vincentk1991 / bert_summarization_1 Goto Github PK

I tried to guess what did you process to get these 2 columns but without the full csv, it is hard. Would you mind explain about it, so i could know how to create the input for preprocessing.py

Hello,

I have this issue in training step
RuntimeError: Expected tensor for argument #1 'indices' to have scalar type Long; but got torch.FloatTensor instead (while checking arguments for embedding)

this issue raised at File "train_GPT2.py", line 105.
outputs = model(input_ids = input_ids, mc_token_ids = mc_token_ids, mc_labels = mc_labels, lm_labels = lm_labels, token_type_ids = token_type_ids)

can you check it, please!

Whenever I run the GPT2_preprocessing.py I get this error:

Traceback (most recent call last):
File "GPT2_preprocessing.py", line 140, in
main(parser.parse_args())
File "GPT2_preprocessing.py", line 113, in main
word_tuple = load_words(df, index)
File "GPT2_preprocessing.py", line 46, in load_words
list_len = np.sort(np.random.choice(
File "mtrand.pyx", line 908, in numpy.random.mtrand.RandomState.choice
ValueError: 'a' cannot be empty unless no samples are taken

Any idea why I get this? Help is appreciated

I downloaded both pre-trained GPT2 and pre-trained Bert Based Uncased, but I am on the receiving end of countless errors. Is there a way to make it more user-friendly? I was thinking you could add "!wget", so users can ensure they are downloaded the correct file. Thank you very much for the help, and I hope to be able to use your code.

"Can't load 'bert-base-uncased'. Make sure that:

• 'bert-base-uncased' is a correct model identifier listed on 'https://huggingface.co/models'

• or 'bert-base-uncased' is the correct path to a directory containing a 'config.json' file"

Hi,
With reference to line:116 in train_GPT2, I am a bit confused that how are you passing "text + summary" to the model?
And you are using lm_loss and mc_loss together? Why it is so?
Why batch size is 1?
This code will be easy to understand with proper documentation.