Search results

Training experiments

Status: Open.

#30 In project-anuvaad/OpenNMT-py;

· aj7tesh opened on Sep 30, 2019

Compare stats for when subwords are giving poor results

Status: Open.

#29 In project-anuvaad/OpenNMT-py;

· aj7tesh opened on Sep 24, 2019

Build one word based model(eng-hindi) to cross check subword based model

Status: Open.

#28 In project-anuvaad/OpenNMT-py;

· aj7tesh opened on Sep 24, 2019

samba integration for models

Status: Open.

#27 In project-anuvaad/OpenNMT-py;

· aj7tesh opened on Sep 24, 2019

Kafka Integration

Status: Open.

#24 In project-anuvaad/OpenNMT-py;

· aj7tesh opened on Sep 23, 2019

english-hindi dictionary for handling 1 word

Status: Open.

#22 In project-anuvaad/OpenNMT-py;

· aj7tesh opened on Sep 18, 2019

Weight fixing for retraining?

Status: Open.

#20 In project-anuvaad/OpenNMT-py;

· aj7tesh opened on Sep 11, 2019

Data Augmentation to handle Named entity

Status: Open.

#17 In project-anuvaad/OpenNMT-py;

· aj7tesh opened on Sep 6, 2019

Incorporate subword-level external embeddings before training NMT

Status: Open.

#15 In project-anuvaad/OpenNMT-py;

· aj7tesh opened on Aug 30, 2019

Experiment-train model with unigram based sentencepiece model as opposed to generic BPE.

Status: Open.

#14 In project-anuvaad/OpenNMT-py;

· aj7tesh opened on Aug 30, 2019

Benchmark UI needs to be working again

Status: Open.

#12 In project-anuvaad/OpenNMT-py;

· aj7tesh opened on Aug 29, 2019

Handling pattern entity like Email address, phone numbers, url, pincode etc. for both soirce as well as target language.

Status: Open.

#11 In project-anuvaad/OpenNMT-py;

· aj7tesh opened on Aug 27, 2019