Skip to content

"TF record not found" #34

Description

@Rafiur

Upon running the "Train your own medical Q&A retrieval model" colab notebook I found that even though the pandas dataframe can read the file in "data/sampleData.csv" but running the
"d = create_dataset_for_bert(
'data/sampleData.csv', tokenizer=tokenizer, batch_size=batch_size,
shuffle_buffer=500000, dynamic_padding=True, max_seq_length=max_seq_len)"
gives an output of "TF record not found"
Also if "d" is printed we can see that all the attributes like question id or question mask etc have none in their value and the end result of running the last cell on the notebook results in a "value error: empty training data"
How can I fix this?

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions