after throwing an instance of 'dmlc::Error'
what(): [11:07:45] /opt/dgl/src/array/cpu/./rowwise_pick.h:89: Check failed: rid < mat.num_rows (2 vs. 1) :
Stack trace:
[bt] (0) /root/miniconda3/lib/python3.8/site-packages/dgl/libdgl.so(dmlc::LogMessageFatal::~LogMessageFatal()+0x4f) [0x7f9ab4cb332f]
[bt] (1) /root/miniconda3/lib/python3.8/site-packages/dgl/libdgl.so(+0x59b122) [0x7f9ab4d03122]
[bt] (2) /root/miniconda3/lib/python3.8/site-packages/torch/lib/libgomp-7c85b1e2.so.1(GOMP_parallel+0x3f) [0x7f9bba37b01f]
[bt] (3) /root/miniconda3/lib/python3.8/site-packages/dgl/libdgl.so(dgl::aten::COOMatrix dgl::aten::impl::CSRRowWisePick(dgl::aten::CSRMatrix, dgl::runtime::NDArray, long, bool, std::function<void (long, long, long, long const*, long const*, long*)>)+0x29a) [0x7f9ab4d0353a]
[bt] (4) /root/miniconda3/lib/python3.8/site-packages/dgl/libdgl.so(dgl::aten::COOMatrix dgl::aten::impl::CSRRowWiseTopk<(DLDeviceType)1, long, long>(dgl::aten::CSRMatrix, dgl::runtime::NDArray, long, dgl::runtime::NDArray, bool)+0x133) [0x7f9ab4d0c273]
[bt] (5) /root/miniconda3/lib/python3.8/site-packages/dgl/libdgl.so(dgl::aten::CSRRowWiseTopk(dgl::aten::CSRMatrix, dgl::runtime::NDArray, long, dgl::runtime::NDArray, bool)+0x426) [0x7f9ab4c93ef6]
[bt] (6) /root/miniconda3/lib/python3.8/site-packages/dgl/libdgl.so(dgl::sampling::SampleNeighborsTopk(std::shared_ptrdgl::BaseHeteroGraph, std::vector<dgl::runtime::NDArray, std::allocatordgl::runtime::NDArray > const&, std::vector<long, std::allocator > const&, dgl::EdgeDir, std::vector<dgl::runtime::NDArray, std::allocatordgl::runtime::NDArray > const&, bool)+0x1364) [0x7f9ab547fc04]
[bt] (7) /root/miniconda3/lib/python3.8/site-packages/dgl/libdgl.so(+0xd1bf3a) [0x7f9ab5483f3a]
[bt] (8) /root/miniconda3/lib/python3.8/site-packages/dgl/libdgl.so(+0xd1c694) [0x7f9ab5484694]

Aborted (core dumped)

作者你好，在阅读代码过程中，发现在为某一个用户构建动态图的过程可能会把别的用户的target item也会包含进去，这是否会导致数据泄露？

Hello,

Thank you for the very interesting paper, really cool concept.

I had a question about how the Train/Validation/Test split happens in the code:

https://github.com/ZM7/DGSR/blob/2293cbdf6e043e9afc02d313e15458124dbd31b7/new_data.py#L112-L123

It seems to me that using this logic:

• items of time == 0 ... time == split_point - 2 are added to the train data
• items of time == split_point - 3 is added to the validation data
• items of time != split_point - 3 is added to the test data

This would mean that the items in the training data are also added to the test data.

In the following table this is illustrated further:

This seems very odd. I was wondering if this was intended or a bug?

Hello,

Thanks again for the interesting research. We were further investigating the paper and I had a question.

Is the weight matrix 'W1' in equation 5 the same as the weight matrix 'W1' in equation 7? And is this also the same as in equation 14?

I was under the the impression this was true, however in equation 5 and 7 the weights are multiplied with the item embedding and in equation 14 it is multiplied with user embedding.

Thank you for your research and kind regards,
Linkerbrain

您好，我想请问一下运行 new_data.py 时会出现 ‘No such file or directory: './Data/sample.csv'的问题要怎么解决呢？

Traceback (most recent call last):
File "new_main.py", line 120, in
for user, batch_graph, label, last_item, neg_tar in test_data:
File "/usr/scratch/yjiang400/miniconda3/envs/dgsr/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 521, in next
data = self._next_data()
File "/usr/scratch/yjiang400/miniconda3/envs/dgsr/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 1203, in _next_data
return self._process_data(data)
File "/usr/scratch/yjiang400/miniconda3/envs/dgsr/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 1229, in _process_data
data.reraise()
File "/usr/scratch/yjiang400/miniconda3/envs/dgsr/lib/python3.6/site-packages/torch/_utils.py", line 434, in reraise
raise exception
KeyError: Caught KeyError in DataLoader worker process 0.
Original Traceback (most recent call last):
File "/usr/scratch/yjiang400/miniconda3/envs/dgsr/lib/python3.6/site-packages/torch/utils/data/_utils/worker.py", line 287, in _worker_loop
data = fetcher.fetch(index)
File "/usr/scratch/yjiang400/miniconda3/envs/dgsr/lib/python3.6/site-packages/torch/utils/data/_utils/fetch.py", line 52, in fetch
return self.collate_fn(data)
File "new_main.py", line 92, in
test_data = DataLoader(dataset=test_set, batch_size=opt.batchSize, collate_fn=lambda x: collate_test(x, data_neg), pin_memory=True, num_workers=8)
File "/export/hdd/scratch/yjiang400/DGSR/DGSR.py", line 341, in collate_test
return torch.tensor(user).long(), dgl.batch(graph), torch.tensor(label).long(), torch.tensor(last_item).long(), torch.Tensor(neg_generate(user, user_neg)).long()
File "/export/hdd/scratch/yjiang400/DGSR/DGSR.py", line 326, in neg_generate
neg[i] = np.random.choice(data_neg[u], neg_num, replace=False)
File "/usr/scratch/yjiang400/miniconda3/envs/dgsr/lib/python3.6/site-packages/pandas/core/series.py", line 871, in getitem
result = self.index.get_value(self, key)
File "/usr/scratch/yjiang400/miniconda3/envs/dgsr/lib/python3.6/site-packages/pandas/core/indexes/base.py", line 4405, in get_value
return self._engine.get_value(s, k, tz=getattr(series.dtype, "tz", None))
File "pandas/_libs/index.pyx", line 80, in pandas._libs.index.IndexEngine.get_value
File "pandas/_libs/index.pyx", line 90, in pandas._libs.index.IndexEngine.get_value
File "pandas/_libs/index.pyx", line 135, in pandas._libs.index.IndexEngine.get_loc
File "pandas/_libs/index_class_helper.pxi", line 109, in pandas._libs.index.Int64Engine._check_type
KeyError: tensor([0])