where is the data?

[baiziyuandyufei]

hello! I run your project in my computer, the command I runned is
python run.py train
but the program prompt
No such file or directory: 'glove_word2id
I think the code not include the data file, is it?
Thank you!

[manueltonneau]

Before running the run.py file, you have to run the glove_embedding.py file on the raw data. This will produce two files needed for run.py: glove_embeddings.npy and a glove_word2id dictionary.

[changtianluckyforever]

Before running the run.py file, you have to run the glove_embedding.py file on the raw data. This will produce two files needed for run.py: glove_embeddings.npy and a glove_word2id dictionary.

Hello! Excuse me, mananeau, I am also reading this project repository, in the text_preprocessing.ipynb file, it imports en_disaster.csv? Could you please let me know where we could find it? thanks!

[manueltonneau]

Unfortunately, it seems that @sebsk did not share the data csv, or at least, I didn't find it. Looking at the paper, you'll get more information about the data sources:

[changtianluckyforever]

Unfortunately, it seems that @sebsk did not share the data csv, or at least, I didn't find it. Looking at the paper, you'll get more information about the data sources:

Thanks!!!!

[rajae-Bens]

Hi,

did u find the data ? can u share it with us?

Thank u

[sebsk]

Hey guys, sorry for the super late reply. I did not anticipate people would ever notice this project (happy to know:)). The Tweets text data I used in this project are here: https://drive.google.com/drive/folders/1HwR_VKiAZsHPbifoxXrMMekfdEW59Mqk?usp=sharing

I have split them into df_train, df_val, and df_test, but it is not necessary to be split this way. While the aforementioned 3 are preprocessed data, "en_disaster" is the raw dataset.

Sorry again for not responding timely, and big thanks for the attention!

[baiziyuandyufei]

@sebsk Thank you!!!……^_^