o1 forces chain of thought so that’s unnecessary now. Something like dspy is likely still be useful for out of distribution tasks or questions though, via examples
Great tutorial! There's something missing tho'. How to define correctly the examples for any DSPy methods that accept a trainset. You need to define a list of dspy.Example(). Putting it out here maybe it well help someone.
Best tutorial on DSPy I've watched so far.
thanks, appreciate it
+1 for video with agents
Great Tutorial, DSPy stands for Declarative Self-Improving Python FYI
Yes! That's so cool! Thank you so much for talking about this!
Ive been busy lately. Thank you as always ! 🎉
Can’t wait
Thank you for sharing.
Thanks for your share :)
Now that o1 is out, how would it affect the value that dspy offers?
o1 forces chain of thought so that’s unnecessary now.
Something like dspy is likely still be useful for out of distribution tasks or questions though, via examples
Great tutorial! There's something missing tho'. How to define correctly the examples for any DSPy methods that accept a trainset. You need to define a list of dspy.Example(). Putting it out here maybe it well help someone.
Yeah true, setting up examples is key. Perhaps in a bit I'll do another deeper DSPy vid
great tip