instructor 1.2.4__tar.gz → 1.2.6__tar.gz

{instructor-1.2.4 → instructor-1.2.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: instructor
-Version: 1.2.4
+Version: 1.2.6
 Summary: structured outputs for llm
 Home-page: https://github.com/jxnl/instructor
 License: MIT
@@ -16,6 +16,7 @@ Classifier: Programming Language :: Python :: 3.12
 Provides-Extra: anthropic
 Provides-Extra: cohere
 Provides-Extra: groq
+Provides-Extra: litellm
 Provides-Extra: mistralai
 Provides-Extra: test-docs
 Requires-Dist: aiohttp (>=3.9.1,<4.0.0)
@@ -25,11 +26,11 @@ Requires-Dist: diskcache (>=5.6.3,<6.0.0) ; extra == "test-docs"
 Requires-Dist: docstring-parser (>=0.16,<0.17)
 Requires-Dist: fastapi (>=0.109.2,<0.110.0) ; extra == "test-docs"
 Requires-Dist: groq (>=0.4.2,<0.5.0) ; extra == "groq" or extra == "test-docs"
-Requires-Dist: litellm (>=1.0.0,<2.0.0) ; extra == "test-docs"
+Requires-Dist: litellm (>=1.35.31,<2.0.0) ; extra == "test-docs" or extra == "litellm"
 Requires-Dist: mistralai (>=0.1.8,<0.2.0) ; extra == "test-docs" or extra == "mistralai"
 Requires-Dist: openai (>=1.1.0,<2.0.0)
 Requires-Dist: pandas (>=2.2.0,<3.0.0) ; extra == "test-docs"
-Requires-Dist: pydantic (==2.7.0)
+Requires-Dist: pydantic (>=2.7.0,<3.0.0)
 Requires-Dist: pydantic-core (>=2.18.0,<3.0.0)
 Requires-Dist: pydantic_extra_types (>=2.6.0,<3.0.0) ; extra == "test-docs"
 Requires-Dist: redis (>=5.0.1,<6.0.0) ; extra == "test-docs"

{instructor-1.2.4 → instructor-1.2.6}/instructor/__init__.py

@@ -1,4 +1,4 @@
-import importlib
+import importlib.util
 
 from .mode import Mode
 from .process_response import handle_response_model

{instructor-1.2.4 → instructor-1.2.6}/instructor/cli/files.py

@@ -1,6 +1,8 @@
+# type: ignore - stub mismatched
+
 import time
 from datetime import datetime
-from typing import List
+from typing import Literal, cast
 
 import openai
 import typer
@@ -14,7 +16,7 @@ console = Console()
 
 
 # Sample response data
-def generate_file_table(files: List[openai.types.FileObject]) -> Table:
+def generate_file_table(files: list[openai.types.FileObject]) -> Table:
     table = Table(
         title="OpenAI Files",
     )
@@ -36,7 +38,7 @@ def generate_file_table(files: List[openai.types.FileObject]) -> Table:
     return table
 
 
-def get_files() -> List[openai.types.FileObject]:
+def get_files() -> list[openai.types.FileObject]:
     files = client.files.list()
     files = files.data
     files = sorted(files, key=lambda x: x.created_at, reverse=True)
@@ -50,15 +52,17 @@ def get_file_status(file_id: str) -> str:
 
 @app.command(
     help="Upload a file to OpenAI's servers, will monitor the upload status until it is processed",
-)  # type: ignore[misc]
+)
 def upload(
-    filepath: str = typer.Argument(..., help="Path to the file to upload"),
+    filepath: str = typer.Argument(help="Path to the file to upload"),
     purpose: str = typer.Option("fine-tune", help="Purpose of the file"),
     poll: int = typer.Option(5, help="Polling interval in seconds"),
 ) -> None:
+    # Literals aren't supported by Typer yet.
+    file_purpose = cast(Literal["fine-tune", "assistants"], purpose)
     with open(filepath, "rb") as file:
-        response = client.files.create(file=file, purpose=purpose)
-    file_id = response["id"]
+        response = client.files.create(file=file, purpose=file_purpose)
+    file_id = response["id"]  # type: ignore - types might be out of date
     with console.status(f"Monitoring upload: {file_id}...") as status:
         status.spinner_style = "dots"
         while True:
@@ -71,10 +75,10 @@ def upload(
 
 @app.command(
     help="Download a file from OpenAI's servers",
-)  # type: ignore[misc]
+)
 def download(
-    file_id: str = typer.Argument(..., help="ID of the file to download"),
-    output: str = typer.Argument(..., help="Output path for the downloaded file"),
+    file_id: str = typer.Argument(help="ID of the file to download"),
+    output: str = typer.Argument(help="Output path for the downloaded file"),
 ) -> None:
     with console.status(f"[bold green]Downloading file {file_id}...", spinner="dots"):
         content = client.files.download(file_id)
@@ -85,8 +89,8 @@ def download(
 
 @app.command(
     help="Delete a file from OpenAI's servers",
-)  # type: ignore[misc]
-def delete(file_id: str = typer.Argument(..., help="ID of the file to delete")) -> None:
+)
+def delete(file_id: str = typer.Argument(help="ID of the file to delete")) -> None:
     with console.status(f"[bold red]Deleting file {file_id}...", spinner="dots"):
         try:
             client.files.delete(file_id)
@@ -98,9 +102,9 @@ def delete(file_id: str = typer.Argument(..., help="ID of the file to delete"))
 
 @app.command(
     help="Monitor the status of a file on OpenAI's servers",
-)  # type: ignore[misc]
+)
 def status(
-    file_id: str = typer.Argument(..., help="ID of the file to check the status of"),
+    file_id: str = typer.Argument(help="ID of the file to check the status of"),
 ) -> None:
     with console.status(f"Monitoring status of file {file_id}...") as status:
         while True:
@@ -113,7 +117,7 @@ def status(
 
 @app.command(
     help="List the files on OpenAI's servers",
-)  # type: ignore[misc]
+)
 def list() -> None:
     files = get_files()
     console.log(generate_file_table(files))

{instructor-1.2.4 → instructor-1.2.6}/instructor/cli/hub.py

@@ -58,7 +58,7 @@ class HubClient:
         else:
             raise Exception(f"Failed to fetch cookbooks: {response.status_code}")
 
-    def get_content_markdown(self, branch, slug):
+    def get_content_markdown(self, branch: str, slug: str) -> str:
         """Get markdown content."""
         url = f"{self.base_url}/api/{branch}/items/{slug}/md/"
         response = httpx.get(url)
@@ -67,7 +67,7 @@ class HubClient:
         else:
             raise Exception(f"Failed to fetch markdown content: {response.status_code}")
 
-    def get_content_python(self, branch, slug):
+    def get_content_python(self, branch: str, slug: str) -> str:
         """Get Python code blocks from content."""
         url = f"{self.base_url}/api/{branch}/items/{slug}/py/"
         response = httpx.get(url)
@@ -76,12 +76,12 @@ class HubClient:
         else:
             raise Exception(f"Failed to fetch Python content: {response.status_code}")
 
-    def get_cookbook_id(self, id: int, branch: str = "main") -> HubPage:
+    def get_cookbook_id(self, id: int, branch: str = "main") -> Optional[HubPage]:
         for cookbook in self.get_cookbooks(branch):
             if cookbook.id == id:
                 return cookbook
 
-    def get_cookbook_slug(self, slug: str, branch: str = "main") -> HubPage:
+    def get_cookbook_slug(self, slug: str, branch: str = "main") -> Optional[HubPage]:
         for cookbook in self.get_cookbooks(branch):
             if cookbook.slug == slug:
                 return cookbook
@@ -155,7 +155,7 @@ def pull(
 
     if file:
         with open(file, "w") as f:
-            f.write(output)
+            f.write(output)  # type: ignore - markdown is writable
             return
 
     if page:

{instructor-1.2.4 → instructor-1.2.6}/instructor/cli/jobs.py

@@ -1,13 +1,13 @@
-from typing import Dict, List, Union
+from typing import Optional, TypedDict
 from openai import OpenAI
 
+from openai.types.fine_tuning.job_create_params import Hyperparameters
 import typer
 import time
 from rich.live import Live
 from rich.table import Table
 from rich.console import Console
 from datetime import datetime
-from typing import cast
 from openai.types.fine_tuning import FineTuningJob
 
 client = OpenAI()
@@ -15,10 +15,15 @@ app = typer.Typer()
 console = Console()
 
 
-def generate_table(jobs: List[FineTuningJob]) -> Table:
+class FuneTuningParams(TypedDict, total=False):
+    hyperparameters: Hyperparameters
+    validation_file: Optional[str]
+    suffix: Optional[str]
+
+
+def generate_table(jobs: list[FineTuningJob]) -> Table:
     # Sorting the jobs by creation time
-    jobs = sorted(jobs, key=lambda x: (cast(FineTuningJob, x)).created_at, reverse=True)
-    jobs = cast(List[FineTuningJob], jobs)
+    jobs = sorted(jobs, key=lambda x: x.created_at, reverse=True)
 
     table = Table(
         title="OpenAI Fine Tuning Job Monitoring",
@@ -66,7 +71,7 @@ def status_color(status: str) -> str:
     )
 
 
-def get_jobs(limit: int = 5) -> List[FineTuningJob]:
+def get_jobs(limit: int = 5) -> list[FineTuningJob]:
     return client.fine_tuning.jobs.list(limit=limit).data
 
 
@@ -78,7 +83,7 @@ def get_file_status(file_id: str) -> str:
 @app.command(
     name="list",
     help="Monitor the status of the most recent fine-tuning jobs.",
-)  # type: ignore[misc]
+)
 def watch(
     limit: int = typer.Option(5, help="Limit the number of jobs to monitor"),
     poll: int = typer.Option(5, help="Polling interval in seconds"),
@@ -97,24 +102,24 @@ def watch(
 
 @app.command(
     help="Create a fine-tuning job from an existing ID.",
-)  # type: ignore[misc]
+)
 def create_from_id(
-    id: str = typer.Argument(..., help="ID of the existing fine-tuning job"),
+    id: str = typer.Argument(help="ID of the existing fine-tuning job"),
     model: str = typer.Option("gpt-3.5-turbo", help="Model to use for fine-tuning"),
-    n_epochs: int = typer.Option(
+    n_epochs: Optional[int] = typer.Option(
         None, help="Number of epochs for fine-tuning", show_default=False
     ),
-    batch_size: int = typer.Option(
+    batch_size: Optional[int] = typer.Option(
         None, help="Batch size for fine-tuning", show_default=False
     ),
-    learning_rate_multiplier: float = typer.Option(
+    learning_rate_multiplier: Optional[float] = typer.Option(
         None, help="Learning rate multiplier for fine-tuning", show_default=False
     ),
-    validation_file_id: str = typer.Option(
+    validation_file_id: Optional[str] = typer.Option(
         None, help="ID of the uploaded validation file"
     ),
 ) -> None:
-    hyperparameters_dict: Dict[str, Union[int, float, str]] = {}
+    hyperparameters_dict: Hyperparameters = {}
     if n_epochs is not None:
         hyperparameters_dict["n_epochs"] = n_epochs
     if batch_size is not None:
@@ -128,7 +133,7 @@ def create_from_id(
         job = client.fine_tuning.jobs.create(
             training_file=id,
             model=model,
-            hyperparameters=hyperparameters_dict if hyperparameters_dict else None,
+            hyperparameters=hyperparameters_dict,
             validation_file=validation_file_id if validation_file_id else None,
         )
         console.log(f"[bold green]Fine-tuning job created with ID: {job.id}")
@@ -137,24 +142,28 @@ def create_from_id(
 
 @app.command(
     help="Create a fine-tuning job from a file.",
-)  # type: ignore[misc]
+)
 def create_from_file(
-    file: str = typer.Argument(..., help="Path to the file for fine-tuning"),
+    file: str = typer.Argument(help="Path to the file for fine-tuning"),
     model: str = typer.Option("gpt-3.5-turbo", help="Model to use for fine-tuning"),
     poll: int = typer.Option(2, help="Polling interval in seconds"),
-    n_epochs: int = typer.Option(
+    n_epochs: Optional[int] = typer.Option(
         None, help="Number of epochs for fine-tuning", show_default=False
     ),
-    batch_size: int = typer.Option(
+    batch_size: Optional[int] = typer.Option(
         None, help="Batch size for fine-tuning", show_default=False
     ),
-    learning_rate_multiplier: float = typer.Option(
+    learning_rate_multiplier: Optional[float] = typer.Option(
         None, help="Learning rate multiplier for fine-tuning", show_default=False
     ),
-    validation_file: str = typer.Option(None, help="Path to the validation file"),
-    model_suffix: str = typer.Option(None, help="Suffix to identify the model"),
+    validation_file: Optional[str] = typer.Option(
+        None, help="Path to the validation file"
+    ),
+    model_suffix: Optional[str] = typer.Option(
+        None, help="Suffix to identify the model"
+    ),
 ) -> None:
-    hyperparameters_dict: Dict[str, Union[int, float, str]] = {}
+    hyperparameters_dict: Hyperparameters = {}
     if n_epochs is not None:
         hyperparameters_dict["n_epochs"] = n_epochs
     if batch_size is not None:
@@ -177,8 +186,9 @@ def create_from_file(
         status.spinner_style = "dots"
         while True:
             file_status = get_file_status(file_id)
-            if validation_file_id:
-                validation_file_status = get_file_status(validation_file_id)
+            validation_file_status = (
+                get_file_status(validation_file_id) if validation_file_id else ""
+            )
 
             if file_status == "processed" and (
                 not validation_file_id or validation_file_status == "processed"
@@ -192,7 +202,7 @@ def create_from_file(
 
             time.sleep(poll)
 
-    additional_params: Dict[str, Union[str, Dict[str, Union[int, float, str]]]] = {}
+    additional_params: FuneTuningParams = {}
     if hyperparameters_dict:
         additional_params["hyperparameters"] = hyperparameters_dict
     if validation_file:
@@ -218,9 +228,9 @@ def create_from_file(
 
