edsl 0.1.37__py3-none-any.whl → 0.1.37.dev1__py3-none-any.whl

edsl/inference_services/InferenceServicesCollection.py

@@ -16,48 +16,25 @@ class InferenceServicesCollection:
 
     @staticmethod
     def _get_service_available(service, warn: bool = False) -> list[str]:
+        from_api = True
         try:
             service_models = service.available()
-        except Exception:
+        except Exception as e:
             if warn:
                 warnings.warn(
                     f"""Error getting models for {service._inference_service_}. 
                     Check that you have properly stored your Expected Parrot API key and activated remote inference, or stored your own API keys for the language models that you want to use.
                     See https://docs.expectedparrot.com/en/latest/api_keys.html for instructions on storing API keys.
-                    Relying on Coop.""",
+                    Relying on cache.""",
                     UserWarning,
                 )
+            from edsl.inference_services.models_available_cache import models_available
 
-            # Use the list of models on Coop as a fallback
-            try:
-                from edsl import Coop
-
-                c = Coop()
-                models_from_coop = c.fetch_models()
-                service_models = models_from_coop.get(service._inference_service_, [])
-
-                # cache results
-                service._models_list_cache = service_models
-
-            # Finally, use the available models cache from the Python file
-            except Exception:
-                if warn:
-                    warnings.warn(
-                        f"""Error getting models for {service._inference_service_}. 
-                        Relying on EDSL cache.""",
-                        UserWarning,
-                    )
-
-                from edsl.inference_services.models_available_cache import (
-                    models_available,
-                )
-
-                service_models = models_available.get(service._inference_service_, [])
-
-                # cache results
-                service._models_list_cache = service_models
-
-        return service_models
+            service_models = models_available.get(service._inference_service_, [])
+            # cache results
+            service._models_list_cache = service_models
+            from_api = False
+        return service_models  # , from_api
 
     def available(self):
         total_models = []

edsl/jobs/Jobs.py

@@ -3,10 +3,9 @@ from __future__ import annotations
 import warnings
 import requests
 from itertools import product
-from typing import Literal, Optional, Union, Sequence, Generator
+from typing import Optional, Union, Sequence, Generator
 
 from edsl.Base import Base
-
 from edsl.exceptions import MissingAPIKeyError
 from edsl.jobs.buckets.BucketCollection import BucketCollection
 from edsl.jobs.interviews.Interview import Interview
@@ -194,7 +193,7 @@ class Jobs(Base):
                     inference_service=invigilator.model._inference_service_,
                     model=invigilator.model.model,
                 )
-                costs.append(prompt_cost["cost_usd"])
+                costs.append(prompt_cost["cost"])
 
         d = Dataset(
             [
@@ -210,14 +209,14 @@ class Jobs(Base):
         )
         return d
 
-    def show_prompts(self, all=False, max_rows: Optional[int] = None) -> None:
+    def show_prompts(self, all=False) -> None:
         """Print the prompts."""
         if all:
-            self.prompts().to_scenario_list().print(format="rich", max_rows=max_rows)
+            self.prompts().to_scenario_list().print(format="rich")
         else:
             self.prompts().select(
                 "user_prompt", "system_prompt"
-            ).to_scenario_list().print(format="rich", max_rows=max_rows)
+            ).to_scenario_list().print(format="rich")
 
     @staticmethod
     def estimate_prompt_cost(
@@ -228,7 +227,6 @@ class Jobs(Base):
         model: str,
     ) -> dict:
         """Estimates the cost of a prompt. Takes piping into account."""
-        import math
 
         def get_piping_multiplier(prompt: str):
             """Returns 2 if a prompt includes Jinja braces, and 1 otherwise."""
@@ -242,25 +240,10 @@ class Jobs(Base):
 
         try:
             relevant_prices = price_lookup[key]
-
-            service_input_token_price = float(
-                relevant_prices["input"]["service_stated_token_price"]
-            )
-            service_input_token_qty = float(
-                relevant_prices["input"]["service_stated_token_qty"]
-            )
-            input_price_per_token = service_input_token_price / service_input_token_qty
-
-            service_output_token_price = float(
-                relevant_prices["output"]["service_stated_token_price"]
-            )
-            service_output_token_qty = float(
-                relevant_prices["output"]["service_stated_token_qty"]
-            )
-            output_price_per_token = (
-                service_output_token_price / service_output_token_qty
+            output_price_per_token = 1 / float(
+                relevant_prices["output"]["one_usd_buys"]
             )
-
+            input_price_per_token = 1 / float(relevant_prices["input"]["one_usd_buys"])
         except KeyError:
             # A KeyError is likely to occur if we cannot retrieve prices (the price_lookup dict is empty)
             # Use a sensible default
@@ -270,8 +253,9 @@ class Jobs(Base):
             warnings.warn(
                 "Price data could not be retrieved. Using default estimates for input and output token prices. Input: $0.15 / 1M tokens; Output: $0.60 / 1M tokens"
             )
-            input_price_per_token = 0.00000015  # $0.15 / 1M tokens
-            output_price_per_token = 0.00000060  # $0.60 / 1M tokens
+
+            output_price_per_token = 0.00000015  # $0.15 / 1M tokens
+            input_price_per_token = 0.00000060  # $0.60 / 1M tokens
 
         # Compute the number of characters (double if the question involves piping)
         user_prompt_chars = len(str(user_prompt)) * get_piping_multiplier(
@@ -283,8 +267,7 @@ class Jobs(Base):
 
         # Convert into tokens (1 token approx. equals 4 characters)
         input_tokens = (user_prompt_chars + system_prompt_chars) // 4
-
-        output_tokens = math.ceil(0.75 * input_tokens)
+        output_tokens = input_tokens
 
         cost = (
             input_tokens * input_price_per_token
@@ -294,17 +277,15 @@ class Jobs(Base):
         return {
             "input_tokens": input_tokens,
             "output_tokens": output_tokens,
-            "cost_usd": cost,
+            "cost": cost,
         }
 
-    def estimate_job_cost_from_external_prices(
-        self, price_lookup: dict, iterations: int = 1
-    ) -> dict:
+    def estimate_job_cost_from_external_prices(self, price_lookup: dict) -> dict:
         """
         Estimates the cost of a job according to the following assumptions:
 
         - 1 token = 4 characters.
-        - For each prompt, output tokens = input tokens * 0.75, rounded up to the nearest integer.
+        - Input tokens = output tokens.
 
         price_lookup is an external pricing dictionary.
         """
@@ -341,7 +322,7 @@ class Jobs(Base):
                         "system_prompt": system_prompt,
                         "estimated_input_tokens": prompt_cost["input_tokens"],
                         "estimated_output_tokens": prompt_cost["output_tokens"],
-                        "estimated_cost_usd": prompt_cost["cost_usd"],
+                        "estimated_cost": prompt_cost["cost"],
                         "inference_service": inference_service,
                         "model": model,
                     }
@@ -353,21 +334,18 @@ class Jobs(Base):
             df.groupby(["inference_service", "model"])
             .agg(
                 {
-                    "estimated_cost_usd": "sum",
+                    "estimated_cost": "sum",
                     "estimated_input_tokens": "sum",
                     "estimated_output_tokens": "sum",
                 }
             )
             .reset_index()
         )
-        df["estimated_cost_usd"] = df["estimated_cost_usd"] * iterations
-        df["estimated_input_tokens"] = df["estimated_input_tokens"] * iterations
-        df["estimated_output_tokens"] = df["estimated_output_tokens"] * iterations
 
         estimated_costs_by_model = df.to_dict("records")
 
         estimated_total_cost = sum(
-            model["estimated_cost_usd"] for model in estimated_costs_by_model
+            model["estimated_cost"] for model in estimated_costs_by_model
         )
         estimated_total_input_tokens = sum(
             model["estimated_input_tokens"] for model in estimated_costs_by_model
@@ -377,7 +355,7 @@ class Jobs(Base):
         )
 
         output = {
-            "estimated_total_cost_usd": estimated_total_cost,
+            "estimated_total_cost": estimated_total_cost,
             "estimated_total_input_tokens": estimated_total_input_tokens,
             "estimated_total_output_tokens": estimated_total_output_tokens,
             "model_costs": estimated_costs_by_model,
@@ -385,12 +363,12 @@ class Jobs(Base):
 
         return output
 
-    def estimate_job_cost(self, iterations: int = 1) -> dict:
+    def estimate_job_cost(self) -> dict:
         """
         Estimates the cost of a job according to the following assumptions:
 
         - 1 token = 4 characters.
-        - For each prompt, output tokens = input tokens * 0.75, rounded up to the nearest integer.
+        - Input tokens = output tokens.
 
