K.N.RAMESH <knra...@gmail.com>: Mar 12 02:25PM +0530
Courtesy: Rohan Pandey
@khoomeik
Sanskrit has >2000 verb roots (dhātus). But do you really need to learn
them all?
I had Claude analyze 270 Sanskrit texts, and it found that with just the
192 most common dhātus, you can understand ~90% of verbs in literature.
Attached are those 192 dhātus, ordered by frequency:
|
|
|
Hrishikesh Terdalkar <terdalk...@gmail.com>: Mar 12 09:22PM +0530
If this was performed on DCS, why do you need AI, in particular LLMs for
that analysis?
Isn't this a simple frequency counting problem?
Not saying the results themselves are invalid or contesting their utility.
Just wondering what role Claude played, and more importantly, why?
Regards,
Hrishikesh
|
|
|
Srikanth <sri...@gmail.com>: Mar 12 10:14PM +0530
That's really helpful. Thanks for sharing. The image shared is of
compressed quality... could you please share a text/pdf format or a
higher quality (png, larger) image so that I can take a print out.
Regards,
Srikanth
On Thu, Mar 12 2026 at 02:25:39 PM +05:30:00, K.N.RAMESH
|
|
|
kenp <drk...@gmail.com>: Mar 12 07:05PM -0700
https://ashtadhyayi.com/dhatu
|
|
|
vishal jaiswal <tovisha...@gmail.com>: Mar 13 07:46AM +0530
The process used as well as the names of the texta would be good to have.
On Thu, Mar 12, 2026, 21:26 Hrishikesh Terdalkar <terdalk...@gmail.com>
wrote:
|
|
|
Sudhir Pattar <sudhir...@gmail.com>: Mar 12 10:31PM -0400
What does the color code mean?
Also, please share a picture of higher resolution.
Thanks,
Sudhir
|
|
|
Raunak Dhar <mynameisr...@gmail.com>: Mar 13 08:18AM +0530
Yes, please upload a higher quality image if you have it.
|