Synthetic Data Generation v2

[krypticmouse]

Usage:
[1] Features:

import dsp
from dspy.datasets import DataLoader
from dspy.experimental import Synthesizer, SynthesizerArguments

class GenerateAnswer(dspy.Signature):
    """Answer questions with short factoid answers."""

    context = dspy.InputField(desc="may contain relevant facts")
    question = dspy.InputField()
    answer = dspy.OutputField(desc="often between 1 and 5 words")

config = SynthesizerArguments()
synthesizer = Synthesizer(config=config)

syn_data = synthesizer.generate(GenerateAnswer, num_data=10)

synthesizer.export(data=syn_data, path="syn_datacsv")

[2] Batched Generation(Faster):

from dspy.datasets import DataLoader
from dspy.experimental import Synthesizer, SynthesizerArguments

dl = DataLoader()

data = dl.from_huggingface(
    "gsm8k", "main",
    fields=("question", "answer"),
    input_keys=("question",),
    split="train[:8]"
)

config = SynthesizerArguments()
synthesizer = Synthesizer(config=config)

synthesizer.generate(
    ground_source=data,
    num_data=10,
    batch_size=2,
)

[3] Feedback Driven Generation

config = SynthesizerArguments(
    feedback_mode="llm", # or "human" for human in a loop generation
    num_example_for_feedback=3,
)
synthesizer = Synthesizer(config=config)

synthesizer.generate(
    ground_source=data,
    num_data=10,
    batch_size=2,
)

[4] Tweakable LM for input and output gen and support for module based output generation

[5] Example based input generation optimization

[krypticmouse]

Thanks a lot for the suggestions and fixes!!

[chiragshah285]

@krypticmouse would it be possible to add this to the vercel documnetation? Also what does refer to?
[4] Tweakable LM for input and output gen and support for module based output generation

[5] Example based input generation optimization

[beltrewilton]

Fix potential import bug in dspy.experimental module

The code may contain a bug in the import statement within /dspy/experimental/init.py. The original import statement is: from module_graph import *

This has been changed to: from .module_graph import *