 @app.command(
     help="Cancel a fine-tuning job.",
-)  # type: ignore[misc]
+)
 def cancel(
-    id: str = typer.Argument(..., help="ID of the fine-tuning job to cancel"),
+    id: str = typer.Argument(help="ID of the fine-tuning job to cancel"),
 ) -> None:
     with console.status(f"[bold red]Cancelling job {id}...", spinner="dots"):
         try:

{instructor-1.2.4 → instructor-1.2.6}/instructor/cli/usage.py

@@ -1,9 +1,11 @@
-from typing import List, Dict, Any, Union, DefaultDict
+from typing import Any, Union
+from collections.abc import Awaitable
 from datetime import datetime, timedelta
 import typer
 import os
 import aiohttp
 import asyncio
+from builtins import list as List
 from collections import defaultdict
 from rich.console import Console
 from rich.table import Table
@@ -18,7 +20,7 @@ console = Console()
 api_key = os.environ.get("OPENAI_API_KEY")
 
 
-async def fetch_usage(date: str) -> Dict[str, Any]:
+async def fetch_usage(date: str) -> dict[str, Any]:
     headers = {"Authorization": f"Bearer {api_key}"}
     url = f"https://api.openai.com/v1/usage?date={date}"
     async with aiohttp.ClientSession() as session:
@@ -26,9 +28,9 @@ async def fetch_usage(date: str) -> Dict[str, Any]:
             return await resp.json()
 
 
-async def get_usage_for_past_n_days(n_days: int) -> List[Dict[str, Any]]:
-    tasks = []
-    all_data = []
+async def get_usage_for_past_n_days(n_days: int) -> list[dict[str, Any]]:
+    tasks: List[Awaitable[dict[str, Any]]] = []  # noqa: UP006 - conflicting with the fn name
+    all_data: List[dict[str, Any]] = []  # noqa: UP006 - conflicting with the fn name
     with Progress() as progress:
         if n_days > 1:
             task = progress.add_task("[green]Fetching usage data...", total=n_days)
@@ -46,13 +48,7 @@ async def get_usage_for_past_n_days(n_days: int) -> List[Dict[str, Any]]:
 
 
 # Define the cost per unit for each model
-# Add temporary body type hint here because mypy may infer the dict type
-# from the first few items (?) in the dict, which may not be representative of
-# the entire dict.
-MODEL_COSTS: Dict[
-    ModelNames,
-    Union[Dict[str, float], float],
-] = {
+MODEL_COSTS = {
     "gpt-4-0125-preview": {"prompt": 0.01 / 1000, "completion": 0.03 / 1000},
     "gpt-4-turbo-preview": {"prompt": 0.01 / 1000, "completion": 0.03 / 1000},
     "gpt-4-1106-preview": {"prompt": 0.01 / 1000, "completion": 0.03 / 1000},
@@ -79,7 +75,7 @@ MODEL_COSTS: Dict[
 
 def get_model_cost(
     model: ModelNames,
-) -> Union[Dict[str, float], float]:
+) -> Union[dict[str, float], float]:
     """Get the cost details for a given model."""
     if model in MODEL_COSTS:
         return MODEL_COSTS[model]
@@ -104,7 +100,7 @@ def calculate_cost(
     """Calculate the cost based on the snapshot ID and number of tokens."""
     cost = get_model_cost(snapshot_id)
 
-    if isinstance(cost, float):
+    if isinstance(cost, (float, int)):
         return cost * (n_context_tokens + n_generated_tokens)
 
     prompt_cost = cost["prompt"] * n_context_tokens
@@ -112,13 +108,13 @@ def calculate_cost(
     return prompt_cost + completion_cost
 
 
-def group_and_sum_by_date_and_snapshot(usage_data: List[Dict[str, Any]]) -> Table:
+def group_and_sum_by_date_and_snapshot(usage_data: list[dict[str, Any]]) -> Table:
     """Group and sum the usage data by date and snapshot, including costs."""
-    summary: DefaultDict[
-        str, DefaultDict[str, Dict[str, Union[int, float]]]
-    ] = defaultdict(
-        lambda: defaultdict(
-            lambda: {"total_requests": 0, "total_tokens": 0, "total_cost": 0.0}
+    summary: defaultdict[str, defaultdict[str, dict[str, Union[int, float]]]] = (
+        defaultdict(
+            lambda: defaultdict(
+                lambda: {"total_requests": 0, "total_tokens": 0, "total_cost": 0.0}
+            )
         )
     )
 
@@ -160,7 +156,7 @@ def group_and_sum_by_date_and_snapshot(usage_data: List[Dict[str, Any]]) -> Tabl
     return table
 
 
-@app.command(help="Displays OpenAI API usage data for the past N days.")  # type: ignore[misc]
+@app.command(help="Displays OpenAI API usage data for the past N days.")
 def list(
     n: int = typer.Option(0, help="Number of days."),
 ) -> None: