Skip to content

armgit / dspy

Public Imported from GitHub
0

DSPy: The framework for programming—not prompting—language models

Python

ci(dspy): Add Ruff linting workflow (#538)

Isaac Millera637bf47b527
diff --git a/.github/workflows/run_tests.yml b/.github/workflows/run_tests.yml
index 4a31b8645..52ac21e3c 100644
--- a/.github/workflows/run_tests.yml
+++ b/.github/workflows/run_tests.yml
@@ -1,4 +1,4 @@
-name: Build and Run Tests
+name: Fix, Test, and Build
on:
push:
@@ -10,24 +10,54 @@ env:
POETRY_VERSION: "1.6.1"
jobs:
+ fix:
+ name: Apply Ruff Fix
+ runs-on: ubuntu-latest
+ permissions:
+ contents: write
+ steps:
+ - uses: actions/checkout@v3
+ - uses: actions/setup-python@v5
+ - uses: chartboost/ruff-action@v1
+ with:
+ args: --fix-only
+ - uses: stefanzweifel/git-auto-commit-action@v5
+ with:
+ commit_message: "Automatic Style fixes"
+
test:
- name: Test
+ name: Run Tests
runs-on: ubuntu-latest
strategy:
matrix:
python-version: ["3.9"]
steps:
- - uses: actions/checkout@v3
+ - uses: actions/checkout@v4
+ with:
+ ref: ${{ github.head_ref }}
+ - name: Load cached Poetry installation
+ id: cached-poetry
+ uses: actions/cache@v3
+ with:
+ path: ~/.local
+ key: poetry-${{ env.POETRY_VERSION }}-${{ hashFiles('**/poetry.lock') }}
+ - name: Install Poetry
+ if: steps.cached-poetry.outputs.cache-hit != 'true'
+ uses: snok/install-poetry@v1
- name: Set up python ${{ matrix.python-version }}
uses: actions/setup-python@v5
with:
python-version: ${{ matrix.python-version }}
- - name: Install Poetry
- uses: snok/install-poetry@v1
+ cache: "poetry"
- name: Install dependencies
- run: poetry install
+ run: poetry install --no-interaction --no-root
+ - name: Run lint with tests
+ uses: chartboost/ruff-action@v1
+ with:
+ args: --fix-only
- name: Run tests with pytest
run: poetry run pytest tests/
+
build_poetry:
name: Build Poetry
runs-on: ubuntu-latest
@@ -35,13 +65,23 @@ jobs:
matrix:
python-version: ["3.9"]
steps:
- - uses: actions/checkout@v3
+ - uses: actions/checkout@v4
+ with:
+ ref: ${{ github.head_ref }}
+ - name: Load cached Poetry installation
+ id: cached-poetry
+ uses: actions/cache@v3
+ with:
+ path: ~/.local
+ key: poetry-${{ env.POETRY_VERSION }}-${{ hashFiles('**/poetry.lock') }}
+ - name: Install Poetry
+ if: steps.cached-poetry.outputs.cache-hit != 'true'
+ uses: snok/install-poetry@v1
- name: Set up python ${{ matrix.python-version }}
uses: actions/setup-python@v5
with:
python-version: ${{ matrix.python-version }}
- - name: Install Poetry
- uses: snok/install-poetry@v1
+ cache: "poetry"
- name: Build
run: poetry build
- name: Install built package
@@ -50,6 +90,7 @@ jobs:
run: python -c "import dspy"
- name: Test import dsp
run: python -c "import dsp"
+
build_setup:
name: Build Setup
runs-on: ubuntu-latest
@@ -57,10 +98,22 @@ jobs:
matrix:
python-version: ["3.9"]
steps:
- - uses: actions/checkout@v3
+ - uses: actions/checkout@v4
+ with:
+ ref: ${{ github.head_ref }}
+ - name: Load cached Poetry installation
+ id: cached-poetry
+ uses: actions/cache@v3
+ with:
+ path: ~/.local
+ key: poetry-${{ env.POETRY_VERSION }}-${{ hashFiles('**/poetry.lock') }}
+ - name: Install Poetry
+ if: steps.cached-poetry.outputs.cache-hit != 'true'
+ uses: snok/install-poetry@v1
- name: Set up python ${{ matrix.python-version }}
uses: actions/setup-python@v5
with:
python-version: ${{ matrix.python-version }}
+ cache: "poetry"
- name: Run setup.py build
run: python setup.py build
diff --git a/dsp/__init__.py b/dsp/__init__.py
index fc8a74241..4f98315f1 100644
--- a/dsp/__init__.py
+++ b/dsp/__init__.py
@@ -3,7 +3,6 @@ from .primitives import *
from .templates import *
from .utils import settings
-
"""
TODO:
diff --git a/dsp/evaluation/utils.py b/dsp/evaluation/utils.py
index 1bfd740e4..7b6a47780 100644
--- a/dsp/evaluation/utils.py
+++ b/dsp/evaluation/utils.py
@@ -1,7 +1,8 @@
-import dsp
-import tqdm
import pandas as pd
+import tqdm
+
+import dsp
try:
from IPython.display import display as ipython_display
diff --git a/dsp/modules/__init__.py b/dsp/modules/__init__.py
index d824d3fa5..241a349c5 100644
--- a/dsp/modules/__init__.py
+++ b/dsp/modules/__init__.py
@@ -1,19 +1,15 @@
-from .cache_utils import *
from .azure_openai import AzureOpenAI
-from .gpt3 import *
+from .bedrock import *
+from .cache_utils import *
+from .clarifai import *
+from .cohere import *
+from .colbertv2 import ColBERTv2
from .databricks import *
+from .google import *
+from .gpt3 import *
from .hf import HFModel
-from .colbertv2 import ColBERTv2
-from .sentence_vectorizer import *
-from .cohere import *
-from .sbert import *
-from .pyserini import *
+from .hf_client import Anyscale, HFClientTGI, Together
from .ollama import *
-from .clarifai import *
-from .bedrock import *
-from .google import *
-
-
-from .hf_client import HFClientTGI
-from .hf_client import Anyscale
-from .hf_client import Together
+from .pyserini import *
+from .sbert import *
+from .sentence_vectorizer import *
diff --git a/dsp/modules/aws_lm.py b/dsp/modules/aws_lm.py
index 83674db10..a44e30a49 100644
--- a/dsp/modules/aws_lm.py
+++ b/dsp/modules/aws_lm.py
@@ -4,10 +4,11 @@ A generalized AWS LLM.
from __future__ import annotations
-from abc import abstractmethod
+import json
import logging
+from abc import abstractmethod
from typing import Any, Literal
-import json
+
from dsp.modules.lm import LM
# Heuristic translating number of chars to tokens
diff --git a/dsp/modules/azurecognitivesearch.py b/dsp/modules/azurecognitivesearch.py
index e4ff2e996..fedc59b01 100644
--- a/dsp/modules/azurecognitivesearch.py
+++ b/dsp/modules/azurecognitivesearch.py
@@ -1,6 +1,7 @@
-from typing import Union, Any
+from typing import Any, Union
from dsp.utils import dotdict
+
try:
from azure.core.credentials import AzureKeyCredential
from azure.search.documents import SearchClient
diff --git a/dsp/modules/bedrock.py b/dsp/modules/bedrock.py
index c52ae6197..ac8b42d2d 100644
--- a/dsp/modules/bedrock.py
+++ b/dsp/modules/bedrock.py
@@ -1,7 +1,8 @@
from __future__ import annotations
-from typing import Any
import json
+from typing import Any
+
from dsp.modules.aws_lm import AWSLM
diff --git a/dsp/modules/cache_utils.py b/dsp/modules/cache_utils.py
index 78270c879..e6c09fe71 100644
--- a/dsp/modules/cache_utils.py
+++ b/dsp/modules/cache_utils.py
@@ -1,12 +1,11 @@
import os
-
+from functools import wraps
from pathlib import Path
+
from joblib import Memory
-from functools import wraps
from dsp.utils import dotdict
-
cache_turn_on = True
diff --git a/dsp/modules/clarifai.py b/dsp/modules/clarifai.py
index 8a691a4be..2d5839c62 100644
--- a/dsp/modules/clarifai.py
+++ b/dsp/modules/clarifai.py
@@ -3,6 +3,7 @@ from typing import Any, Optional
from dsp.modules.lm import LM
+
class ClarifaiLLM(LM):
"""Integration to call models hosted in clarifai platform.
diff --git a/dsp/modules/cohere.py b/dsp/modules/cohere.py
index 405eaea48..7308d5935 100644
--- a/dsp/modules/cohere.py
+++ b/dsp/modules/cohere.py
@@ -1,5 +1,6 @@
import math
from typing import Any, Optional
+
import backoff
from dsp.modules.lm import LM
diff --git a/dsp/modules/colbertv2.py b/dsp/modules/colbertv2.py
index 1b33d077a..8ff3c1622 100644
--- a/dsp/modules/colbertv2.py
+++ b/dsp/modules/colbertv2.py
@@ -1,11 +1,11 @@
import functools
-from typing import Optional, Union, Any
+from typing import Any, Optional, Union
+
import requests
from dsp.modules.cache_utils import CacheMemory, NotebookCacheMemory
from dsp.utils import dotdict
-
# TODO: Ideally, this takes the name of the index and looks up its port.
diff --git a/dsp/modules/databricks.py b/dsp/modules/databricks.py
index 8f7d7d6e8..7d90e4fc2 100644
--- a/dsp/modules/databricks.py
+++ b/dsp/modules/databricks.py
@@ -19,8 +19,8 @@ from dsp.modules.cache_utils import CacheMemory, NotebookCacheMemory, cache_turn
from dsp.modules.gpt3 import GPT3
try:
- from openai.openai_object import OpenAIObject
import openai.error
+ from openai.openai_object import OpenAIObject
ERRORS = (openai.error.RateLimitError, openai.error.ServiceUnavailableError, openai.error.APIError)
except Exception:
ERRORS = (openai.RateLimitError, openai.APIError)
diff --git a/dsp/modules/finetuning/finetune_hf.py b/dsp/modules/finetuning/finetune_hf.py
index 0ade213ba..a22ab10fc 100644
--- a/dsp/modules/finetuning/finetune_hf.py
+++ b/dsp/modules/finetuning/finetune_hf.py
@@ -1,28 +1,30 @@
# Adapted from: https://www.philschmid.de/fine-tune-flan-t5#3-fine-tune-and-evaluate-flan-t5
-import os
-import json
import copy
import glob
-import torch
+import json
+import os
import warnings
+from dataclasses import dataclass
+
import evaluate
import numpy as np
+import torch
from datasets import Dataset
-from dataclasses import dataclass
from transformers import (
- set_seed,
AutoConfig,
- AutoModelForSeq2SeqLM,
AutoModelForCausalLM,
+ AutoModelForSeq2SeqLM,
AutoTokenizer,
+ DataCollatorForSeq2Seq,
PreTrainedTokenizer,
- Trainer,
Seq2SeqTrainer,
- TrainingArguments,
Seq2SeqTrainingArguments,
- DataCollatorForSeq2Seq,
+ Trainer,
+ TrainingArguments,
+ set_seed,
)
+
# from peft import get_peft_model, LoraConfig, TaskType
from transformers.trainer_callback import TrainerCallback
diff --git a/dsp/modules/google.py b/dsp/modules/google.py
index 9a1c3937c..5d97534ea 100644
--- a/dsp/modules/google.py
+++ b/dsp/modules/google.py
@@ -1,6 +1,7 @@
import os
-from typing import Any, Optional
from collections.abc import Iterable
+from typing import Any, Optional
+
import backoff
from dsp.modules.lm import LM
diff --git a/dsp/modules/gpt3.py b/dsp/modules/gpt3.py
index ad440f234..d0b343af3 100644
--- a/dsp/modules/gpt3.py
+++ b/dsp/modules/gpt3.py
@@ -11,10 +11,10 @@ import functools
import json
from typing import Any, Literal, Optional, cast
-import dsp
import backoff
import openai
+import dsp
from dsp.modules.cache_utils import CacheMemory, NotebookCacheMemory, cache_turn_on
from dsp.modules.lm import LM
@@ -24,8 +24,8 @@ except Exception:
OPENAI_LEGACY = True
try:
- from openai.openai_object import OpenAIObject
import openai.error
+ from openai.openai_object import OpenAIObject
ERRORS = (
openai.error.RateLimitError,
diff --git a/dsp/modules/hf.py b/dsp/modules/hf.py
index 3c9306a8f..aad0c0e36 100644
--- a/dsp/modules/hf.py
+++ b/dsp/modules/hf.py
@@ -1,8 +1,9 @@
# from peft import PeftConfig, PeftModel
# from transformers import AutoModelForSeq2SeqLM, AutoModelForCausalLM, AutoTokenizer, AutoConfig
-from typing import Optional, Literal
+from typing import Literal, Optional
from dsp.modules.lm import LM
+
# from dsp.modules.finetuning.finetune_hf import preprocess_prompt
def openai_to_hf(**kwargs):
@@ -43,8 +44,8 @@ class HFModel(LM):
self.device_map = hf_device_map
if not self.is_client:
try:
- from transformers import AutoModelForSeq2SeqLM, AutoModelForCausalLM, AutoTokenizer, AutoConfig
import torch
+ from transformers import AutoConfig, AutoModelForCausalLM, AutoModelForSeq2SeqLM, AutoTokenizer
except ImportError as exc:
raise ModuleNotFoundError(
"You need to install Hugging Face transformers library to use HF models.",
diff --git a/dsp/modules/hf_client.py b/dsp/modules/hf_client.py
index 4def62819..fb92c48d5 100644
--- a/dsp/modules/hf_client.py
+++ b/dsp/modules/hf_client.py
@@ -1,15 +1,15 @@
import os
import random
-import requests
-from dsp.modules.hf import HFModel, openai_to_hf
-from dsp.modules.cache_utils import CacheMemory, NotebookCacheMemory
-import subprocess
import re
import shutil
+import subprocess
# from dsp.modules.adapter import TurboAdapter, DavinciAdapter, LlamaAdapter
-
import backoff
+import requests
+
+from dsp.modules.cache_utils import CacheMemory, NotebookCacheMemory
+from dsp.modules.hf import HFModel, openai_to_hf
ERRORS = (Exception)
@@ -173,7 +173,7 @@ class HFServerTGI:
container_id = match.group(1)
port_mapping = subprocess.check_output(['docker', 'port', container_id]).decode().strip()
if f'0.0.0.0:{port}' in port_mapping:
- subprocess.run(['docker', 'stop', container_id])
+ subprocess.run(['docker', 'stop', container_id], check=False)
def run_server(self, port, model_name=None, model_path=None, env_variable=None, gpus="all", num_shard=1, max_input_length=4000, max_total_tokens=4096, max_best_of=100):
self.close_server(port)
@@ -355,8 +355,7 @@ class ChatModuleClient(HFModel):
def __init__(self, model, model_path):
super().__init__(model=model, is_client=True)
- from mlc_chat import ChatModule
- from mlc_chat import ChatConfig
+ from mlc_chat import ChatConfig, ChatModule
self.cm = ChatModule(
model=model, lib_path=model_path, chat_config=ChatConfig(conv_template="LM"),
diff --git a/dsp/modules/ollama.py b/dsp/modules/ollama.py
index a2e3e0139..27304d271 100644
--- a/dsp/modules/ollama.py
+++ b/dsp/modules/ollama.py
@@ -1,10 +1,10 @@
-from dsp.modules.lm import LM
-from typing import Any, Literal
-
import datetime
import hashlib
