search

Saoge123 / ccgnet

MIT License
38 stars 12 forks source link

https://github.com/Saoge123/ccgnet #5

Closed sutapaucd closed 1 year ago

sutapaucd commented 1 year ago

Hi, In CCGnet-OB directory, I was trying to execute predict.py. However, I got the following error: File "predict.py", line 99, in make_graph_dataset self.data_attr_names = D[0].keys() AttributeError: 'NoneType' object has no attribute 'keys'

Also, I am using ccgnet-main/data/Test/Test_Table.tab and ccgnet-main/data/Test/coformers as pwd of input files. If you could please suggest probable solution. Thanks.

y31apw00 commented 1 year ago

Hi sutapaucd, I am a new user of CCGNet and I met the same problem with you.

Using the example table and files as input to Predict.py in CCGnet-OB showed "AttributeError: 'NoneType' object has no attribute 'keys'". I tried to trace source code and input file. Input data can be opened at initial code of "class Dataset(object)". I do not know why the data structure seems empty at "make_graph_dataset".

Do you have any ideas about what's going on?

Saoge123 commented 1 year ago

Hi: I am sorry for the late replay. We recommand to use the codes in origin ccgnet instead of ccgnet-OB, because the ccgnet-OB can not reproduce the results in the artical. You can try using the predict.py in origin ccgnet to the ccgnet-OB. Please contact us if you have any questions.

sutapaucd commented 1 year ago

Hi, Even I use ccgnet-ob, I got the same attibute none type error. Thanks.

On Wed, 21 Jun 2023, 07:27 Saoge123, @.***> wrote:

Hi: I am sorry for the late replay. We recommand to use the codes in origin ccgnet instead of ccgnet-OB, because the ccgnet-OB can not reproduce the results in the artical. You can try using the predict.py in origin ccgnet to the ccgnet-OB. Please contact us if you have any questions.

— Reply to this email directly, view it on GitHub https://github.com/Saoge123/ccgnet/issues/5#issuecomment-1599954280, or unsubscribe https://github.com/notifications/unsubscribe-auth/A75IDO6NHRMIVIUHKZZESF3XMJIKFANCNFSM6AAAAAAY5S6TBU . You are receiving this because you authored the thread.Message ID: @.***>

sutapaucd commented 1 year ago

Hi, so which version of ccdc python api is compatiable with ccgnet.

sutapaucd commented 1 year ago

Which version of ccdc python api is compatible with ccgnet?

On Wed, 21 Jun 2023, 07:29 sutapa dutta, @.***> wrote:

Hi, Even I use ccgnet-ob, I got the same attibute none type error. Thanks.

On Wed, 21 Jun 2023, 07:27 Saoge123, @.***> wrote:

Hi: I am sorry for the late replay. We recommand to use the codes in origin ccgnet instead of ccgnet-OB, because the ccgnet-OB can not reproduce the results in the artical. You can try using the predict.py in origin ccgnet to the ccgnet-OB. Please contact us if you have any questions.

— Reply to this email directly, view it on GitHub https://github.com/Saoge123/ccgnet/issues/5#issuecomment-1599954280, or unsubscribe https://github.com/notifications/unsubscribe-auth/A75IDO6NHRMIVIUHKZZESF3XMJIKFANCNFSM6AAAAAAY5S6TBU . You are receiving this because you authored the thread.Message ID: @.***>

Saoge123 commented 1 year ago

ccgnet use the 2021 version of CSD，so you can find the corresponding version of ccdc.

---Original--- From: @.> Date: Wed, Jun 21, 2023 17:20 PM To: @.>; Cc: @.**@.>; Subject: Re: [Saoge123/ccgnet] https://github.com/Saoge123/ccgnet (Issue #5)

Which version of ccdc python api is compatible with ccgnet?

On Wed, 21 Jun 2023, 07:29 sutapa dutta, @.***> wrote:

> Hi, > Even I use ccgnet-ob, I got the same attibute none type error. > Thanks. > > On Wed, 21 Jun 2023, 07:27 Saoge123, @.> wrote: > >> Hi: >> I am sorry for the late replay. We recommand to use the codes in origin >> ccgnet instead of ccgnet-OB, because the ccgnet-OB can not reproduce the >> results in the artical. You can try using the predict.py in origin ccgnet >> to the ccgnet-OB. Please contact us if you have any questions. >> >> — >> Reply to this email directly, view it on GitHub >> <https://github.com/Saoge123/ccgnet/issues/5#issuecomment-1599954280&gt;, >> or unsubscribe >> <https://github.com/notifications/unsubscribe-auth/A75IDO6NHRMIVIUHKZZESF3XMJIKFANCNFSM6AAAAAAY5S6TBU&gt; >> . >> You are receiving this because you authored the thread.Message ID: >> @.> >> >

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.***>