edsl 0.1.36.dev1__tar.gz → 0.1.36.dev5__tar.gz

{edsl-0.1.36.dev1 → edsl-0.1.36.dev5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: edsl
-Version: 0.1.36.dev1
+Version: 0.1.36.dev5
 Summary: Create and analyze LLM-based surveys
 Home-page: https://www.expectedparrot.com/
 License: MIT

{edsl-0.1.36.dev1 → edsl-0.1.36.dev5}/edsl/Base.py

@@ -7,6 +7,8 @@ import json
 from typing import Any, Optional, Union
 from uuid import UUID
 
+# from edsl.utilities.MethodSuggesterMixin import MethodSuggesterMixin
+
 
 class RichPrintingMixin:
     """Mixin for rich printing and persistence of objects."""
@@ -274,6 +276,9 @@ class Base(
         """This method should be implemented by subclasses."""
         raise NotImplementedError("This method is not implemented yet.")
 
+    def to_json(self):
+        return json.dumps(self.to_dict())
+
     @abstractmethod
     def from_dict():
         """This method should be implemented by subclasses."""

edsl-0.1.36.dev5/edsl/__version__.py

@@ -0,0 +1 @@
+__version__ = "0.1.36.dev5"

{edsl-0.1.36.dev1 → edsl-0.1.36.dev5}/edsl/agents/Agent.py

@@ -111,7 +111,11 @@ class Agent(Base):
         self.name = name
         self._traits = traits or dict()
         self.codebook = codebook or dict()
-        self.instruction = instruction or self.default_instruction
+        if instruction is None:
+            self.instruction = self.default_instruction
+        else:
+            self.instruction = instruction
+        # self.instruction = instruction or self.default_instruction
         self.dynamic_traits_function = dynamic_traits_function
 
         # Deal with dynamic traits function

{edsl-0.1.36.dev1 → edsl-0.1.36.dev5}/edsl/agents/PromptConstructor.py

@@ -212,9 +212,10 @@ class PromptConstructor:
             )
 
             if relevant_instructions != []:
-                preamble_text = Prompt(
-                    text="Before answer this question, you were given the following instructions: "
-                )
+                # preamble_text = Prompt(
+                #    text="You were given the following instructions: "
+                # )
+                preamble_text = Prompt(text="")
                 for instruction in relevant_instructions:
                     preamble_text += instruction.text
                 rendered_instructions = preamble_text + rendered_instructions

{edsl-0.1.36.dev1 → edsl-0.1.36.dev5}/edsl/coop/PriceFetcher.py

@@ -16,30 +16,26 @@ class PriceFetcher:
         if self._cached_prices is not None:
             return self._cached_prices
 
+        import os
         import requests
-        import csv
-        from io import StringIO
-
-        sheet_id = "1SAO3Bhntefl0XQHJv27rMxpvu6uzKDWNXFHRa7jrUDs"
-
-        # Construct the URL to fetch the CSV
-        url = f"https://docs.google.com/spreadsheets/d/{sheet_id}/export?format=csv"
+        from edsl import CONFIG
 
         try:
-            # Fetch the CSV data
-            response = requests.get(url)
+            # Fetch the pricing data
+            url = f"{CONFIG.EXPECTED_PARROT_URL}/api/v0/prices"
+            api_key = os.getenv("EXPECTED_PARROT_API_KEY")
+            headers = {}
+            if api_key:
+                headers["Authorization"] = f"Bearer {api_key}"
+            else:
+                headers["Authorization"] = f"Bearer None"
+
+            response = requests.get(url, headers=headers, timeout=20)
             response.raise_for_status()  # Raise an exception for bad responses
 
-            # Parse the CSV data
-            csv_data = StringIO(response.text)
-            reader = csv.reader(csv_data)
-
-            # Convert to list of dictionaries
-            headers = next(reader)
-            data = [dict(zip(headers, row)) for row in reader]
+            # Parse the data
+            data = response.json()
 
-            # self._cached_prices = data
-            # return data
             price_lookup = {}
             for entry in data:
                 service = entry.get("service", None)

