-
Notifications
You must be signed in to change notification settings - Fork 96
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Watsonx Prompt Templates Updated documentation and examples Retry mechanism update
- Loading branch information
Showing
56 changed files
with
1,285 additions
and
271 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,86 @@ | ||
| import asyncio | ||
| import logging | ||
| import os | ||
| import random | ||
|
|
||
| import httpx | ||
| from dotenv import load_dotenv | ||
|
|
||
| from genai.model import Credentials, Model | ||
| from genai.schemas import GenerateParams, ModelType, TokenParams | ||
| from genai.services.connection_manager import ConnectionManager | ||
| from genai.services.request_handler import RequestHandler | ||
|
|
||
| logging.basicConfig(level="ERROR") | ||
|
|
||
|
|
||
| class FlakyRequestHandler(RequestHandler): | ||
| @staticmethod | ||
| async def flaky_async_generate( | ||
| endpoint: str, key: str, model_id: str = None, inputs: list = None, parameters: dict = None, options=None | ||
| ): | ||
| """Low level API for async /generate request to REST API. | ||
| Args: | ||
| endpoint (str): Remote endpoint to be queried. | ||
| key (str): API key for authorization. | ||
| model_id (str, optional): The id of the language model to be queried. Defaults to None. | ||
| inputs (list, optional): List of inputs to be queried. Defaults to None. | ||
| parameters (dict, optional): Key-value pairs for model parameters. Defaults to None. | ||
| options (Options, optional): Additional parameters to pass in the query payload. Defaults to None. | ||
| Returns: | ||
| httpx.Response: Response from the REST API. | ||
| """ | ||
| headers, json_data = RequestHandler._metadata( | ||
| method="POST", | ||
| key=key, | ||
| model_id=model_id, | ||
| inputs=inputs, | ||
| parameters=parameters, | ||
| options=options, | ||
| ) | ||
|
|
||
| response = None | ||
| totalsleep = 0 | ||
| for attempt in range(0, ConnectionManager.MAX_RETRIES_GENERATE): | ||
| response = await ConnectionManager.async_generate_client.post(endpoint, headers=headers, json=json_data) | ||
| choice = random.randint(0, 2) | ||
| if choice == 0: | ||
| response.status_code = httpx.codes.SERVICE_UNAVAILABLE | ||
| elif choice == 1: | ||
| response.status_code = httpx.codes.TOO_MANY_REQUESTS | ||
| else: | ||
| pass | ||
| if response.status_code in [httpx.codes.SERVICE_UNAVAILABLE, httpx.codes.TOO_MANY_REQUESTS]: | ||
| totalsleep += 2 ** (attempt + 1) | ||
| print( | ||
| "Prompt = {}, Choice = {}, Current sleep {} seconds, Total sleep = {}".format( | ||
| json_data["inputs"], choice, 2 ** (attempt + 1), totalsleep | ||
| ) | ||
| ) | ||
| await asyncio.sleep(2 ** (attempt + 1)) | ||
| else: | ||
| break | ||
| return response | ||
|
|
||
|
|
||
| RequestHandler.async_generate = FlakyRequestHandler.flaky_async_generate | ||
|
|
||
| # make sure you have a .env file under genai root with | ||
| # GENAI_KEY=<your-genai-key> | ||
| load_dotenv() | ||
| api_key = os.getenv("GENAI_KEY", None) | ||
| api_url = os.getenv("GENAI_API", None) | ||
| creds = Credentials(api_key=api_key, api_endpoint=api_url) # credentials object to access GENAI | ||
|
|
||
|
|
||
| # Instantiate parameters for text generation | ||
| generate_params = GenerateParams(decoding_method="sample", max_new_tokens=5, min_new_tokens=1) | ||
| tokenize_params = TokenParams(return_tokens=True) | ||
|
|
||
|
|
||
| flan_ul2 = Model(ModelType.FLAN_UL2, params=generate_params, credentials=creds) | ||
| prompts = ["Generate a random number > {}: ".format(i) for i in range(25)] | ||
| for response in flan_ul2.generate_async(prompts, ordered=True): | ||
| pass |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.