+from typing import Any, Literal
+
import requests
+from dsp.modules.lm import LM
def post_request_metadata(model_name, prompt):
diff --git a/dsp/modules/pyserini.py b/dsp/modules/pyserini.py
index 94ce970ea..5523a76e7 100644
--- a/dsp/modules/pyserini.py
+++ b/dsp/modules/pyserini.py
@@ -1,5 +1,6 @@
-from typing import Union
import json
+from typing import Union
+
from datasets import Dataset
from dsp.utils import dotdict
@@ -30,8 +31,8 @@ class PyseriniRetriever:
"""
# Keep pyserini as an optional dependency
+ from pyserini.prebuilt_index_info import FAISS_INDEX_INFO, IMPACT_INDEX_INFO, TF_INDEX_INFO
from pyserini.search import FaissSearcher
- from pyserini.prebuilt_index_info import TF_INDEX_INFO, FAISS_INDEX_INFO, IMPACT_INDEX_INFO
self.encoder = FaissSearcher._init_encoder_from_str(query_encoder)
self.dataset = dataset
diff --git a/dsp/primitives/__init__.py b/dsp/primitives/__init__.py
index 3798444a1..ef278efa6 100644
--- a/dsp/primitives/__init__.py
+++ b/dsp/primitives/__init__.py
@@ -1,6 +1,6 @@
-from .primitives import *
+from .compiler import *
from .demonstrate import *
-from .search import *
-from .predict import *
from .inspect import *
-from .compiler import *
+from .predict import *
+from .primitives import *
+from .search import *
diff --git a/dsp/primitives/compiler.py b/dsp/primitives/compiler.py
index 996919e85..625ae23ac 100644
--- a/dsp/primitives/compiler.py
+++ b/dsp/primitives/compiler.py
@@ -1,12 +1,13 @@
import os
+import random
+import subprocess
import time
+
import tqdm
import ujson
-import random
-import subprocess
+from datasets.fingerprint import Hasher
import dsp
-from datasets.fingerprint import Hasher
if os.environ.get('DSP_NOTEBOOK_CACHEDIR'):
training_data_directory = os.path.join(os.environ.get('DSP_NOTEBOOK_CACHEDIR'), 'compiler')
@@ -24,7 +25,7 @@ def openai_check_finetune(jobname):
command = f"""openai api fine_tunes.get -i {jobname}"""
print(command)
- result = subprocess.run(command.split(), stdout=subprocess.PIPE)
+ result = subprocess.run(command.split(), stdout=subprocess.PIPE, check=False)
output = result.stdout.decode("utf-8").strip()
try:
diff --git a/dsp/primitives/demonstrate.py b/dsp/primitives/demonstrate.py
index ae08e28ba..7dfe126b5 100644
--- a/dsp/primitives/demonstrate.py
+++ b/dsp/primitives/demonstrate.py
@@ -1,5 +1,5 @@
import random
-from typing import Callable, Any
+from typing import Any, Callable
import numpy as np
diff --git a/dsp/primitives/inspect.py b/dsp/primitives/inspect.py
index 3fbb05c55..ebcd3b787 100644
--- a/dsp/primitives/inspect.py
+++ b/dsp/primitives/inspect.py
@@ -1,8 +1,9 @@
import inspect
-import string
+import json
import random
+import string
+
import requests
-import json
class FuncInspector:
diff --git a/dsp/primitives/predict.py b/dsp/primitives/predict.py
index c89256b91..41edf6c51 100644
--- a/dsp/primitives/predict.py
+++ b/dsp/primitives/predict.py
@@ -1,11 +1,11 @@
from collections import Counter
-from typing import Callable, Any, Optional
+from typing import Any, Callable, Optional
import dsp
-from dsp.utils import zipstar, normalize_text
-from dsp.utils.utils import dotdict
-from dsp.templates.template_v3 import Template
from dsp.primitives.demonstrate import Example
+from dsp.templates.template_v3 import Template
+from dsp.utils import normalize_text, zipstar
+from dsp.utils.utils import dotdict
class Completions:
diff --git a/dsp/primitives/primitives.py b/dsp/primitives/primitives.py
index a7851ea33..2e4c4bfc0 100644
--- a/dsp/primitives/primitives.py
+++ b/dsp/primitives/primitives.py
@@ -1,6 +1,8 @@
-import dsp
from functools import wraps
+import dsp
+
+
# applied right to left (innermost first, like function calls)
def compose_decorators(*decorators):
def decorator(func):
diff --git a/dsp/templates/__init__.py b/dsp/templates/__init__.py
index b4aed7643..5eaae95f7 100644
--- a/dsp/templates/__init__.py
+++ b/dsp/templates/__init__.py
@@ -1,4 +1,4 @@
-from .utils import *
from .template_v2 import *
from .template_v3 import *
+from .utils import *
diff --git a/dsp/templates/template_v2.py b/dsp/templates/template_v2.py
index d6e61642e..236fd1d64 100644
--- a/dsp/templates/template_v2.py
+++ b/dsp/templates/template_v2.py
@@ -1,9 +1,11 @@
-from collections import namedtuple
import re
-from typing import Union, Any
+from collections import namedtuple
+from typing import Any, Union
+
import dsp
from dsp.primitives.demonstrate import Example
-from .utils import passages2text, format_answers
+
+from .utils import format_answers, passages2text
Field = namedtuple("Field", "name separator input_variable output_variable description")
diff --git a/dsp/templates/template_v3.py b/dsp/templates/template_v3.py
index fb097f877..605705c15 100644
--- a/dsp/templates/template_v3.py
+++ b/dsp/templates/template_v3.py
@@ -1,5 +1,6 @@
from typing import Callable
-from dsp.templates import TemplateV2, passages2text, format_answers, Field
+
+from dsp.templates import Field, TemplateV2, format_answers, passages2text
class Type:
diff --git a/dsp/templates/utils.py b/dsp/templates/utils.py
index 0b0d4891f..53624ad2d 100644
--- a/dsp/templates/utils.py
+++ b/dsp/templates/utils.py
@@ -1,4 +1,4 @@
-from typing import Union, Optional
+from typing import Optional, Union
def passages2text(passages: Union[str, list, tuple]) -> str:
diff --git a/dsp/utils/__init__.py b/dsp/utils/__init__.py
index c41f74832..63624350a 100644
--- a/dsp/utils/__init__.py
+++ b/dsp/utils/__init__.py
@@ -1,4 +1,4 @@
+from dsp.utils.dpr import *
+from dsp.utils.metrics import *
from dsp.utils.settings import *
from dsp.utils.utils import *
-from dsp.utils.metrics import *
-from dsp.utils.dpr import *
diff --git a/dsp/utils/dpr.py b/dsp/utils/dpr.py
index 2174ea4da..d4d18f84f 100644
--- a/dsp/utils/dpr.py
+++ b/dsp/utils/dpr.py
@@ -4,9 +4,10 @@
Original license: https://github.com/facebookresearch/DPR/blob/main/LICENSE
"""
-import regex
import unicodedata
+import regex
+
class Tokens:
"""A class to represent a list of tokenized text."""
diff --git a/dsp/utils/metrics.py b/dsp/utils/metrics.py
index f0d69f00f..ddbd51ae9 100644
--- a/dsp/utils/metrics.py
+++ b/dsp/utils/metrics.py
@@ -1,8 +1,8 @@
import re
import string
import unicodedata
-
from collections import Counter
+
from dsp.utils.utils import print_message
diff --git a/dsp/utils/settings.py b/dsp/utils/settings.py
index 462f99d42..1de0ada7e 100644
--- a/dsp/utils/settings.py
+++ b/dsp/utils/settings.py
@@ -1,6 +1,7 @@
+import threading
from contextlib import contextmanager
+
from dsp.utils.utils import dotdict
-import threading
class Settings:
diff --git a/dsp/utils/settings_v2.py b/dsp/utils/settings_v2.py
index 2998ca7cd..6652474d3 100644
--- a/dsp/utils/settings_v2.py
+++ b/dsp/utils/settings_v2.py
@@ -1,7 +1,8 @@
+import copy
import threading
-from contextlib import contextmanager
from concurrent.futures import ThreadPoolExecutor, as_completed
-import copy
+from contextlib import contextmanager
+
class Settings:
def __init__(self):
diff --git a/dsp/utils/utils.py b/dsp/utils/utils.py
index aeba5c970..d5a30b0e3 100644
--- a/dsp/utils/utils.py
+++ b/dsp/utils/utils.py
@@ -1,10 +1,10 @@
-import os
-import tqdm
import datetime
import itertools
-
+import os
from collections import defaultdict
+import tqdm
+
def print_message(*s, condition=True, pad=False, sep=None):
s = " ".join([str(x) for x in s])
@@ -77,6 +77,7 @@ def batch(group, bsize, provide_offset=False):
import copy
+
class dotdict(dict):
def __getattr__(self, key):
if key.startswith('__') and key.endswith('__'):
diff --git a/dspy/__init__.py b/dspy/__init__.py
index 98fe6fba8..b61ea8a01 100644
--- a/dspy/__init__.py
+++ b/dspy/__init__.py
@@ -1,18 +1,12 @@
-from dsp.modules.hf_client import ChatModuleClient
-from dsp.modules.hf_client import HFServerTGI, HFClientVLLM, HFClientSGLang
-from .signatures import *
-
-from .retrieve import *
-from .predict import *
-from .primitives import *
-
# from .evaluation import *
-
-
# FIXME:
-
-
import dsp
+from dsp.modules.hf_client import ChatModuleClient, HFClientSGLang, HFClientVLLM, HFServerTGI
+
+from .predict import *
+from .primitives import *
+from .retrieve import *
+from .signatures import *
settings = dsp.settings
diff --git a/dspy/datasets/__init__.py b/dspy/datasets/__init__.py
index 62409d6ed..bf6e3dd01 100644
--- a/dspy/datasets/__init__.py
+++ b/dspy/datasets/__init__.py
@@ -1,4 +1,4 @@
+from .colors import Colors
+from .dataloader import DataLoader
from .dataset import Dataset
from .hotpotqa import HotPotQA
-from .colors import Colors
-from .dataloader import DataLoader
\ No newline at end of file
diff --git a/dspy/datasets/colors.py b/dspy/datasets/colors.py
index 265badf27..edf2e4169 100644
--- a/dspy/datasets/colors.py
+++ b/dspy/datasets/colors.py
@@ -1,4 +1,5 @@
import random
+
from dspy.datasets.dataset import Dataset
### A bunch of colors, originally from matplotlib
diff --git a/dspy/datasets/dataloader.py b/dspy/datasets/dataloader.py
index 8900c107f..c9da0270d 100644
--- a/dspy/datasets/dataloader.py
+++ b/dspy/datasets/dataloader.py
@@ -1,10 +1,12 @@
-import dspy
import random
-from dspy.datasets import Dataset
+from collections.abc import Mapping
+from typing import List, Tuple, Union
from datasets import load_dataset
-from typing import Union, List, Tuple
-from collections.abc import Mapping
+
+import dspy
+from dspy.datasets.dataset import Dataset
+
class DataLoader(Dataset):
def __init__(self,):
diff --git a/dspy/datasets/dataset.py b/dspy/datasets/dataset.py
index 9cd279199..c66c5edc2 100644
--- a/dspy/datasets/dataset.py
+++ b/dspy/datasets/dataset.py
@@ -1,8 +1,9 @@
-import uuid
import random
+import uuid
-from dspy import Example
from dsp.utils import dotdict
+from dspy import Example
+
class Dataset:
def __init__(self, train_seed=0, train_size=None, eval_seed=0, dev_size=None, test_size=None):
diff --git a/dspy/datasets/gsm8k.py b/dspy/datasets/gsm8k.py
index 3b3514862..328d1aee8 100644
--- a/dspy/datasets/gsm8k.py
+++ b/dspy/datasets/gsm8k.py
@@ -1,8 +1,9 @@
-import tqdm
import random
+import tqdm
from datasets import load_dataset
+
class GSM8K:
def __init__(self) -> None:
super().__init__()
diff --git a/dspy/datasets/hotpotqa.py b/dspy/datasets/hotpotqa.py
index f1cc73482..9fbf08617 100644
--- a/dspy/datasets/hotpotqa.py
+++ b/dspy/datasets/hotpotqa.py
@@ -1,6 +1,7 @@
import random
from datasets import load_dataset
+
from dspy.datasets.dataset import Dataset
diff --git a/dspy/evaluate/__init__.py b/dspy/evaluate/__init__.py
index eea386c59..2526c2be0 100644
--- a/dspy/evaluate/__init__.py
+++ b/dspy/evaluate/__init__.py
@@ -1,4 +1,5 @@
+from dsp.utils import EM, normalize_text
+
+from .auto_evaluation import *
from .evaluate import Evaluate
from .metrics import *
-from .auto_evaluation import *
-from dsp.utils import EM, normalize_text
diff --git a/dspy/evaluate/auto_evaluation.py b/dspy/evaluate/auto_evaluation.py
index 1ec659540..569a5b16c 100644
--- a/dspy/evaluate/auto_evaluation.py
+++ b/dspy/evaluate/auto_evaluation.py
@@ -1,4 +1,5 @@
-import dspy
+import dspy
+
class AnswerCorrectnessSignature(dspy.Signature):
"""Verify that the predicted answer matches the gold answer."""
diff --git a/dspy/evaluate/evaluate.py b/dspy/evaluate/evaluate.py
index 0be330375..f38bcff38 100644
--- a/dspy/evaluate/evaluate.py
+++ b/dspy/evaluate/evaluate.py
@@ -1,11 +1,14 @@
-import dsp
-import tqdm
-import types
import threading
+import types
+
import pandas as pd
+import tqdm
+
+import dsp
try:
- from IPython.display import display as ipython_display, HTML
+ from IPython.display import HTML
+ from IPython.display import display as ipython_display
except ImportError:
ipython_display = print
HTML = lambda x: x
diff --git a/dspy/evaluate/metrics.py b/dspy/evaluate/metrics.py
index 79c6208af..0e25b5a11 100644
--- a/dspy/evaluate/metrics.py
+++ b/dspy/evaluate/metrics.py
@@ -2,6 +2,7 @@
import dsp
+
def answer_exact_match(example, pred, trace=None, frac=1.0):
assert(type(example.answer) is str or type(example.answer) is list)
diff --git a/dspy/experimental/__init__.py b/dspy/experimental/__init__.py
index b4385639d..36477304c 100644
--- a/dspy/experimental/__init__.py
+++ b/dspy/experimental/__init__.py
@@ -1,2 +1,2 @@
-from .synthetic_data import *
from .synthesizer import *
+from .synthetic_data import *
diff --git a/dspy/experimental/synthesizer.py b/dspy/experimental/synthesizer.py
index e247d35f3..a66bf4d2b 100644
--- a/dspy/experimental/synthesizer.py
+++ b/dspy/experimental/synthesizer.py
@@ -1,8 +1,11 @@
-import dspy
import random
from typing import List
-from tqdm import tqdm, trange
+
from datasets import Dataset
+from tqdm import tqdm, trange
+
+import dspy
+
def format_examples(examples: List[dspy.Example]):
if isinstance(examples, str):
@@ -14,11 +17,11 @@ def format_examples(examples: List[dspy.Example]):
input_keys = example.inputs().keys()
label_keys = example.labels().keys()
- formatted_example += f"Inputs:\n"
+ formatted_example += "Inputs:\n"
for key in input_keys:
formatted_example += f"{key}: {example[key]}\n"
- formatted_example += f"Outputs:\n"
+ formatted_example += "Outputs:\n"
for key in label_keys:
formatted_example += f"{key}: {example[key]}\n"
@@ -94,7 +97,7 @@ class Synthesizer:
self.generate_input_data = self.generate_input_data.insert(
-1,
field_name,
- output_field
+ output_field,
)
input_field = dspy.InputField(
@@ -104,7 +107,7 @@ class Synthesizer:
self.generate_output_data = self.generate_output_data.insert(
-1,
field_name,
- input_field
+ input_field,
)
for key in tqdm(output_keys, desc="Preparing Output Fields"):
@@ -123,7 +126,7 @@ class Synthesizer:
self.generate_output_data = self.generate_output_data.insert(
-1,
field_name,
- output_field
+ output_field,
)
return dspy.ChainOfThought(self.generate_input_data), dspy.Predict(self.generate_output_data)
@@ -167,7 +170,7 @@ class Synthesizer:
extention = mode or path.split(".")[-1]
dataset = Dataset.from_list(
- [example.toDict() for example in data]
+ [example.toDict() for example in data],
)
if extention == "csv":
diff --git a/dspy/experimental/synthetic_data.py b/dspy/experimental/synthetic_data.py
index d5177fcac..37c140352 100644
--- a/dspy/experimental/synthetic_data.py
+++ b/dspy/experimental/synthetic_data.py
@@ -1,8 +1,11 @@
-from pydantic import BaseModel
-import dspy
import random
from typing import List, Optional
+from pydantic import BaseModel
+
+import dspy
+
+
class descriptionSignature(dspy.Signature):
field_name = dspy.InputField(desc="name of a field")
example = dspy.InputField(desc="an example value for the field")
@@ -55,7 +58,7 @@ class SyntheticDataGenerator:
'__doc__': f"Generates the following outputs: {{{', '.join(properties.keys())}}}.",
'sindex': dspy.InputField(desc="a random string"),
