synth-ai 0.2.6.dev1__py3-none-any.whl → 0.4.3__py3-none-any.whl

synth_ai/v0/tracing_v1/upload.py

@@ -1,525 +0,0 @@
-import asyncio
-import json
-import logging
-import os
-import ssl
-import time
-import warnings
-from typing import Any, TypedDict
-
-import requests
-from dotenv import load_dotenv
-from pydantic import BaseModel, ConfigDict, field_validator
-from requests.adapters import HTTPAdapter
-from urllib3.poolmanager import PoolManager
-
-from .abstractions import Dataset, SystemTrace
-from .events.store import event_store
-
-load_dotenv()
-
-
-# Issue deprecation warning and raise not supported error for backend upload functionality
-def _raise_not_supported_error():
-    """Raise an error for backend upload functionality that is no longer supported."""
-    warnings.warn(
-        "Backend upload functionality in synth_ai.tracing_v1 is no longer supported. "
-        "Please use synth_ai.tracing_v2 for new tracing functionality.",
-        DeprecationWarning,
-        stacklevel=3,
-    )
-    raise NotImplementedError(
-        "Backend upload functionality is no longer supported in synth_ai.tracing_v1. "
-        "Please use synth_ai.tracing_v2 instead."
-    )
-
-
-# NOTE: This may cause memory issues in the future
-def validate_json(data: dict[str, Any]) -> None:
-    """Validate that a dictionary contains only JSON-serializable values.
-
-    Args:
-        data: Dictionary to validate for JSON serialization
-
-    Raises:
-        ValueError: If the dictionary contains non-serializable values
-    """
-
-    try:
-        json.dumps(data)
-    except (TypeError, OverflowError) as e:
-        raise ValueError(f"Contains non-JSON-serializable values: {e}. {data}")
-
-
-def createPayload(dataset: Dataset, traces: list[SystemTrace]) -> dict[str, Any]:
-    payload = {
-        "traces": [trace.to_dict() for trace in traces],  # Convert SystemTrace objects to dicts
-        "dataset": dataset.to_dict(),
-    }
-    return payload
-
-
-class TLSAdapter(HTTPAdapter):
-    def init_poolmanager(self, connections, maxsize, block=False):
-        """Create and initialize the urllib3 PoolManager."""
-        ctx = ssl.create_default_context()
-        ctx.set_ciphers("DEFAULT@SECLEVEL=1")
-        self.poolmanager = PoolManager(
-            num_pools=connections,
-            maxsize=maxsize,
-            block=block,
-            ssl_version=ssl.PROTOCOL_TLSv1_2,
-            ssl_context=ctx,
-        )
-
-
-def load_signed_url(signed_url: str, dataset: Dataset, traces: list[SystemTrace]) -> None:
-    payload = createPayload(dataset, traces)
-    validate_json(payload)
-
-    session = requests.Session()
-    adapter = TLSAdapter()
-    session.mount("https://", adapter)
-
-    try:
-        response = session.put(
-            signed_url, json=payload, headers={"Content-Type": "application/json"}
-        )
-        response.raise_for_status()
-    except requests.exceptions.RequestException as e:
-        print(f"Error making request: {str(e)}")
-        print(f"Request payload: {payload}")  # Debugging info
-        raise
-
-    if response.status_code != 200:
-        raise ValueError(
-            f"Failed to load signed URL Status Code: {response.status_code} "
-            f"Response: {response.text}, Signed URL: {signed_url}"
-        )
-    else:
-        pass
-
-
-def send_system_traces_s3(
-    dataset: Dataset,
-    traces: list[SystemTrace],
-    base_url: str,
-    api_key: str,
-    system_id: str,
-    system_name: str,
-    verbose: bool = False,
-):
-    upload_id, signed_url = get_upload_id(base_url, api_key, system_id, system_name, verbose)
-    load_signed_url(signed_url, dataset, traces)
-
-    token_url = f"{base_url}/v1/auth/token"
-    try:
-        token_response = requests.get(token_url, headers={"customer_specific_api_key": api_key})
-        token_response.raise_for_status()
-        access_token = token_response.json()["access_token"]
-    except requests.exceptions.RequestException as e:
-        logging.error(f"Error obtaining access token: {e}")
-        raise
-
-    api_url = f"{base_url}/v1/uploads/process-upload/{upload_id}"
-    data = {"signed_url": signed_url}
-    headers = {
-        "Content-Type": "application/json",
-        "Authorization": f"Bearer {access_token}",
-    }
-
-    try:
-        response = requests.post(api_url, headers=headers, json=data)
-        response.raise_for_status()
-
-        response_data = response.json()
-        upload_id = response_data.get("upload_id")
-        signed_url = response_data.get("signed_url")
-        return upload_id, signed_url
-    except requests.exceptions.HTTPError as e:
-        logging.error(f"HTTP error occurred: {e}")
-        raise
-    except Exception as e:
-        logging.error(f"An error occurred: {e}")
-        raise
-
-
-def get_upload_id(
-    base_url: str, api_key: str, system_id: str, system_name: str, verbose: bool = False
-) -> tuple[str, str]:
-    """
-    Modified client-side function to send both system_id and system_name.
-    """
-    token_url = f"{base_url}/v1/auth/token"
-    token_response = requests.get(token_url, headers={"customer_specific_api_key": api_key})
-    token_response.raise_for_status()
-    access_token = token_response.json()["access_token"]
-
-    # Include system_name in the query parameters
-    api_url = (
-        f"{base_url}/v1/uploads/get-upload-id-signed-url?"
-        f"system_id={system_id}&system_name={system_name}"
-    )
-    headers = {
-        "Content-Type": "application/json",
-        "Authorization": f"Bearer {access_token}",
-    }
-
-    try:
-        response = requests.get(api_url, headers=headers)
-        response.raise_for_status()
-        upload_id = response.json()["upload_id"]
-        signed_url = response.json()["signed_url"]
-        return upload_id, signed_url
-    except requests.exceptions.HTTPError as e:
-        logging.error(f"HTTP error occurred: {e}")
-        raise
-    except Exception as e:
-        logging.error(f"An error occurred: {e}")
-        raise
-
-
-class UploadValidator(BaseModel):
-    """Validator for upload data format."""
-
-    model_config = ConfigDict(
-        from_attributes=True,  # Replaces the deprecated orm_mode
-        validate_assignment=True,
-        extra="forbid",  # Prevent additional fields
-    )
-
-    traces: list[dict[str, Any]]
-    dataset: dict[str, Any]
-
-    @field_validator("traces")
-    @classmethod
-    def validate_traces(cls, traces: list[dict[str, Any]]) -> list[dict[str, Any]]:
-        if not traces:
-            raise ValueError("Traces list cannot be empty")
-
-        for trace in traces:
-            # Validate required fields in each trace
-            if "system_instance_id" not in trace:
-                raise ValueError("Each trace must have a system_instance_id")
-            if "partition" not in trace:
-                raise ValueError("Each trace must have a partition")
-
-            # Validate metadata if present
-            if "metadata" in trace and trace["metadata"] is not None:
-                if not isinstance(trace["metadata"], dict):
-                    raise ValueError("Metadata must be a dictionary")
-
-            # Validate partition structure
-            partition = trace["partition"]
-            if not isinstance(partition, list):
-                raise ValueError("Partition must be a list")
-
-            for part in partition:
-                if "partition_index" not in part:
-                    raise ValueError("Each partition element must have a partition_index")
-                if "events" not in part:
-                    raise ValueError("Each partition element must have an events list")
-
-                # Validate events
-                events = part["events"]
-                if not isinstance(events, list):
-                    raise ValueError("Events must be a list")
-
-                for event in events:
-                    required_fields = [
-                        "event_type",
-                        "opened",
-                        "closed",
-                        "partition_index",
-                    ]
-                    missing_fields = [f for f in required_fields if f not in event]
-                    if missing_fields:
-                        raise ValueError(f"Event missing required fields: {missing_fields}")
-
-        return traces
-
-    @field_validator("dataset")
-    @classmethod
-    def validate_dataset(cls, dataset: dict[str, Any]) -> dict[str, Any]:
-        required_fields = ["questions", "reward_signals"]
-        missing_fields = [f for f in required_fields if f not in dataset]
-        if missing_fields:
-            raise ValueError(f"Dataset missing required fields: {missing_fields}")
-
-        # Validate questions
-        questions = dataset["questions"]
-        if not isinstance(questions, list):
-            raise ValueError("Questions must be a list")
-
-        for question in questions:
-            if "intent" not in question or "criteria" not in question:
-                raise ValueError("Each question must have intent and criteria")
-
-        # Validate reward signals
-        reward_signals = dataset["reward_signals"]
-        if not isinstance(reward_signals, list):
-            raise ValueError("Reward signals must be a list")
-
-        return dataset
-
-
-def validate_upload(traces: list[dict[str, Any]], dataset: dict[str, Any]):
-    # Validate the upload format before sending to server.
-    # Raises ValueError if validation fails.
-    try:
-        UploadValidator(traces=traces, dataset=dataset)
-        return True
-    except ValueError as e:
-        raise ValueError(f"Upload validation failed: {str(e)}")
-
-
-def is_event_loop_running():
-    try:
-        asyncio.get_running_loop()  # Check if there's a running event loop
-        return True
-    except RuntimeError:
-        # This exception is raised if no event loop is running
-        return False
-
-
-def format_upload_output(
-    dataset: Dataset, traces: list[SystemTrace]
-) -> tuple[list[dict[str, Any]], list[dict[str, Any]], list[dict[str, Any]]]:
-    # Format questions array
-    questions_data = [
-        {"intent": q.intent, "criteria": q.criteria, "id": q.id} for q in dataset.questions
-    ]
-
-    # Format reward signals array with error handling
-    reward_signals_data = [
-        {
-            "system_instance_id": rs.system_instance_id,
-            "reward": rs.reward,
-            "question_id": rs.question_id,
-            "annotation": rs.annotation if hasattr(rs, "annotation") else None,
-        }
-        for rs in dataset.reward_signals
-    ]
-
-    # Format traces array
-    traces_data = [
-        {
-            "system_instance_id": t.system_instance_id,
-            "metadata": t.metadata if t.metadata else None,
-            "partition": [
-                {
-                    "partition_index": p.partition_index,
-                    "events": [e.to_dict() for e in p.events],
-                }
-                for p in t.partition
-            ],
-        }
-        for t in traces
-    ]
-
-    return questions_data, reward_signals_data, traces_data
-
-
-class UploadIdResponse(TypedDict):
-    message: str
-    upload_id: str
-    signed_url: str
-
-
-class ProcessUploadResponse(TypedDict):
-    status: str
-    upload_id: str
-    signed_url: str
-
-
-def upload(
-    dataset: Dataset,
-    traces: list[SystemTrace] = [],
-    verbose: bool = False,
-    show_payload: bool = False,
-) -> tuple[
-    ProcessUploadResponse,
-    list[dict[str, Any]],
-    list[dict[str, Any]],
-    list[dict[str, Any]],
-]:
-    """Upload all system traces and dataset to the server.
-
-    DEPRECATED: Backend upload functionality is no longer supported in v1.
-
-    Args:
-        dataset: Dataset containing questions and reward signals
-        traces: List of system traces to upload
-        verbose: Whether to print verbose output
-        show_payload: Whether to show the payload being sent
-
-    Returns:
-        Tuple containing:
-        - response: Server response with status, upload_id, and signed_url
-        - questions_json: List of formatted questions
-        - reward_signals_json: List of formatted reward signals
-        - traces_json: List of formatted traces
-
-    Raises:
-        ValueError: If no system traces found or validation fails
-        requests.exceptions.HTTPError: If server request fails
-        RuntimeError: If SYNTH_API_KEY environment variable not set
-    """
-    _raise_not_supported_error()
-    return upload_helper(dataset, traces, verbose, show_payload)
-
-
-def upload_helper(
-    dataset: Dataset,
-    traces: list[SystemTrace] = [],
-    verbose: bool = False,
-    show_payload: bool = False,
-) -> tuple[
-    ProcessUploadResponse,
-    list[dict[str, Any]],
-    list[dict[str, Any]],
-    list[dict[str, Any]],
-]:
-    """Helper function to handle the upload process.
-
-    Returns same type as upload() function.
-    """
-    _raise_not_supported_error()
-    api_key = os.getenv("SYNTH_API_KEY")
-    if not api_key:
-        raise ValueError("SYNTH_API_KEY environment variable not set")
-    base_url = os.getenv("SYNTH_ENDPOINT_OVERRIDE", "https://agent-learning.onrender.com")
-
-    from .decorators import _local, active_events_var
-    from .trackers import synth_tracker_async
-
-    # First close any tracker events
-    if hasattr(synth_tracker_async, "active_events"):
-        for event_type, event in list(synth_tracker_async.active_events.items()):
-            if event and event.closed is None:
-                event.closed = time.time()
-                try:
-                    event_store.add_event(
-                        event.system_name,
-                        event.system_id,
-                        event.system_instance_id,
-                        event,
-                    )
-                    if verbose:
-                        print(f"Closed and stored tracker async event: {event_type}")
-                except Exception as e:
-                    logging.error(f"Failed to store tracker event {event_type}: {str(e)}")
-        synth_tracker_async.active_events.clear()
-
-    # End all active events before uploading
-    if hasattr(_local, "active_events"):
-        for event_type, event in _local.active_events.items():
-            if event and event.closed is None:
-                event.closed = time.time()
-                if hasattr(_local, "system_instance_id"):
-                    try:
-                        event_store.add_event(
-                            _local.system_name,
-                            _local.system_id,
-                            _local.system_instance_id,
-                            event,
-                        )
-                        if verbose:
-                            print(f"Closed and stored active event: {event_type}")
-                    except Exception as e:
-                        logging.error(f"Failed to store event {event_type}: {str(e)}")
-        _local.active_events.clear()
-
-    # NEW: Close all open asynchronous events
-    active_events_async = active_events_var.get()
-    if active_events_async:
-        current_time = time.time()
-        for event_type, event in list(active_events_async.items()):
-            if event and event.closed is None:
-                event.closed = current_time
-                try:
-                    event_store.add_event(
-                        event.system_name,
-                        event.system_id,
-                        event.system_instance_id,
-                        event,
-                    )
-                    if verbose:
-                        print(f"Closed and stored async event: {event_type}")
-                except Exception as e:
-                    logging.error(f"Failed to store async event {event_type}: {str(e)}")
-        active_events_var.set({})
-
-    # Also close any unclosed events in existing traces
-    logged_traces = event_store.get_system_traces()
-    traces = logged_traces + traces
-    # traces = event_store.get_system_traces() if len(traces) == 0 else traces
-    current_time = time.time()
-    for trace in traces:
-        for partition in trace.partition:
-            for event in partition.events:
-                if event.closed is None:
-                    event.closed = current_time
-                    event_store.add_event(
-                        trace.system_name,
-                        trace.system_id,
-                        trace.system_instance_id,
-                        event,
-                    )
-                    if verbose:
-                        print(f"Closed existing unclosed event: {event.event_type}")
-
-    try:
-        # Get traces and convert to dict format
-        if len(traces) == 0:
-            raise ValueError("No system traces found")
-        traces_dict = [trace.to_dict() for trace in traces]
-        dataset_dict = dataset.to_dict()
-
-        # Validate upload format
-        if verbose:
-            print("Validating upload format...")
-        validate_upload(traces_dict, dataset_dict)
-        if verbose:
-            print("Upload format validation successful")
-
-        # Send to server
-        upload_id, signed_url = send_system_traces_s3(
-            dataset=dataset,
-            traces=traces,
-            base_url=base_url,
-            api_key=api_key,
-            system_id=traces[0].system_id,
-            system_name=traces[0].system_name,
-            verbose=verbose,
-        )
-
-        questions_json, reward_signals_json, traces_json = format_upload_output(dataset, traces)
-        return (
-            {
-                "status": "success",
-                "upload_id": upload_id,
-                "signed_url": signed_url,
-            },
-            questions_json,
-            reward_signals_json,
-            traces_json,
-        )
-
-    except ValueError as e:
-        if verbose:
-            print("Validation error:", str(e))
-            print("\nTraces:")
-            print(json.dumps(traces_dict, indent=2))
-            print("\nDataset:")
-            print(json.dumps(dataset_dict, indent=2))
-        raise
-    except requests.exceptions.HTTPError as e:
-        if verbose:
-            print("HTTP error occurred:", e)
-            print("\nTraces:")
-            print(json.dumps(traces_dict, indent=2))
-            print("\nDataset:")
-            print(json.dumps(dataset_dict, indent=2))
-        raise
-# ruff: noqa

synth_ai/v0/tracing_v1/utils.py

@@ -1,9 +0,0 @@
-import uuid
-
-
-def get_system_id(system_name: str) -> str:
-    """Create a deterministic system_instance_id from system_name using UUID5."""
-    if not system_name:
-        raise ValueError("system_name cannot be empty")
-    system_id = uuid.uuid5(uuid.NAMESPACE_DNS, system_name)
-    return str(system_id)

synth_ai/zyk/__init__.py

@@ -1,30 +0,0 @@
-"""
-DEPRECATED: This module has been moved to synth_ai.lm
-
-The synth_ai.zyk module is deprecated and will be removed in a future version.
-Please update your imports:
-
-OLD: from synth_ai.zyk import LM
-NEW: from synth_ai.lm.core.main import LM
-
-or
-
-NEW: from synth_ai import LM  # (recommended)
-"""
-
-# ruff: noqa: E402
-import warnings
-
-# Issue deprecation warning
-warnings.warn(
-    "synth_ai.zyk is deprecated and will be removed in a future version. "
-    "Please use 'from synth_ai import LM' or 'from synth_ai.lm.core.main import LM' instead.",
-    DeprecationWarning,
-    stacklevel=2,
-)
-
-# Import from new location for backward compatibility
-from synth_ai.lm.core.main import LM
-from synth_ai.lm.vendors.base import BaseLMResponse
-
-__all__ = ["LM", "BaseLMResponse"]

synth_ai-0.2.6.dev1.dist-info/METADATA

@@ -1,106 +0,0 @@
-Metadata-Version: 2.4
-Name: synth-ai
-Version: 0.2.6.dev1
-Summary: RL as a service SDK - Core AI functionality and tracing
-Author-email: Synth AI <josh@usesynth.ai>
-License-Expression: MIT
-Project-URL: Homepage, https://github.com/synth-laboratories/synth-ai
-Project-URL: Repository, https://github.com/synth-laboratories/synth-ai
-Project-URL: Issues, https://github.com/synth-laboratories/synth-ai/issues
-Requires-Python: >=3.11
-Description-Content-Type: text/markdown
-License-File: LICENSE
-Requires-Dist: pydantic>=2.0.0
-Requires-Dist: python-dotenv>=1.0.1
-Requires-Dist: requests>=2.32.3
-Requires-Dist: urllib3>=2.3.0
-Requires-Dist: tqdm>=4.66.4
-Requires-Dist: jsonschema>=4.23.0
-Requires-Dist: backoff>=2.0.0
-Requires-Dist: typing_extensions>=4.0.0
-Requires-Dist: openai>=1.99.0
-Requires-Dist: anthropic>=0.42.0
-Requires-Dist: langfuse<3.0.0,>=2.53.9
-Requires-Dist: opentelemetry-api<1.27.0,>=1.26.0
-Requires-Dist: opentelemetry-sdk<1.27.0,>=1.26.0
-Requires-Dist: diskcache>=5.6.3
-Requires-Dist: groq>=0.30.0
-Requires-Dist: google-genai>=1.26.0
-Requires-Dist: together>=1.5.21
-Requires-Dist: mistralai>=1.9.2
-Requires-Dist: fastapi>=0.115.12
-Requires-Dist: uvicorn>=0.34.2
-Requires-Dist: numpy>=2.2.3
-Requires-Dist: networkx>=3.4.2
-Requires-Dist: redis>=6.2.0
-Requires-Dist: duckdb>=1.0.0
-Requires-Dist: pandas>=2.2.3
-Requires-Dist: ty>=0.0.1a5
-Requires-Dist: toml>=0.10.2
-Requires-Dist: sqlalchemy>=2.0.42
-Requires-Dist: aiosqlite>=0.21.0
-Requires-Dist: greenlet>=3.2.3
-Requires-Dist: libsql>=0.1.8
-Requires-Dist: pynacl>=1.5.0
-Requires-Dist: google-api-core>=2.25.1
-Requires-Dist: google-generativeai>=0.8.5
-Requires-Dist: crafter>=1.8.3
-Requires-Dist: click>=8.1.0
-Requires-Dist: textual>=1.1.0
-Requires-Dist: openai-harmony>=0.0.1
-Requires-Dist: asyncpg>=0.30.0
-Requires-Dist: aiohttp>=3.8.0
-Requires-Dist: datasets>=4.0.0
-Requires-Dist: transformers>=4.56.1
-Requires-Dist: modal>=1.1.4
-Provides-Extra: dev
-Requires-Dist: build>=1.2.2.post1; extra == "dev"
-Requires-Dist: twine>=4.0.0; extra == "dev"
-Requires-Dist: keyring>=24.0.0; extra == "dev"
-Requires-Dist: pytest>=8.3.3; extra == "dev"
-Requires-Dist: pytest-asyncio>=0.24.0; extra == "dev"
-Requires-Dist: pytest-cov>=4.1.0; extra == "dev"
-Requires-Dist: pyright>=1.1.350; extra == "dev"
-Requires-Dist: coverage[toml]>=7.3.0; extra == "dev"
-Requires-Dist: ruff>=0.1.0; extra == "dev"
-Provides-Extra: research
-Requires-Dist: crafter>=1.8.3; extra == "research"
-Requires-Dist: datasets>=4.0.0; extra == "research"
-Provides-Extra: all
-Requires-Dist: crafter>=1.8.3; extra == "all"
-Requires-Dist: datasets>=4.0.0; extra == "all"
-Dynamic: license-file
-
-# Synth-AI
-
-[![Python](https://img.shields.io/badge/python-3.11+-blue)](https://www.python.org/)
-[![License](https://img.shields.io/badge/license-MIT-green)](LICENSE)
-[![PyPI](https://img.shields.io/badge/PyPI-0.2.4.dev9-orange)](https://pypi.org/project/synth-ai/)
-![Coverage](https://img.shields.io/badge/coverage-0.0%25-red)
-![Tests](https://img.shields.io/badge/tests-17%2F17%20passing-brightgreen)
-
-Docs: [Synth‑AI Documentation](https://docs.usesynth.ai/synth-ai/introduction)
-
-Fast and effective reinforcement learning for agents, via an API
-
-## Highlights
-
-- Easily scale gpu topologies - train on 3 a10gs or 8 H100s (multi-node available upon request)
-- Requires only a thin fastapi wrapper to integrate with existing agent software.
-- Supports the best OSS models like Qwen3. (gpt-oss available upon request, GA soon)
-- Own your trained models
-
-## Getting Started
-
-synth-ai comes with a built-in RL example tailored for training Qwen/Qwen3-0.6B to succeed at Math.
-
-Please create an account at [Synth](https://usesynth.ai) and [Modal](https://modal.com) for the Math hello‑world test run. Then run:
-
-```bash
-uvx synth-ai rl_demo check
-uvx synth-ai rl_demo deploy
-uvx synth-ai rl_demo configure
-uvx synth-ai rl_demo run
-```
-
-To walk through kicking off your first RL run, see the [Synth‑AI Documentation](https://docs.usesynth.ai/synth-ai/introduction).