delft
delft copied to clipboard
kermitt2
a Deep Learning Framework for Text
Do you have a plan to reproduce the BERT NER model? I tried, but with Bert_base, the best micro-avg Test F1 on CoNLL-2003 is 91.37, while the reported in the...
This PR is to try to implement a way to load TF checkpoints and torch models in a "plain" way, so having just the `ckpt files` / `pytorch_model.bin` and the...
This PR should solve #124 and #126: 1. In the wrappers, the `tmp_directory` (usually `tmp/model-architecture`) is deleted at the beginning 3. The wrappers get an additional parameter for the temporary...
I've implemented this some time ago and it worked relatively well for normal models, however for transformers we realised it was not taken in account. With the new update to...
When we perform the n-fold cross validation or holdout evaluation, we would like to have the possibility to output the raw results (on a separate file) as we do in...
This PR setup DeLFT to use tensorboard: logs, callbacks... The idea is to cover the interesting metrics so that everything is visualized on Tensorboard. How to use: basically nothing special,...
I'm opening a new issue as I've stumble upon the same issue and went back the same pain again #14. For certain cases (e.g. development on windows or mac), we...
I'm running under Windows 10, following along the instructions given by the readme document. When trying to retrain the model using this command `python nerTagger.py --dataset-type conll2003 train_eval` I ran...
This PR aims to make updates to the applications scripts. I will add the other changes while working in this draft Changes: - validate `--architecture` using the argparse - for...
This issue is to cover a revision of the directory management on the training or n-fold training. Right now the training write directly under `/data/models/model-name`. For training is not a...