Hugging Face's logo

Spaces:
PlayfulTechnology
/
QARAC
Build error

App Files Community
2
QARAC / scripts.py

Commit History

  1. Created HuggingFace Space

    PeteBleackley
    cce945c

  1. Removed diagnostic

    PeteBleackley
    3340c72

  2. Ensure consitency value is 32 bit

    PeteBleackley
    452e9a6

  1. Removed diagnostics

    PeteBleackley
    355ad7d

  2. Wrap CrossEntropyLoss in callable that makes it appplicable to sequences

    PeteBleackley
    4d8fdac

  3. Using torch.nn.CosineSimilarity to simplify code

    PeteBleackley
    798488e

  4. Use builtin sum rather than torch.sum for a generator

    PeteBleackley
    1d1a876

  5. Use keepdim option when normalising vectors

    PeteBleackley
    738b546

  6. Make EPSILON a tensor

    PeteBleackley
    cf5f935

  7. torch.maximum, not torch.max

    PeteBleackley
    98ad67d

  8. Learning rate scheduler

    PeteBleackley
    ae82fc7

  9. Fixed typo

    PeteBleackley
    1d631bc

  10. Minor error in setting up tokenizer

    PeteBleackley
    fb4c0b0

  11. PyTorch implementation of HugggingFace PreTrainedModel class does not allow direct setting of base_model. Rejig constructors accordingly

    PeteBleackley
    519dfd1

  1. Modified training scripts to use PyTorch

    PeteBleackley
    c8625dc

  1. Removed diagnostics

    PeteBleackley
    8561f47

  2. Use dictionary coprehensions to do padding and return dictionaries instead of default dictionaries

    PeteBleackley
    7a61dc8

  1. Completed script for testing consistency

    PeteBleackley
    dd9c3ed

  1. Fixed name of argument

    PeteBleackley
    1b76f7d

  2. Tensor is logits

    PeteBleackley
    888010e

  3. Incoplete testing script for consistency. Fixed typo

    PeteBleackley
    14f1a57

  4. Testing script for reasoning

    PeteBleackley
    6d6bb62

  5. Testing scripts

    PeteBleackley
    65ae142

  1. Attention masks, generation, and testing script

    PeteBleackley
    6ebe943

  1. Invoke class, not module

    PeteBleackley
    9946f4d

  2. Superclass __init__

    PeteBleackley
    8626c30

  3. Fix typo

    PeteBleackley
    e219af7

  4. Authentication

    PeteBleackley
    75a64b9

  5. Training script

    PeteBleackley
    33285bd

  6. Training script

    PeteBleackley
    63b2c6a

  1. Bug fixes

    PeteBleackley
    4f504fa

  2. Bug fixes

    PeteBleackley
    47b39f3

  3. Bug fixes

    PeteBleackley
    ab0523b

  4. Script for preparing training datasets

    PeteBleackley
    d63afc6

  1. Coreference Resolution for WikiQA dataset

    PeteBleackley
    e149b0f

  1. Data preparation for WikiQA

    PeteBleackley
    c106121

  2. More work on models

    PeteBleackley
    f16a715

  1. GlobalAttentionPoolingHead layer

    PeteBleackley
    8f1745b

  1. Datasets documentation

    PeteBleackley
    58e8b0b

  1. Code for building and training base modkels.

    PeteBleackley
    83d5adb