**{field_name: dspy.OutputField(desc=properties[field_name].get('description', 'No description'))
- for field_name in properties.keys()}
+ for field_name in properties.keys()},
}
# # Usage example
diff --git a/dspy/functional/__init__.py b/dspy/functional/__init__.py
index 3ccced47b..746ac9318 100644
--- a/dspy/functional/__init__.py
+++ b/dspy/functional/__init__.py
@@ -1 +1 @@
-from .functional import cot, predictor, FunctionalModule, TypedPredictor, TypedChainOfThought
+from .functional import FunctionalModule, TypedChainOfThought, TypedPredictor, cot, predictor
diff --git a/dspy/functional/functional.py b/dspy/functional/functional.py
index cb7b3669a..97bbdb8d8 100644
--- a/dspy/functional/functional.py
+++ b/dspy/functional/functional.py
@@ -1,18 +1,15 @@
-from collections import defaultdict
import inspect
-import os
-import openai
-import dspy
+import json
import typing
+from typing import Annotated, List, Tuple
+
import pydantic
-from typing import Annotated, List, Tuple # noqa: UP035
-from dsp.templates import passages2text
-import json
-from dspy.primitives.prediction import Prediction
+import dspy
+from dsp.templates.utils import passages2text
+from dspy.primitives.prediction import Prediction
from dspy.signatures.signature import ensure_signature, make_signature
-
MAX_RETRIES = 3
@@ -199,7 +196,7 @@ class TypedPredictor(dspy.Module):
else:
# If there are no errors, we return the parsed results
return Prediction.from_completions(
- {key: [r[key] for r in parsed_results] for key in signature.output_fields}
+ {key: [r[key] for r in parsed_results] for key in signature.output_fields},
)
raise ValueError(
"Too many retries trying to get the correct output format. " + "Try simplifying the requirements.",
@@ -326,8 +323,8 @@ def gold_passages_retrieved(example, pred, _trace=None) -> bool:
def hotpot() -> None:
- from dsp.utils import deduplicate
import dspy.evaluate
+ from dsp.utils import deduplicate
from dspy.datasets import HotPotQA
from dspy.evaluate.evaluate import Evaluate
from dspy.teleprompt.bootstrap import BootstrapFewShot
diff --git a/dspy/predict/__init__.py b/dspy/predict/__init__.py
index 8b0770150..b176d2a0e 100644
--- a/dspy/predict/__init__.py
+++ b/dspy/predict/__init__.py
@@ -1,9 +1,9 @@
-from .predict import Predict
+from .aggregation import majority
from .chain_of_thought import ChainOfThought
-from .multi_chain_comparison import MultiChainComparison
from .chain_of_thought_with_hint import ChainOfThoughtWithHint
-from .react import ReAct
-from .aggregation import majority
+from .knn import KNN
+from .multi_chain_comparison import MultiChainComparison
+from .predict import Predict
from .program_of_thought import ProgramOfThought
+from .react import ReAct
from .retry import Retry
-from .knn import KNN
\ No newline at end of file
diff --git a/dspy/predict/aggregation.py b/dspy/predict/aggregation.py
index ca4154aa2..4cb6df3f9 100644
--- a/dspy/predict/aggregation.py
+++ b/dspy/predict/aggregation.py
@@ -1,6 +1,5 @@
-from dspy.primitives.prediction import Prediction, Completions
from dsp.utils import normalize_text
-
+from dspy.primitives.prediction import Completions, Prediction
default_normalize = lambda s: normalize_text(s) or None
diff --git a/dspy/predict/chain_of_thought.py b/dspy/predict/chain_of_thought.py
index f374aaabe..70bab0a7a 100644
--- a/dspy/predict/chain_of_thought.py
+++ b/dspy/predict/chain_of_thought.py
@@ -4,7 +4,6 @@ from dspy.signatures.signature import ensure_signature
from .predict import Predict
-
# TODO: FIXME: Insert this right before the *first* output field. Also rewrite this to use the new signature system.
# TODO: This shouldn't inherit from Predict. It should be a module that has one or two predictors.
diff --git a/dspy/predict/chain_of_thought_with_hint.py b/dspy/predict/chain_of_thought_with_hint.py
index 6f7856632..484a2a38a 100644
--- a/dspy/predict/chain_of_thought_with_hint.py
+++ b/dspy/predict/chain_of_thought_with_hint.py
@@ -3,8 +3,6 @@ import dspy
from .predict import Predict
-
-
# TODO: FIXME: Insert this right before the *first* output field. Also rewrite this to use the new signature system.
class ChainOfThoughtWithHint(Predict):
diff --git a/dspy/predict/knn.py b/dspy/predict/knn.py
index e0e49fe2b..e275ccb6a 100644
--- a/dspy/predict/knn.py
+++ b/dspy/predict/knn.py
@@ -1,7 +1,10 @@
from typing import List
+
import numpy as np
+
import dsp
+
class KNN:
def __init__(self, k: int, trainset: List[dsp.Example]):
self.k = k
diff --git a/dspy/predict/langchain.py b/dspy/predict/langchain.py
index 4be855e8d..86d439f50 100644
--- a/dspy/predict/langchain.py
+++ b/dspy/predict/langchain.py
@@ -1,18 +1,17 @@
import copy
import random
+from langchain_core.pydantic_v1 import Extra
+from langchain_core.runnables import Runnable
+
import dsp
import dspy
-
from dspy.predict.parameter import Parameter
from dspy.predict.predict import Predict
from dspy.primitives.prediction import Prediction
from dspy.signatures.field import InputField, OutputField
from dspy.signatures.signature import infer_prefix
-from langchain_core.pydantic_v1 import Extra
-from langchain_core.runnables import Runnable
-
# TODO: This class is currently hard to test, because it hardcodes gpt-4 usage:
# gpt4T = dspy.OpenAI(model='gpt-4-1106-preview', max_tokens=4000, model_type='chat')
diff --git a/dspy/predict/multi_chain_comparison.py b/dspy/predict/multi_chain_comparison.py
index ad69b646e..37b66afad 100644
--- a/dspy/predict/multi_chain_comparison.py
+++ b/dspy/predict/multi_chain_comparison.py
@@ -1,8 +1,8 @@
import dspy
from dspy.signatures.signature import ensure_signature
-from .predict import Predict
-from ..primitives.program import Module
+from ..primitives.program import Module
+from .predict import Predict
class MultiChainComparison(Module):
diff --git a/dspy/predict/predict.py b/dspy/predict/predict.py
index 60e668a83..fe6beb448 100644
--- a/dspy/predict/predict.py
+++ b/dspy/predict/predict.py
@@ -1,11 +1,11 @@
-import dsp
import random
+import dsp
from dspy.predict.parameter import Parameter
from dspy.primitives.prediction import Prediction
-
from dspy.signatures.signature import ensure_signature, signature_to_template
+
class Predict(Parameter):
def __init__(self, signature, **config):
self.stage = random.randbytes(8).hex()
diff --git a/dspy/predict/program_of_thought.py b/dspy/predict/program_of_thought.py
index f7e80eed7..1cfb54bd9 100644
--- a/dspy/predict/program_of_thought.py
+++ b/dspy/predict/program_of_thought.py
@@ -1,8 +1,10 @@
+import re
+
import dspy
from dspy.signatures.signature import ensure_signature
+
from ..primitives.program import Module
from ..primitives.python_interpreter import CodePrompt, PythonInterpreter
-import re
class ProgramOfThought(Module):
diff --git a/dspy/predict/react.py b/dspy/predict/react.py
index cf3be8b73..a465ee710 100644
--- a/dspy/predict/react.py
+++ b/dspy/predict/react.py
@@ -1,6 +1,7 @@
import dsp
import dspy
from dspy.signatures.signature import ensure_signature
+
from ..primitives.program import Module
from .predict import Predict
diff --git a/dspy/predict/retry.py b/dspy/predict/retry.py
index af1d37f98..dcef1488f 100644
--- a/dspy/predict/retry.py
+++ b/dspy/predict/retry.py
@@ -1,6 +1,7 @@
import copy
-import dspy
+
import dsp
+import dspy
from .predict import Predict
diff --git a/dspy/primitives/__init__.py b/dspy/primitives/__init__.py
index b7baf229b..e66783716 100644
--- a/dspy/primitives/__init__.py
+++ b/dspy/primitives/__init__.py
@@ -1,5 +1,5 @@
+from .assertions import *
from .example import *
-from .program import *
from .prediction import *
-from .assertions import *
+from .program import *
from .python_interpreter import *
diff --git a/dspy/primitives/assertions.py b/dspy/primitives/assertions.py
index 140f8bf6a..19f1aea3a 100644
--- a/dspy/primitives/assertions.py
+++ b/dspy/primitives/assertions.py
@@ -1,9 +1,10 @@
import inspect
+import logging
+import uuid
from typing import Any
+
import dsp
import dspy
-import logging
-import uuid
#################### Assertion Helpers ####################
diff --git a/dspy/primitives/module.py b/dspy/primitives/module.py
index 1b4a342f7..bee80338f 100644
--- a/dspy/primitives/module.py
+++ b/dspy/primitives/module.py
@@ -1,4 +1,5 @@
import copy
+
import ujson
diff --git a/dspy/primitives/program.py b/dspy/primitives/program.py
index 5e063bfb0..aa499d908 100644
--- a/dspy/primitives/program.py
+++ b/dspy/primitives/program.py
@@ -1,8 +1,9 @@
-from dspy.primitives.module import BaseModule
-from dspy.primitives.assertions import *
import re
+from dspy.primitives.assertions import *
+from dspy.primitives.module import BaseModule
+
class ProgramMeta(type):
pass
diff --git a/dspy/primitives/python_interpreter.py b/dspy/primitives/python_interpreter.py
index fd6166db5..1b7456c7b 100644
--- a/dspy/primitives/python_interpreter.py
+++ b/dspy/primitives/python_interpreter.py
@@ -12,10 +12,12 @@
# limitations under the License.
# =========== Copyright 2023 @ CAMEL-AI.org. All Rights Reserved. ===========
import ast
+import builtins
import difflib
import importlib
import re
import typing
+from collections.abc import Mapping
from typing import (
Any,
Dict,
@@ -24,8 +26,6 @@ from typing import (
Set,
Tuple,
)
-from collections.abc import Mapping
-import builtins
class InterpreterError(ValueError):
diff --git a/dspy/retrieve/chromadb_rm.py b/dspy/retrieve/chromadb_rm.py
index b4fc0bcf9..07ef407d1 100644
--- a/dspy/retrieve/chromadb_rm.py
+++ b/dspy/retrieve/chromadb_rm.py
@@ -2,10 +2,12 @@
Retriever model for chromadb
"""
-from typing import Optional, List, Union
+from typing import List, Optional, Union
+
+import backoff
import openai
+
import dspy
-import backoff
from dsp.utils import dotdict
try:
@@ -16,13 +18,13 @@ except Exception:
try:
import chromadb
- from chromadb.config import Settings
- from chromadb.utils import embedding_functions
+ import chromadb.utils.embedding_functions as ef
from chromadb.api.types import (
Embeddable,
EmbeddingFunction,
)
- import chromadb.utils.embedding_functions as ef
+ from chromadb.config import Settings
+ from chromadb.utils import embedding_functions
except ImportError:
chromadb = None
diff --git a/dspy/retrieve/databricks_rm.py b/dspy/retrieve/databricks_rm.py
index 74066ec5b..c275bdddc 100644
--- a/dspy/retrieve/databricks_rm.py
+++ b/dspy/retrieve/databricks_rm.py
@@ -1,10 +1,13 @@
-import dspy
import os
-import requests
-from typing import Union, List
from collections import defaultdict
+from typing import List, Union
+
+import requests
+
+import dspy
from dspy.primitives.prediction import Prediction
+
class DatabricksRM(dspy.Retrieve):
"""
A retrieval module that uses Databricks Vector Search Endpoint to return the top-k embeddings for a given query.
diff --git a/dspy/retrieve/deeplake_rm.py b/dspy/retrieve/deeplake_rm.py
index 87beb4856..235108a91 100644
--- a/dspy/retrieve/deeplake_rm.py
+++ b/dspy/retrieve/deeplake_rm.py
@@ -2,10 +2,12 @@
Retriever model for deeplake
"""
-from typing import Optional, List, Union
+from collections import defaultdict
+from typing import List, Optional, Union
+
import openai
+
import dspy
-from collections import defaultdict
from dsp.utils import dotdict
try:
diff --git a/dspy/retrieve/marqo_rm.py b/dspy/retrieve/marqo_rm.py
index a090e47f5..29c52fdb4 100644
--- a/dspy/retrieve/marqo_rm.py
+++ b/dspy/retrieve/marqo_rm.py
@@ -1,5 +1,6 @@
from collections import defaultdict
from typing import List, Union
+
import dspy
from dspy import dotdict
diff --git a/dspy/retrieve/mongodb_atlas_rm.py b/dspy/retrieve/mongodb_atlas_rm.py
index 70e784730..3cafb05f6 100644
--- a/dspy/retrieve/mongodb_atlas_rm.py
+++ b/dspy/retrieve/mongodb_atlas_rm.py
@@ -1,23 +1,25 @@
-from typing import List, Any
-import dspy
import os
+from typing import Any, List
+
+import backoff
from openai import (
- OpenAI,
APITimeoutError,
InternalServerError,
+ OpenAI,
RateLimitError,
UnprocessableEntityError,
)
-import backoff
+
+import dspy
try:
from pymongo import MongoClient
from pymongo.errors import (
- ConnectionFailure,
ConfigurationError,
- ServerSelectionTimeoutError,
+ ConnectionFailure,
InvalidURI,
OperationFailure,
+ ServerSelectionTimeoutError,
)
except ImportError:
raise ImportError(
diff --git a/dspy/retrieve/pgvector_rm.py b/dspy/retrieve/pgvector_rm.py
index 8403efb2e..cf1773f17 100644
--- a/dspy/retrieve/pgvector_rm.py
+++ b/dspy/retrieve/pgvector_rm.py
@@ -1,10 +1,12 @@
-import dspy
-import openai
from typing import List, Optional
+import openai
+
+import dspy
+
try:
- from pgvector.psycopg2 import register_vector
import psycopg2
+ from pgvector.psycopg2 import register_vector
from psycopg2 import sql
except ImportError:
raise ImportError(
diff --git a/dspy/retrieve/pinecone_rm.py b/dspy/retrieve/pinecone_rm.py
index b520ed5fa..0328bc6ba 100644
--- a/dspy/retrieve/pinecone_rm.py
+++ b/dspy/retrieve/pinecone_rm.py
@@ -3,11 +3,13 @@ Retriever model for Pinecone
Author: Dhar Rawal (@drawal1)
"""
-from dsp.utils import dotdict
-from typing import Optional, List, Union
-import dspy
+from typing import List, Optional, Union
+
import backoff
+import dspy
+from dsp.utils import dotdict
+
try:
import pinecone
except ImportError:
@@ -19,6 +21,7 @@ if pinecone is None:
)
import openai
+
try:
OPENAI_LEGACY = int(openai.version.__version__[0]) == 0
except Exception:
diff --git a/dspy/retrieve/qdrant_rm.py b/dspy/retrieve/qdrant_rm.py
index 46977a252..5c2af050b 100644
--- a/dspy/retrieve/qdrant_rm.py
+++ b/dspy/retrieve/qdrant_rm.py
@@ -1,11 +1,12 @@
from collections import defaultdict
-from typing import List, Union, Optional
+from typing import List, Optional, Union
+
import dspy
from dsp.utils import dotdict
try:
- from qdrant_client import QdrantClient
import fastembed
+ from qdrant_client import QdrantClient
except ImportError:
raise ImportError(
"The 'qdrant' extra is required to use QdrantRM. Install it with `pip install dspy-ai[qdrant]`",
diff --git a/dspy/retrieve/retrieve.py b/dspy/retrieve/retrieve.py
index 910858241..71b4f5734 100644
--- a/dspy/retrieve/retrieve.py
+++ b/dspy/retrieve/retrieve.py
@@ -1,8 +1,7 @@
-from typing import List, Union, Optional
-
-import dsp
import random
+from typing import List, Optional, Union
+import dsp
from dspy.predict.parameter import Parameter
from dspy.primitives.prediction import Prediction
diff --git a/dspy/retrieve/vectara_rm.py b/dspy/retrieve/vectara_rm.py
index 4d72ed7e8..047c70d6c 100644
--- a/dspy/retrieve/vectara_rm.py
+++ b/dspy/retrieve/vectara_rm.py
@@ -1,11 +1,11 @@
-from collections import defaultdict
-from typing import List, Union
-import dspy
-from typing import Optional
import json
import os
+from collections import defaultdict
+from typing import List, Optional, Union
+
import requests
+import dspy
from dsp.utils import dotdict
START_SNIPPET = "<%START%>"
diff --git a/dspy/retrieve/weaviate_rm.py b/dspy/retrieve/weaviate_rm.py
index 1ef7950c1..61e8d1ef0 100644
--- a/dspy/retrieve/weaviate_rm.py
+++ b/dspy/retrieve/weaviate_rm.py
@@ -1,7 +1,7 @@
-from typing import List, Union
+from typing import List, Optional, Union
+
import dspy
from dsp.utils import dotdict
-from typing import Optional
try:
import weaviate
diff --git a/dspy/retrieve/you_rm.py b/dspy/retrieve/you_rm.py
index 7321e67bd..25a62ccea 100644
--- a/dspy/retrieve/you_rm.py
+++ b/dspy/retrieve/you_rm.py
@@ -1,9 +1,10 @@
-import dspy
import os
+from typing import List, Optional, Union
