jiapengwu / temp Goto Github PK
View Code? Open in Web Editor NEWTemporal Message Passing Network for Temporal Knowledge Graph Completion
Temporal Message Passing Network for Temporal Knowledge Graph Completion
do you run in linux, i follow the readme but can not run the code
你好,首先非常感谢您提供代码。在复现代码的过程中,发现只有GRRGCN模型使用了impute创新点,BiGRRGCN、SARGCN、BiSARGCN并没有使用impute,是这样么?频率门也是这样的
我debug模型时只有GRRGCN模型停留在断点处了(在RRGCN.py的276行,使用impute的地方),而另外三个模型没有。是我没有找到正确的代码位置么?希望大佬指教,非常感谢
您好,我对您的文章很感兴趣,里面有很多值得研究的内容,我按照您的readme部署,但是一直报错,有如下问题:1、必须通过git才能运行吗?2、是否对设备需求比较大,服务器120G还不太够,能否告知部署代码的设备要求?3、如果这边不能训练的话,能否分享一个训练好的模型以供复现学习?非常期待您的回复!!!
Hi
I'm wondering what is those modes in score function
ex) complex(mode = 'tail')
what does it mean???
whats the difference between mode "tail", mode "head" and mode "single"?
Hi, thank you for sharing your trained model. But when I use the trained model of BiSARGCN-icews14-complex, there would raise an error. It seems like model parameters in the code are not match with the trained model. And there also have the same problem on GRRGCN.
Following are details about traceback. Thanks .
command:python -u test.py --checkpoint-path ./to_upload/BiSARGCN-icews14-complex-8-0.1-time-embed-only-last-layer-random-dropout/version_202004080949
./to_upload/BiSARGCN-icews14-complex-8-0.1-time-embed-only-last-layer-random-dropout/version_202004080949/to_upload-BiSARGCN-icews14-complex-8-0.1-time-embed-only-last-layer-random-dropout-version_202004080949-predictions.pk
Traceback (most recent call last):
File "test.py", line 423, in
predictions = inference()
File "test.py", line 406, in inference
model.load_state_dict(checkpoint['state_dict'])
File "/data/***/conda-env/.conda/envs/py-lighting/lib/python3.6/site-packages/torch/nn/modules/module.py", line 839, in load_state_dict
self.class.name, "\n\t".join(error_msgs)))
RuntimeError: Error(s) in loading state_dict for BiSelfAttentionRGCN:
Unexpected key(s) in state_dict: "ent_encoder.layer_2.q_linear.bias", "ent_encoder.layer_2.v_linear.bias", "ent_encoder.layer_2.k_linear.bias".
command : python -u test.py --checkpoint-path ./to_upload/GRRGCN-icews14-complex-15-0.1-time-embed-only-last-layer-random-dropout/version_202004300606
./to_upload/GRRGCN-icews14-complex-15-0.1-time-embed-only-last-layer-random-dropout/version_202004300606/to_upload-GRRGCN-icews14-complex-15-0.1-time-embed-only-last-layer-random-dropout-version_202004300606-predictions.pk
Traceback (most recent call last):
File "test.py", line 423, in
predictions = inference()
File "test.py", line 406, in inference
model.load_state_dict(checkpoint['state_dict'])
File "/data/***/conda-env/.conda/envs/py-lighting/lib/python3.6/site-packages/torch/nn/modules/module.py", line 839, in load_state_dict
self.class.name, "\n\t".join(error_msgs)))
RuntimeError: Error(s) in loading state_dict for PostDynamicRGCN:
Missing key(s) in state_dict: "subject_query_subject_embed_linear.0.weight", "subject_query_subject_embed_linear.0.bias", "subject_query_subject_embed_linear.2.weight", "subject_query_subject_embed_linear.2.bias", "object_query_subject_embed_linear.0.weight", "object_query_subject_embed_linear.0.bias", "object_query_subject_embed_linear.2.weight", "object_query_subject_embed_linear.2.bias", "subject_query_object_embed_linear.0.weight", "subject_query_object_embed_linear.0.bias", "subject_query_object_embed_linear.2.weight", "subject_query_object_embed_linear.2.bias", "object_query_object_embed_linear.0.weight", "object_query_object_embed_linear.0.bias", "object_query_object_embed_linear.2.weight", "object_query_object_embed_linear.2.bias".
Unexpected key(s) in state_dict: "subject_query_subject_embed_linear.weight", "subject_query_subject_embed_linear.bias", "object_query_subject_embed_linear.weight", "object_query_subject_embed_linear.bias", "subject_query_object_embed_linear.weight", "subject_query_object_embed_linear.bias", "object_query_object_embed_linear.weight", "object_query_object_embed_linear.bias".
When I run your code with your provided config files, some models get stuck in a certain epoch of training. For example SRGCN get stuck in 69 epoch. I hope your can check and run it. Thanks!
Hi ,when I run the code with the command “ python -u main.py -c grid/icews14/config_srgcn.json --fast-dev-run”,there raise an error "AttributeError: 'NoneType' object has no attribute 'module'"in trianner.test() . I found your code has the "test.py". Does the trianner.test() function need to be called in "main.py"? Thanks !
Thank you for your code, but i can't understand this code without readme file.
Thanks for sharing the codes. But actually, it is hard to follow and understand the codes without a useful readme file.
Hi, Jiapeng, thanks for released code. I want to know some detail about SARGCN.
TeMP/models/SelfAttentionRGCN.py
Lines 97 to 102 in 53e90c1
In dataset ICEWS14, ICEWS05-15, do you remove edges containing entity with only a single mention in the subgraph?
Hi
I'm curious about your loss
because when i read your paper, the loss's shape is like
but when i printed your loss in TKG_Module.py, like
def training_step(self, batch_time, batch_idx):
# gc.collect()
loss = self.forward(batch_time)
print("\n",loss.item())
and i see the loss values are all positive values.
is the loss your model using for training exactly same with the paper???
(I'm just a student. I'm just curious about it
so please don't feel bad about the question)
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.