{edsl-0.1.36.dev1 → edsl-0.1.36.dev5}/edsl/coop/coop.py

@@ -6,6 +6,7 @@ from typing import Any, Optional, Union, Literal
 from uuid import UUID
 import edsl
 from edsl import CONFIG, CacheEntry, Jobs, Survey
+from edsl.exceptions.coop import CoopNoUUIDError, CoopServerResponseError
 from edsl.coop.utils import (
     EDSLObject,
     ObjectRegistry,
@@ -99,7 +100,7 @@ class Coop:
             if "Authorization" in message:
                 print(message)
                 message = "Please provide an Expected Parrot API key."
-            raise Exception(message)
+            raise CoopServerResponseError(message)
 
     def _json_handle_none(self, value: Any) -> Any:
         """
@@ -116,7 +117,7 @@ class Coop:
         Resolve the uuid from a uuid or a url.
         """
         if not url and not uuid:
-            raise Exception("No uuid or url provided for the object.")
+            raise CoopNoUUIDError("No uuid or url provided for the object.")
         if not uuid and url:
             uuid = url.split("/")[-1]
         return uuid
@@ -521,7 +522,7 @@ class Coop:
         self._resolve_server_response(response)
         response_json = response.json()
         return {
-            "uuid": response_json.get("jobs_uuid"),
+            "uuid": response_json.get("job_uuid"),
             "description": response_json.get("description"),
             "status": response_json.get("status"),
             "iterations": response_json.get("iterations"),
@@ -529,29 +530,41 @@ class Coop:
             "version": self._edsl_version,
         }
 
-    def remote_inference_get(self, job_uuid: str) -> dict:
+    def remote_inference_get(
+        self, job_uuid: Optional[str] = None, results_uuid: Optional[str] = None
+    ) -> dict:
         """
         Get the details of a remote inference job.
+        You can pass either the job uuid or the results uuid as a parameter.
+        If you pass both, the job uuid will be prioritized.
 
         :param job_uuid: The UUID of the EDSL job.
+        :param results_uuid: The UUID of the results associated with the EDSL job.
 
         >>> coop.remote_inference_get("9f8484ee-b407-40e4-9652-4133a7236c9c")
         {'jobs_uuid': '9f8484ee-b407-40e4-9652-4133a7236c9c', 'results_uuid': 'dd708234-31bf-4fe1-8747-6e232625e026', 'results_url': 'https://www.expectedparrot.com/content/dd708234-31bf-4fe1-8747-6e232625e026', 'status': 'completed', 'reason': None, 'price': 16, 'version': '0.1.29.dev4'}
         """
+        if job_uuid is None and results_uuid is None:
+            raise ValueError("Either job_uuid or results_uuid must be provided.")
+        elif job_uuid is not None:
+            params = {"job_uuid": job_uuid}
+        else:
+            params = {"results_uuid": results_uuid}
+
         response = self._send_server_request(
             uri="api/v0/remote-inference",
             method="GET",
-            params={"uuid": job_uuid},
+            params=params,
         )
         self._resolve_server_response(response)
         data = response.json()
         return {
-            "jobs_uuid": data.get("jobs_uuid"),
+            "job_uuid": data.get("job_uuid"),
             "results_uuid": data.get("results_uuid"),
             "results_url": f"{self.url}/content/{data.get('results_uuid')}",
             "status": data.get("status"),
             "reason": data.get("reason"),
-            "price": data.get("price"),
+            "credits_consumed": data.get("price"),
             "version": data.get("version"),
         }
 
@@ -584,7 +597,10 @@ class Coop:
         )
         self._resolve_server_response(response)
         response_json = response.json()
-        return response_json.get("cost")
+        return {
+            "credits": response_json.get("cost_in_credits"),
+            "usd": response_json.get("cost_in_usd"),
+        }
 
     ################
     # Remote Errors
@@ -649,6 +665,10 @@ class Coop:
         return response_json
 
     def fetch_prices(self) -> dict:
+        """
+        Fetch model prices from Coop. If the request fails, return an empty dict.
+        """
+
         from edsl.coop.PriceFetcher import PriceFetcher
 
         from edsl.config import CONFIG
@@ -659,6 +679,20 @@ class Coop:
         else:
             return {}
 
+    def fetch_rate_limit_config_vars(self) -> dict:
+        """
+        Fetch a dict of rate limit config vars from Coop.
+
+        The dict keys are RPM and TPM variables like EDSL_SERVICE_RPM_OPENAI.
+        """
+        response = self._send_server_request(
+            uri="api/v0/config-vars",
+            method="GET",
+        )
+        self._resolve_server_response(response)
+        data = response.json()
+        return data
+
 
 if __name__ == "__main__":
     sheet_data = fetch_sheet_data()

edsl-0.1.36.dev5/edsl/data/RemoteCacheSync.py

@@ -0,0 +1,84 @@
+class RemoteCacheSync:
+    def __init__(self, coop, cache, output_func, remote_cache=True, remote_cache_description=""):
+        self.coop = coop
+        self.cache = cache
+        self._output = output_func
+        self.remote_cache = remote_cache
+        self.old_entry_keys = []
+        self.new_cache_entries = []
+        self.remote_cache_description = remote_cache_description
+
+    def __enter__(self):
+        if self.remote_cache:
+            self._sync_from_remote()
+            self.old_entry_keys = list(self.cache.keys())
+        return self
+
+    def __exit__(self, exc_type, exc_value, traceback):
+        if self.remote_cache:
+            self._sync_to_remote()
+        return False  # Propagate exceptions
+
+    def _sync_from_remote(self):
+        cache_difference = self.coop.remote_cache_get_diff(self.cache.keys())
+        client_missing_cacheentries = cache_difference.get("client_missing_cacheentries", [])
+        missing_entry_count = len(client_missing_cacheentries)
+        
+        if missing_entry_count > 0:
+            self._output(
+                f"Updating local cache with {missing_entry_count:,} new "
+                f"{'entry' if missing_entry_count == 1 else 'entries'} from remote..."
+            )
+            self.cache.add_from_dict({entry.key: entry for entry in client_missing_cacheentries})
+            self._output("Local cache updated!")
+        else:
+            self._output("No new entries to add to local cache.")
+
+    def _sync_to_remote(self):
+        cache_difference = self.coop.remote_cache_get_diff(self.cache.keys())
+        server_missing_cacheentry_keys = cache_difference.get("server_missing_cacheentry_keys", [])
+        server_missing_cacheentries = [
+            entry
+            for key in server_missing_cacheentry_keys
+            if (entry := self.cache.data.get(key)) is not None
+        ]
+        
+        new_cache_entries = [
+            entry for entry in self.cache.values() if entry.key not in self.old_entry_keys
+        ]
+        server_missing_cacheentries.extend(new_cache_entries)
+        new_entry_count = len(server_missing_cacheentries)
+        
+        if new_entry_count > 0:
+            self._output(
+                f"Updating remote cache with {new_entry_count:,} new "
+                f"{'entry' if new_entry_count == 1 else 'entries'}..."
+            )
+            self.coop.remote_cache_create_many(
+                server_missing_cacheentries,
+                visibility="private",
+                description=self.remote_cache_description,
+            )
+            self._output("Remote cache updated!")
+        else:
+            self._output("No new entries to add to remote cache.")
+        
+        self._output(f"There are {len(self.cache.keys()):,} entries in the local cache.")
+
+# # Usage example
+# def run_job(self, n, progress_bar, cache, stop_on_exception, sidecar_model, print_exceptions, raise_validation_errors, use_remote_cache=True):
+#     with RemoteCacheSync(self.coop, cache, self._output, remote_cache=use_remote_cache):
+#         self._output("Running job...")
+#         results = self._run_local(
+#             n=n,
+#             progress_bar=progress_bar,
+#             cache=cache,
+#             stop_on_exception=stop_on_exception,
+#             sidecar_model=sidecar_model,
+#             print_exceptions=print_exceptions,
+#             raise_validation_errors=raise_validation_errors,
+#         )
+#         self._output("Job completed!")
+    
+#     results.cache = cache.new_entries_cache()
+#     return results

edsl-0.1.36.dev5/edsl/exceptions/coop.py

@@ -0,0 +1,10 @@
+class CoopErrors(Exception):
+    pass
+
+
+class CoopNoUUIDError(CoopErrors):
+    pass
+
+
+class CoopServerResponseError(CoopErrors):
+    pass

{edsl-0.1.36.dev1 → edsl-0.1.36.dev5}/edsl/inference_services/InferenceServiceABC.py

@@ -1,6 +1,7 @@
 from abc import abstractmethod, ABC
 import os
 import re
+from datetime import datetime, timedelta
 from edsl.config import CONFIG
 
 
@@ -10,6 +11,8 @@ class InferenceServiceABC(ABC):
     Anthropic: https://docs.anthropic.com/en/api/rate-limits
     """
 
+    _coop_config_vars = None
+
     default_levels = {
         "google": {"tpm": 2_000_000, "rpm": 15},
         "openai": {"tpm": 2_000_000, "rpm": 10_000},
@@ -31,12 +34,37 @@ class InferenceServiceABC(ABC):
                 f"Class {cls.__name__} must have a 'model_exclude_list' attribute."
             )
 
+    @classmethod
+    def _should_refresh_coop_config_vars(cls):
+        """
+        Returns True if config vars have been fetched over 24 hours ago, and False otherwise.
+        """
+
+        if cls._last_config_fetch is None:
+            return True
+        return (datetime.now() - cls._last_config_fetch) > timedelta(hours=24)
+
     @classmethod
     def _get_limt(cls, limit_type: str) -> int:
         key = f"EDSL_SERVICE_{limit_type.upper()}_{cls._inference_service_.upper()}"
         if key in os.environ:
             return int(os.getenv(key))
 
+        if cls._coop_config_vars is None or cls._should_refresh_coop_config_vars():
+            try:
+                from edsl import Coop
+
+                c = Coop()
+                cls._coop_config_vars = c.fetch_rate_limit_config_vars()
+                cls._last_config_fetch = datetime.now()
+                if key in cls._coop_config_vars:
+                    return cls._coop_config_vars[key]
+            except Exception:
+                cls._coop_config_vars = None
+        else:
+            if key in cls._coop_config_vars:
+                return cls._coop_config_vars[key]
+
         if cls._inference_service_ in cls.default_levels:
             return int(cls.default_levels[cls._inference_service_][limit_type])
 

{edsl-0.1.36.dev1 → edsl-0.1.36.dev5}/edsl/inference_services/registry.py

@@ -11,21 +11,29 @@ from edsl.inference_services.AwsBedrock import AwsBedrockService
 from edsl.inference_services.AzureAI import AzureAIService
 from edsl.inference_services.OllamaService import OllamaService
 from edsl.inference_services.TestService import TestService
-from edsl.inference_services.MistralAIService import MistralAIService
 from edsl.inference_services.TogetherAIService import TogetherAIService
 
-default = InferenceServicesCollection(
-    [
-        OpenAIService,
-        AnthropicService,
-        DeepInfraService,
-        GoogleService,
-        GroqService,
-        AwsBedrockService,
-        AzureAIService,
-        OllamaService,
-        TestService,
-        MistralAIService,
-        TogetherAIService,
-    ]
-)
+try:
+    from edsl.inference_services.MistralAIService import MistralAIService
+
+    mistral_available = True
+except Exception as e:
+    mistral_available = False
+
+services = [
+    OpenAIService,
+    AnthropicService,
+    DeepInfraService,
+    GoogleService,
+    GroqService,
+    AwsBedrockService,
+    AzureAIService,
+    OllamaService,
+    TestService,
+    TogetherAIService,
+]
+
+if mistral_available:
+    services.append(MistralAIService)
+
+default = InferenceServicesCollection(services)