Cahya Wirawan

Multiclass classification and pre-trained word embedding (word2vec & GloVe) support and it's comparison

36

Hi Denny, Thanks very much for the great blog and the source code. I just started with machine learning and this article helped me a lot to learn machine learning,...

Support for multiclass, word embeddings, configuration file and new datasets

4

Hi, I added following functionalities: - multiclass classification - pre-trained word embedding using word2vec and GloVe - configuration file in yaml format - new dataset 20newsgroup (loaded using sklearn.datasets) -...

sync doesn't have option "-f"

2

Hi, When I tried livecd-iso-to-disk.sh, it exec "sync -f", but in red hat/centos the sync command doesn't have the option -f

Kernel restarted

1

I have some other problems to run the notebook CLS-DE.ipynb. If I use conda and install the default pytorch (1.3.1), after the command `exp.finetune_lm.train_(cls_dataset, num_epochs=20)` I get following error message:...

Racing name: DodolGarut This interactive Visualization provides more helpful informations if we hover on the specific line or point of the picture. It can display the waypoint index, steps, progress,...

Hi, I would like to do speed comparison among several Forced Aligner softwares. I already tried MAUS with around 30 minutes of an audio file. The alignment process took more...

The dataset kopi_cc with the config name kopi_cc_2022_05-neardup_clean_nusantara_ssp is not complete

2

## Describe the bug It seems that the dataset **kopi_cc** with the config name **kopi_cc_2022_05-neardup_clean_nusantara_ssp** is not complete. It can't find the file cleaned_oscar-neardup-0000000000**19**.json.gz and above. According to the source...

script inference for a single image

6

Hi, Thanks for the source code, the result looks great. Btw, do you have a simple script to do the inference for a single image (and mask)? Thanks.

The fine tuning script run_speech_recognition_seq2seq_streaming.py use interleave_datasets which will truncate the train split

2

The fine tuning script run_speech_recognition_seq2seq_streaming.py use interleave_dataset function to add the train and validation split together. But I think what we really want to use is concatenate_dataset, because according to...

Sometimes my model repeats a whole paragraph again and again endlessly. Would it be possible to ban repetition of specific ngram size, like this NoRepeatNgramLogitsProcessor from transformer https://huggingface.co/docs/transformers/v4.36.1/en/internal/generation_utils#transformers.NoRepeatNGramLogitsProcessor ? Thanks.