We should add this to SymPy Gamma once you have this working. Aaron Meurer
On Wed, May 13, 2020 at 1:03 PM Moses Paul <[email protected]> wrote: > > (ps I'm aware that the examples (sum, Max) I gave up there use iterables ) > Here's an excerpt from the model training dataset > what be the maximum of D, m => Max ( D , m ) > what be the max of D, m => Max ( D , m ) > what be the biggest of D, m => Max ( D , m ) > find the sum of D, m => sum ( D , m ) > find the total of D, m => sum ( D , m ) > find the minimum of D, m => Min ( D , m ) > find the min of D, m => Min ( D , m ) > find the smallest of D, m => Min ( D , m ) > find the maximum of D, m => Max ( D , m ) > find the max of D, m => Max ( D , m ) > > The above dataset is from a lemmatized version of natural language queries > (is -> be) > > and this is how the output looks like when passed to sympify > > >>> sympify('Max(1, 2, 3)') > 3 > >>> sympify('Max(1, 2, x)') > Max(2, x) > > > > On Thursday, May 14, 2020 at 12:23:45 AM UTC+5:30, Moses Paul wrote: >> >> >> >> On Wednesday, May 13, 2020 at 11:47:22 PM UTC+5:30, Aaron Meurer wrote: >>> >>> What sorts of things is it able to parse? >> >> >> As of now, it can do stuff like >> >> "What is the maximum of x,3,4,5,y" which returns Max(3,4,5,x,y) (passed to >> sympify) >> "Find the sum of x, x+y, x^3" -> sum(x, x+y, x**3) >> "Calculate the integral of x^2 + 3x" -> Integral(x**2+3*x) >> "Sum of x from 0 to 100" -> Sum(x, (x, 0, 100)) >> >> Kinda like that. >> ( I'm structuring work I've done so far, I'll post a link to my repo here, >> once I finish that ) >>> >>> >>> I don't know if there is a well structured glossary of SymPy >>> functions. The default namespace (what gets imported with "from sympy >>> import *") is the best place to start. >>> >> Gotcha! >> >>> Aaron Meurer >>> >>> On Wed, May 13, 2020 at 11:19 AM Moses Paul <[email protected]> wrote: >>> > >>> > So I've been working on an NLP parser for sympy. >>> > This is how it works, >>> > >>> > The Input is first "cleaned up" and rewritten into a structure that is >>> > comprehended by a NMT model (seq2seq) >>> > The processed input is passed on to the model which then gives a specific >>> > type of output, which is then "processed". >>> > The final result is one that works when used inside >>> > sympify('Expression') >>> > >>> > So Far I've been able to train using data generated from Functions >>> > similar to Sum, Max, Min i.e functions with a list of inputs and also >>> > with functions such as Summations and Integrals. >>> > Since I haven't gone through SymPy's entire codebase, it would be really >>> > useful if I had sort of a Glossary or an equivalent structure from which >>> > I can glean information about the various functions SymPy has, like a >>> > list of single parameter functions, two parameters, multiple parameters >>> > and so on. >>> > >>> > I haven't been able to find anything so far, help would be much >>> > appreciated >>> > >>> > Cheers >>> > Moses Paul >>> > >>> > -- >>> > You received this message because you are subscribed to the Google Groups >>> > "sympy" group. >>> > To unsubscribe from this group and stop receiving emails from it, send an >>> > email to [email protected]. >>> > To view this discussion on the web visit >>> > https://groups.google.com/d/msgid/sympy/9fef4da7-aa4f-4c47-ac44-932efacb1dcd%40googlegroups.com. > > -- > You received this message because you are subscribed to the Google Groups > "sympy" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/sympy/e88b5533-19d6-4b0f-9ce7-222735b1adfc%40googlegroups.com. -- You received this message because you are subscribed to the Google Groups "sympy" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/CAKgW%3D6%2B1JcViPiiZrk%2BYha%3Dzi3fNdJtYs7%2BhQAG8Zv%2Bt2G%2B-Bg%40mail.gmail.com.
