Public Synth MIPROv2 optimizer package.
Project description
synth-optimizers
Public Synth optimizer package source.
PyPI distribution: synth-optimizers
Current package version: 0.1.1
The initial public package contains the Synth Lab-derived MIPROv2 optimizer
surface under synth_optimizers.miprov2.
Install
pip install synth-optimizers
MIPROv2 Usage
from synth_optimizers.miprov2 import MiproCompatRunConfig, optimize
result = optimize(
seed_candidate={"system_prompt": "You are a careful assistant."},
trainset=train_rows,
valset=heldout_rows,
adapter=my_adapter,
task_lm="openai/gpt-4.1-mini",
reflection_lm="openai/gpt-5",
config=MiproCompatRunConfig(
dataset="banking77",
task="first5_intents",
optimizer_budget=8,
max_concurrency=4,
),
)
print(result.best_candidate)
print(result.val_aggregate_scores[result.best_idx])
Layout
src/
synth_optimizers/
miprov2/
Release
This package releases independently from other packages in this monorepo. See
RELEASE.md for the manual release workflow.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file synth_optimizers-0.1.1.tar.gz.
File metadata
- Download URL: synth_optimizers-0.1.1.tar.gz
- Upload date:
- Size: 135.1 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.14.2
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
0b0476f49f8bbe411d79453fa69ad2464f37f16471acbbe1103de854770d1aaa
|
|
| MD5 |
7b28d5e7109ea2f3d3ca9d01f161a98c
|
|
| BLAKE2b-256 |
9f524cad1330e9cad1771217dc7e2e79dafadcaea10bff5b0214dcfb1f8d34d2
|
File details
Details for the file synth_optimizers-0.1.1-py3-none-any.whl.
File metadata
- Download URL: synth_optimizers-0.1.1-py3-none-any.whl
- Upload date:
- Size: 149.4 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.14.2
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
6e1f1cfd22e137831fd9649223752c4004c85610dddd0c8204bf998ed6f0ac9a
|
|
| MD5 |
704fd886c3681c3439e92ddff1541470
|
|
| BLAKE2b-256 |
04bbb2d10c275acf298b59fee92470f1c9e8907175c3f9fbbece5b32ee648e3e
|