Karthikeyan Madathil

Yes, we need a better package than argparse. click looks like a good option. I've committed a workaround for now, but this is another one which needs a good solution.

Will do. I don't think I've aggressively optimized many parts of the parser, so I'll take those up.

I concur

@codito - Not sure how the whitespace problem and this issue are related? This is about evaluating accuracy, is it not. Your issue is picking one split over another.

यो घटको यस्य उत्पन्नम् उपयुङ्क्ते इति क्रमेण सोपनक्रमः कृतः‌ |‌ अत्र, L0 घटकस्य उत्पन्नं L1 इत्यनेन प्रयुज्यते | > भवता सूचितयोर् अपवादयोर् अन्तर्भावः प्रथमस्तरे (नाम ल-० इत्यत्र) एवास्तु नावगतं मया...

पश्यतु इमानि सूत्राणि ८ . ३ . ४० नमस्पुरसो*र्गत्योः* ८ . ३ . ४१ इदुदुपधस्य*चाप्रत्ययस्य* । ८ . ३ . ४२ तिरसोऽन्यतरस्याम् । ८ . ३ . ४३ द्विस्त्रिश्चतुरिति *कृत्वोऽर्थे*...

> अधुना तु majority विषयान् समीचीनतया कर्तुं प्रयत्नः विधेयः (यतो हि तेन बहूनाम् उपकारो भवेत्) । तदनु exceptions विषये चिन्तयाम । समीचीनम् |‌ इदानीं अपवादानां चिन्ता न करोमि :-)

Awsome! Dr. Dhaval! Once you create a package, we can use this to show detailed derivations of verb forms.Is there something planned for noun forms? @vvasuki @avinashvarna @codito : Thoughts...

An even better option may be the smaller 1300 sentence testset found in this [even later paper.](https://www.aclweb.org/anthology/2020.emnlp-main.388.pdf) . Advantage, it is available on github The paper author has provided another...

I have received the DCS10K and KISS datasets from Amrith Krishna. KISS has been committed into the DB. DCS10K will be added after I figure out how to (too many...