         Fetches prices from Coop.
         """
@@ -399,9 +377,7 @@ class Jobs(Base):
         c = Coop()
         price_lookup = c.fetch_prices()
 
-        return self.estimate_job_cost_from_external_prices(
-            price_lookup=price_lookup, iterations=iterations
-        )
+        return self.estimate_job_cost_from_external_prices(price_lookup=price_lookup)
 
     @staticmethod
     def compute_job_cost(job_results: "Results") -> float:
@@ -723,11 +699,7 @@ class Jobs(Base):
         return self._raise_validation_errors
 
     def create_remote_inference_job(
-        self,
-        iterations: int = 1,
-        remote_inference_description: Optional[str] = None,
-        remote_inference_results_visibility: Optional[VisibilityType] = "unlisted",
-        verbose=False,
+        self, iterations: int = 1, remote_inference_description: Optional[str] = None
     ):
         """ """
         from edsl.coop.coop import Coop
@@ -739,11 +711,9 @@ class Jobs(Base):
             description=remote_inference_description,
             status="queued",
             iterations=iterations,
-            initial_results_visibility=remote_inference_results_visibility,
         )
         job_uuid = remote_job_creation_data.get("uuid")
-        if self.verbose:
-            print(f"Job sent to server. (Job uuid={job_uuid}).")
+        print(f"Job sent to server. (Job uuid={job_uuid}).")
         return remote_job_creation_data
 
     @staticmethod
@@ -754,7 +724,7 @@ class Jobs(Base):
         return coop.remote_inference_get(job_uuid)
 
     def poll_remote_inference_job(
-        self, remote_job_creation_data: dict, verbose=False, poll_interval=5
+        self, remote_job_creation_data: dict
     ) -> Union[Results, None]:
         from edsl.coop.coop import Coop
         import time
@@ -771,46 +741,42 @@ class Jobs(Base):
             remote_job_data = coop.remote_inference_get(job_uuid)
             status = remote_job_data.get("status")
             if status == "cancelled":
-                if self.verbose:
-                    print("\r" + " " * 80 + "\r", end="")
-                    print("Job cancelled by the user.")
-                    print(
-                        f"See {expected_parrot_url}/home/remote-inference for more details."
-                    )
+                print("\r" + " " * 80 + "\r", end="")
+                print("Job cancelled by the user.")
+                print(
+                    f"See {expected_parrot_url}/home/remote-inference for more details."
+                )
                 return None
             elif status == "failed":
-                if self.verbose:
-                    print("\r" + " " * 80 + "\r", end="")
-                    print("Job failed.")
-                    print(
-                        f"See {expected_parrot_url}/home/remote-inference for more details."
-                    )
+                print("\r" + " " * 80 + "\r", end="")
+                print("Job failed.")
+                print(
+                    f"See {expected_parrot_url}/home/remote-inference for more details."
+                )
                 return None
             elif status == "completed":
                 results_uuid = remote_job_data.get("results_uuid")
                 results = coop.get(results_uuid, expected_object_type="results")
-                if self.verbose:
-                    print("\r" + " " * 80 + "\r", end="")
-                    url = f"{expected_parrot_url}/content/{results_uuid}"
-                    print(f"Job completed and Results stored on Coop: {url}.")
+                print("\r" + " " * 80 + "\r", end="")
+                url = f"{expected_parrot_url}/content/{results_uuid}"
+                print(f"Job completed and Results stored on Coop: {url}.")
                 return results
             else:
-                duration = poll_interval
+                duration = 5
                 time_checked = datetime.now().strftime("%Y-%m-%d %I:%M:%S %p")
                 frames = ["⠋", "⠙", "⠹", "⠸", "⠼", "⠴", "⠦", "⠧", "⠇", "⠏"]
                 start_time = time.time()
                 i = 0
                 while time.time() - start_time < duration:
-                    if self.verbose:
-                        print(
-                            f"\r{frames[i % len(frames)]} Job status: {status} - last update: {time_checked}",
-                            end="",
-                            flush=True,
-                        )
+                    print(
+                        f"\r{frames[i % len(frames)]} Job status: {status} - last update: {time_checked}",
+                        end="",
+                        flush=True,
+                    )
                     time.sleep(0.1)
                     i += 1
 
-    def use_remote_inference(self, disable_remote_inference: bool) -> bool:
+    def use_remote_inference(self, disable_remote_inference: bool):
         if disable_remote_inference:
             return False
         if not disable_remote_inference:
@@ -826,23 +792,20 @@ class Jobs(Base):
 
         return False
 
-    def use_remote_cache(self, disable_remote_cache: bool) -> bool:
-        if disable_remote_cache:
-            return False
-        if not disable_remote_cache:
-            try:
-                from edsl import Coop
+    def use_remote_cache(self):
+        try:
+            from edsl import Coop
 
-                user_edsl_settings = Coop().edsl_settings
-                return user_edsl_settings.get("remote_caching", False)
-            except requests.ConnectionError:
-                pass
-            except CoopServerResponseError as e:
-                pass
+            user_edsl_settings = Coop().edsl_settings
+            return user_edsl_settings.get("remote_caching", False)
+        except requests.ConnectionError:
+            pass
+        except CoopServerResponseError as e:
+            pass
 
         return False
 
-    def check_api_keys(self) -> None:
+    def check_api_keys(self):
         from edsl import Model
 
         for model in self.models + [Model()]:
@@ -852,86 +815,6 @@ class Jobs(Base):
                     inference_service=model._inference_service_,
                 )
 
-    def get_missing_api_keys(self) -> set:
-        """
-        Returns a list of the api keys that a user needs to run this job, but does not currently have in their .env file.
-        """
-
-        missing_api_keys = set()
-
-        from edsl import Model
-        from edsl.enums import service_to_api_keyname
-
-        for model in self.models + [Model()]:
-            if not model.has_valid_api_key():
-                key_name = service_to_api_keyname.get(
-                    model._inference_service_, "NOT FOUND"
-                )
-                missing_api_keys.add(key_name)
-
-        return missing_api_keys
-
-    def user_has_all_model_keys(self):
-        """
-        Returns True if the user has all model keys required to run their job.
-
-        Otherwise, returns False.
-        """
-
-        try:
-            self.check_api_keys()
-            return True
-        except MissingAPIKeyError:
-            return False
-        except Exception:
-            raise
-
-    def user_has_ep_api_key(self) -> bool:
-        """
-        Returns True if the user has an EXPECTED_PARROT_API_KEY in their env.
-
-        Otherwise, returns False.
-        """
-
-        import os
-
-        coop_api_key = os.getenv("EXPECTED_PARROT_API_KEY")
-
-        if coop_api_key is not None:
-            return True
-        else:
-            return False
-
-    def needs_external_llms(self) -> bool:
-        """
-        Returns True if the job needs external LLMs to run.
-
-        Otherwise, returns False.
-        """
-        # These cases are necessary to skip the API key check during doctests
-
-        # Accounts for Results.example()
-        all_agents_answer_questions_directly = len(self.agents) > 0 and all(
-            [hasattr(a, "answer_question_directly") for a in self.agents]
-        )
-
-        # Accounts for InterviewExceptionEntry.example()
-        only_model_is_test = set([m.model for m in self.models]) == set(["test"])
-
-        # Accounts for Survey.__call__
-        all_questions_are_functional = set(
-            [q.question_type for q in self.survey.questions]
-        ) == set(["functional"])
-
-        if (
-            all_agents_answer_questions_directly
-            or only_model_is_test
-            or all_questions_are_functional
-        ):
-            return False
-        else:
-            return True
-
     def run(
         self,
         n: int = 1,
@@ -944,28 +827,22 @@ class Jobs(Base):
         print_exceptions=True,
         remote_cache_description: Optional[str] = None,
         remote_inference_description: Optional[str] = None,
-        remote_inference_results_visibility: Optional[
-            Literal["private", "public", "unlisted"]
-        ] = "unlisted",
         skip_retry: bool = False,
         raise_validation_errors: bool = False,
-        disable_remote_cache: bool = False,
         disable_remote_inference: bool = False,
     ) -> Results:
         """
         Runs the Job: conducts Interviews and returns their results.
 
-        :param n: How many times to run each interview
-        :param progress_bar: Whether to show a progress bar
-        :param stop_on_exception: Stops the job if an exception is raised
-        :param cache: A Cache object to store results
-        :param check_api_keys: Raises an error if API keys are invalid
-        :param verbose: Prints extra messages
-        :param remote_cache_description: Specifies a description for this group of entries in the remote cache
-        :param remote_inference_description: Specifies a description for the remote inference job
-        :param remote_inference_results_visibility: The initial visibility of the Results object on Coop. This will only be used for remote jobs!
-        :param disable_remote_cache: If True, the job will not use remote cache. This only works for local jobs!
-        :param disable_remote_inference: If True, the job will not use remote inference
+        :param n: how many times to run each interview
+        :param progress_bar: shows a progress bar
+        :param stop_on_exception: stops the job if an exception is raised
+        :param cache: a cache object to store results
+        :param check_api_keys: check if the API keys are valid
+        :param batch_mode: run the job in batch mode i.e., no expecation of interaction with the user
+        :param verbose: prints messages
+        :param remote_cache_description: specifies a description for this group of entries in the remote cache
+        :param remote_inference_description: specifies a description for the remote inference job
         """
         from edsl.coop.coop import Coop
 
@@ -975,54 +852,9 @@ class Jobs(Base):
 
         self.verbose = verbose
 
-        if (
-            not self.user_has_all_model_keys()
-            and not self.user_has_ep_api_key()
-            and self.needs_external_llms()
-        ):
-            import secrets
-            from dotenv import load_dotenv
-            from edsl import CONFIG
-            from edsl.coop.coop import Coop
-            from edsl.utilities.utilities import write_api_key_to_env
-
-            missing_api_keys = self.get_missing_api_keys()
-
-            edsl_auth_token = secrets.token_urlsafe(16)
-
-            print("You're missing some of the API keys needed to run this job:")
-            for api_key in missing_api_keys:
-                print(f"     🔑 {api_key}")
-            print(
-                "\nYou can either add the missing keys to your .env file, or use remote inference."
-            )
-            print("Remote inference allows you to run jobs on our server.")
-            print("\n🚀 To use remote inference, sign up at the following link:")
-
-            coop = Coop()
-            coop._display_login_url(edsl_auth_token=edsl_auth_token)
-
-            print(
-                "\nOnce you log in, we will automatically retrieve your Expected Parrot API key and continue your job remotely."
-            )
-
-            api_key = coop._poll_for_api_key(edsl_auth_token)
-
-            if api_key is None:
-                print("\nTimed out waiting for login. Please try again.")
-                return
-
-            write_api_key_to_env(api_key)
-            print("✨ API key retrieved and written to .env file.\n")
-
-            # Retrieve API key so we can continue running the job
-            load_dotenv()
-
         if remote_inference := self.use_remote_inference(disable_remote_inference):
             remote_job_creation_data = self.create_remote_inference_job(
-                iterations=n,
-                remote_inference_description=remote_inference_description,
-                remote_inference_results_visibility=remote_inference_results_visibility,
+                iterations=n, remote_inference_description=remote_inference_description
             )
             results = self.poll_remote_inference_job(remote_job_creation_data)
             if results is None:
@@ -1042,7 +874,7 @@ class Jobs(Base):
 
             cache = Cache()
 
-        remote_cache = self.use_remote_cache(disable_remote_cache)
+        remote_cache = self.use_remote_cache()
         with RemoteCacheSync(
             coop=Coop(),
             cache=cache,
@@ -1063,84 +895,17 @@ class Jobs(Base):
         results.cache = cache.new_entries_cache()
         return results
 
-    async def create_and_poll_remote_job(
-        self,
-        iterations: int = 1,
-        remote_inference_description: Optional[str] = None,
-        remote_inference_results_visibility: Optional[
-            Literal["private", "public", "unlisted"]
-        ] = "unlisted",
-    ) -> Union[Results, None]:
-        """
-        Creates and polls a remote inference job asynchronously.
-        Reuses existing synchronous methods but runs them in an async context.
-
-        :param iterations: Number of times to run each interview
-        :param remote_inference_description: Optional description for the remote job
-        :param remote_inference_results_visibility: Visibility setting for results
-        :return: Results object if successful, None if job fails or is cancelled
-        """
-        import asyncio
-        from functools import partial
-
-        # Create job using existing method
-        loop = asyncio.get_event_loop()
-        remote_job_creation_data = await loop.run_in_executor(
-            None,
-            partial(
-                self.create_remote_inference_job,
-                iterations=iterations,
-                remote_inference_description=remote_inference_description,
-                remote_inference_results_visibility=remote_inference_results_visibility,
-            ),
-        )
-
-        # Poll using existing method but with async sleep
-        return await loop.run_in_executor(
-            None, partial(self.poll_remote_inference_job, remote_job_creation_data)
-        )
-
-    async def run_async(
-        self,
-        cache=None,
-        n=1,
-        disable_remote_inference: bool = False,
-        remote_inference_description: Optional[str] = None,
-        remote_inference_results_visibility: Optional[
-            Literal["private", "public", "unlisted"]
-        ] = "unlisted",
-        **kwargs,
-    ):
-        """Run the job asynchronously, either locally or remotely.
-
-        :param cache: Cache object or boolean
-        :param n: Number of iterations
-        :param disable_remote_inference: If True, forces local execution
-        :param remote_inference_description: Description for remote jobs
-        :param remote_inference_results_visibility: Visibility setting for remote results
-        :param kwargs: Additional arguments passed to local execution
-        :return: Results object
-        """
-        # Check if we should use remote inference
-        if remote_inference := self.use_remote_inference(disable_remote_inference):
-            results = await self.create_and_poll_remote_job(
-                iterations=n,
-                remote_inference_description=remote_inference_description,
-                remote_inference_results_visibility=remote_inference_results_visibility,
-            )
-            if results is None:
-                self._output("Job failed.")
-            return results
-
-        # If not using remote inference, run locally with async
-        return await JobsRunnerAsyncio(self).run_async(cache=cache, n=n, **kwargs)
-
     def _run_local(self, *args, **kwargs):
         """Run the job locally."""
 
         results = JobsRunnerAsyncio(self).run(*args, **kwargs)
         return results
 
+    async def run_async(self, cache=None, n=1, **kwargs):
+        """Run asynchronously."""
+        results = await JobsRunnerAsyncio(self).run_async(cache=cache, n=n, **kwargs)
+        return results
+
     def all_question_parameters(self):
         """Return all the fields in the questions in the survey.
         >>> from edsl.jobs import Jobs

edsl/jobs/interviews/InterviewExceptionEntry.py

@@ -67,11 +67,7 @@ class InterviewExceptionEntry:
         m = LanguageModel.example(test_model=True)
         q = QuestionFreeText.example(exception_to_throw=ValueError)
         results = q.by(m).run(
-            skip_retry=True,
-            print_exceptions=False,
-            raise_validation_errors=True,
-            disable_remote_cache=True,
-            disable_remote_inference=True,
+            skip_retry=True, print_exceptions=False, raise_validation_errors=True
         )
         return results.task_history.exceptions[0]["how_are_you"][0]
 

edsl/jobs/tasks/TaskHistory.py

@@ -39,7 +39,6 @@ class TaskHistory:
             skip_retry=True,
             cache=False,
             raise_validation_errors=True,
-            disable_remote_cache=True,
             disable_remote_inference=True,
         )