+
import requests
-from dsp.utils import dotdict
-from typing import Union, List, Optional
+import dspy
+from dsp.utils import dotdict
class YouRM(dspy.Retrieve):
diff --git a/dspy/signatures/signature.py b/dspy/signatures/signature.py
index 8c1f16164..22bc6475d 100644
--- a/dspy/signatures/signature.py
+++ b/dspy/signatures/signature.py
@@ -1,12 +1,13 @@
import ast
-from copy import deepcopy
+import re
import typing
-import dsp
+from copy import deepcopy
+from typing import Any, Dict, Tuple, Type, Union # noqa: UP035
+
from pydantic import BaseModel, Field, create_model
from pydantic.fields import FieldInfo
-from typing import Any, Type, Union, Dict, Tuple # noqa: UP035
-import re
+import dsp
from dspy.signatures.field import InputField, OutputField, new_to_old_field
diff --git a/dspy/teleprompt/__init__.py b/dspy/teleprompt/__init__.py
index c4454725b..61425e5b9 100644
--- a/dspy/teleprompt/__init__.py
+++ b/dspy/teleprompt/__init__.py
@@ -1,9 +1,9 @@
-from .teleprompt import *
from .bootstrap import *
-from .vanilla import *
-from .random_search import *
from .finetune import *
-from .teleprompt_optuna import *
from .knn_fewshot import *
+from .random_search import *
from .signature_opt import SignatureOptimizer
-from .signature_opt_bayesian import BayesianSignatureOptimizer
\ No newline at end of file
+from .signature_opt_bayesian import BayesianSignatureOptimizer
+from .teleprompt import *
+from .teleprompt_optuna import *
+from .vanilla import *
diff --git a/dspy/teleprompt/bootstrap.py b/dspy/teleprompt/bootstrap.py
index 5aaefa271..6fc9dd2ef 100644
--- a/dspy/teleprompt/bootstrap.py
+++ b/dspy/teleprompt/bootstrap.py
@@ -1,16 +1,15 @@
-import dsp
-import tqdm
import random
import threading
-import dspy
+import tqdm
+import dsp
+import dspy
from dspy.primitives import Example
from .teleprompt import Teleprompter
from .vanilla import LabeledFewShot
-
# TODO: metrics should return an object with __bool__ basically, but fine if they're more complex.
# They can also be sortable.
diff --git a/dspy/teleprompt/finetune.py b/dspy/teleprompt/finetune.py
index 29b7887ab..8400fe774 100644
--- a/dspy/teleprompt/finetune.py
+++ b/dspy/teleprompt/finetune.py
@@ -1,17 +1,18 @@
import os
-import time
-import dsp
import random
+import time
import ujson
from datasets.fingerprint import Hasher
+import dsp
from dspy.signatures.signature import signature_to_template
-# from dspy.primitives import Example
+from .bootstrap import BootstrapFewShot
+# from dspy.primitives import Example
from .teleprompt import Teleprompter
-from .bootstrap import BootstrapFewShot
+
# from .vanilla import LabeledFewShot
# from dspy.evaluate.evaluate import Evaluate
diff --git a/dspy/teleprompt/knn_fewshot.py b/dspy/teleprompt/knn_fewshot.py
index 4bc447abd..b99944707 100644
--- a/dspy/teleprompt/knn_fewshot.py
+++ b/dspy/teleprompt/knn_fewshot.py
@@ -1,9 +1,11 @@
-from typing import List
import types
+from typing import List
+
import dsp
+from dspy.teleprompt import BootstrapFewShot
from .teleprompt import Teleprompter
-from dspy.teleprompt import BootstrapFewShot
+
class KNNFewShot(Teleprompter):
def __init__(self, KNN, k: int, trainset: List[dsp.Example]):
diff --git a/dspy/teleprompt/random_search.py b/dspy/teleprompt/random_search.py
index 3f7f2cd63..bc0e5ef61 100644
--- a/dspy/teleprompt/random_search.py
+++ b/dspy/teleprompt/random_search.py
@@ -1,13 +1,11 @@
import random
+from dspy.evaluate.evaluate import Evaluate
from dspy.teleprompt.teleprompt import Teleprompter
from .bootstrap import BootstrapFewShot
from .vanilla import LabeledFewShot
-from dspy.evaluate.evaluate import Evaluate
-
-
# TODO: Don't forget dealing with the raw demos.
# TODO: Deal with the (pretty common) case of having a metric for filtering and a separate metric for eval.
# The metric itself may tell though by the presence of trace.
diff --git a/dspy/teleprompt/signature_opt.py b/dspy/teleprompt/signature_opt.py
index 3163acf3f..d71aa69a7 100644
--- a/dspy/teleprompt/signature_opt.py
+++ b/dspy/teleprompt/signature_opt.py
@@ -1,9 +1,10 @@
+from collections import defaultdict
+
import dsp
import dspy
-from dspy.teleprompt.teleprompt import Teleprompter
-from dspy.signatures import Signature
from dspy.evaluate.evaluate import Evaluate
-from collections import defaultdict
+from dspy.signatures import Signature
+from dspy.teleprompt.teleprompt import Teleprompter
"""
USAGE SUGGESTIONS:
diff --git a/dspy/teleprompt/signature_opt_bayesian.py b/dspy/teleprompt/signature_opt_bayesian.py
index 85298276a..e316462c9 100644
--- a/dspy/teleprompt/signature_opt_bayesian.py
+++ b/dspy/teleprompt/signature_opt_bayesian.py
@@ -1,14 +1,16 @@
+import math
+import random
+from collections import defaultdict
+
+import optuna
+
import dsp
import dspy
-from dspy.signatures.signature import signature_to_template
-from dspy.teleprompt.teleprompt import Teleprompter
-from dspy.signatures import Signature
from dspy.evaluate.evaluate import Evaluate
-from collections import defaultdict
-import random
+from dspy.signatures import Signature
+from dspy.signatures.signature import signature_to_template
from dspy.teleprompt import BootstrapFewShot
-import optuna
-import math
+from dspy.teleprompt.teleprompt import Teleprompter
"""
USAGE SUGGESTIONS:
diff --git a/dspy/teleprompt/teleprompt_optuna.py b/dspy/teleprompt/teleprompt_optuna.py
index 3686e66b7..501bd71fd 100644
--- a/dspy/teleprompt/teleprompt_optuna.py
+++ b/dspy/teleprompt/teleprompt_optuna.py
@@ -1,10 +1,10 @@
import optuna
+from dspy.evaluate.evaluate import Evaluate
from dspy.teleprompt.teleprompt import Teleprompter
from .bootstrap import BootstrapFewShot
-from dspy.evaluate.evaluate import Evaluate
class BootstrapFewShotWithOptuna(Teleprompter):
def __init__(self, metric, teacher_settings={}, max_bootstrapped_demos=4, max_labeled_demos=16, max_rounds=1, num_candidate_programs=16, num_threads=6):
diff --git a/dspy/utils/dummies.py b/dspy/utils/dummies.py
index 1f6390536..d97d59b5a 100644
--- a/dspy/utils/dummies.py
+++ b/dspy/utils/dummies.py
@@ -1,9 +1,11 @@
import random
+import re
from typing import Union
-from dsp.modules import LM
+
import numpy as np
+
+from dsp.modules import LM
from dsp.utils.utils import dotdict
-import re
class DummyLM(LM):
diff --git a/examples/longformqa/utils.py b/examples/longformqa/utils.py
index e4ebadc82..bdae81a24 100644
--- a/examples/longformqa/utils.py
+++ b/examples/longformqa/utils.py
@@ -1,8 +1,10 @@
-import regex as re
import nltk
+import regex as re
+
nltk.download('punkt')
from nltk.tokenize import sent_tokenize
+
def extract_text_by_citation(paragraph):
citation_regex = re.compile(r'(.*?)(\[\d+\]\.)', re.DOTALL)
parts_with_citation = citation_regex.findall(paragraph)
diff --git a/inspect-app/app.py b/inspect-app/app.py
index 5ea50dd92..14ea40551 100644
--- a/inspect-app/app.py
+++ b/inspect-app/app.py
@@ -1,8 +1,9 @@
-from flask import Flask, request, jsonify
-from flask_cors import CORS
-import boto3
import time
+import boto3
+from flask import Flask, jsonify, request
+from flask_cors import CORS
+
app = Flask(__name__)
CORS(app)
diff --git a/pyproject.toml b/pyproject.toml
index 1c0f75968..e2c719eaf 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -218,6 +218,8 @@ select = [
"ERA",
# pandas-vet
"PD",
+ # Import sort
+ "I",
# avoid shadowing
"PLW",
]
diff --git a/setup.py b/setup.py
index d31e9c791..3f3bb1858 100644
--- a/setup.py
+++ b/setup.py
@@ -1,4 +1,4 @@
-from setuptools import setup, find_packages
+from setuptools import find_packages, setup
# Read the content of the README file
with open('README.md', encoding='utf-8') as f:
diff --git a/testing/optimizer_tester.py b/testing/optimizer_tester.py
index e55f4dd4c..68d394601 100644
--- a/testing/optimizer_tester.py
+++ b/testing/optimizer_tester.py
@@ -1,18 +1,21 @@
-from .tasks.scone import ScoNeTask
-from .tasks.hotpotqa import HotPotQATask
-from .tasks.gsm8k import GSM8KTask
-from .tasks.biodex import BioDexTask
-from .tasks.tweet import TweetTask
-from .tasks.tweet_metric import TweetMetricTask
-from dspy.evaluate import Evaluate
-from dotenv import load_dotenv
-import openai
-import os
-import dspy
import csv
import datetime
+import os
from timeit import default_timer as timer
+import openai
+from dotenv import load_dotenv
+
+import dspy
+from dspy.evaluate import Evaluate
+
+from .tasks.biodex import BioDexTask
+from .tasks.gsm8k import GSM8KTask
+from .tasks.hotpotqa import HotPotQATask
+from .tasks.scone import ScoNeTask
+from .tasks.tweet import TweetTask
+from .tasks.tweet_metric import TweetMetricTask
+
datasets = ["ScoNe", "HotPotQA", "GSM8K", "BioDex", "Tweet"]
class OptimizerTester:
diff --git a/testing/tasks/__init__.py b/testing/tasks/__init__.py
index f5034e36f..0b1ee32ec 100644
--- a/testing/tasks/__init__.py
+++ b/testing/tasks/__init__.py
@@ -1,7 +1,7 @@
from .base_task import BaseTask
+from .biodex import BioDexTask
from .gsm8k import GSM8KTask
from .hotpotqa import HotPotQATask
from .scone import ScoNeTask
-from .biodex import BioDexTask
from .tweet import TweetTask
-from .tweet_metric import TweetMetricTask
\ No newline at end of file
+from .tweet_metric import TweetMetricTask
diff --git a/testing/tasks/base_task.py b/testing/tasks/base_task.py
index c7f8f42b6..089e6d7be 100644
--- a/testing/tasks/base_task.py
+++ b/testing/tasks/base_task.py
@@ -1,5 +1,6 @@
from abc import ABC, abstractmethod
+
class BaseTask(ABC):
def __init__(self):
pass
diff --git a/testing/tasks/biodex.py b/testing/tasks/biodex.py
index f98da1141..6e51745b3 100644
--- a/testing/tasks/biodex.py
+++ b/testing/tasks/biodex.py
@@ -1,20 +1,24 @@
from __future__ import annotations
-from .base_task import BaseTask
-import dspy
-from dspy.evaluate import Evaluate
-from dsp.utils import deduplicate
-import tqdm
-import datasets
+
import math
-from functools import lru_cache
import os
-import re
import pickle
+import re
+from collections import defaultdict
+from enum import Enum
+from functools import lru_cache
+from typing import DefaultDict, Dict, List, Optional, Union
+
+import datasets
+import tqdm
from pydantic import BaseModel
from rapidfuzz import process
-from typing import Union, Optional, Dict, DefaultDict, List
-from enum import Enum
-from collections import defaultdict
+
+import dspy
+from dsp.utils import deduplicate
+from dspy.evaluate import Evaluate
+
+from .base_task import BaseTask
# Must point to a MedDra download with mdhier.asc
data_dir = '/future/u/okhattab/data/2023/MedDraV2/meddra_23_0_english/MedAscii' # NOTE: EDIT THIS LINE
@@ -529,6 +533,7 @@ class CompilationMetric:
metricR = CompilationMetric(field='reactions', metric='recall')
from collections import Counter
+
def reduce_grounded_reactions(grounded_reactions):
scores = {}
for score, reaction in grounded_reactions:
diff --git a/testing/tasks/gsm8k.py b/testing/tasks/gsm8k.py
index 6724ebdcd..04f20d177 100644
--- a/testing/tasks/gsm8k.py
+++ b/testing/tasks/gsm8k.py
@@ -1,11 +1,14 @@
-import dspy
-from .base_task import BaseTask
-from dspy.datasets.gsm8k import gsm8k_metric
import random
-import tqdm
+import tqdm
from datasets import load_dataset
+import dspy
+from dspy.datasets.gsm8k import gsm8k_metric
+
+from .base_task import BaseTask
+
+
class CoT(dspy.Module):
def __init__(self):
super().__init__()
diff --git a/testing/tasks/hotpotqa.py b/testing/tasks/hotpotqa.py
index 19a5075fc..f0bf17078 100644
--- a/testing/tasks/hotpotqa.py
+++ b/testing/tasks/hotpotqa.py
@@ -1,7 +1,9 @@
import dspy
+from dspy.datasets import HotPotQA
from dspy.evaluate import Evaluate
+
from .base_task import BaseTask
-from dspy.datasets import HotPotQA
+
class MultiHop(dspy.Module):
def __init__(self,passages_per_hop):
diff --git a/testing/tasks/scone.py b/testing/tasks/scone.py
index 0ee5f0391..27e8691ac 100644
--- a/testing/tasks/scone.py
+++ b/testing/tasks/scone.py
@@ -1,9 +1,13 @@
import glob
import os
+import random
+
import pandas as pd
+
import dspy
+
from .base_task import BaseTask
-import random
+
def load_scone(dirname):
dfs = []
diff --git a/testing/tasks/tweet.py b/testing/tasks/tweet.py
index 30bdef68a..73870a059 100644
--- a/testing/tasks/tweet.py
+++ b/testing/tasks/tweet.py
@@ -1,10 +1,14 @@
-import dspy
-from dspy.datasets import HotPotQA
-from .base_task import BaseTask
+import os
from functools import lru_cache
+
import openai
from dotenv import load_dotenv
-import os
+
+import dspy
+from dspy.datasets import HotPotQA
+
+from .base_task import BaseTask
+
class TweetSignature(dspy.Signature):
("""Given context and a question, answer with a tweet""")
diff --git a/testing/tasks/tweet_metric.py b/testing/tasks/tweet_metric.py
index baa7024ab..b91458c48 100644
--- a/testing/tasks/tweet_metric.py
+++ b/testing/tasks/tweet_metric.py
@@ -1,14 +1,18 @@
-import dspy
-from dspy.datasets import HotPotQA
-from .base_task import BaseTask
-from dspy import Example
+import os
+import uuid
from functools import lru_cache
+
import openai
from dotenv import load_dotenv
-import os
-import uuid
from tqdm import tqdm
+import dspy
+from dspy import Example
+from dspy.datasets import HotPotQA
+
+from .base_task import BaseTask
+
+
class TweetSignature(dspy.Signature):
("""Given context and a question, answer with a tweet""")