OpenAI Online no key halts

[CharlieJCJ]

After #234, this happens

Minimal Reproduce

from bespokelabs.curator import Prompter
from datasets import Dataset
import logging

logger = logging.getLogger("bespokelabs.curator")
logger.setLevel(logging.DEBUG)

dataset = Dataset.from_dict({"prompt": ["write me a poem"] * 100})

prompter = Prompter(
    prompt_func=lambda row: row["prompt"],
    model_name="gpt-4o-mini",
    response_format=None,
)

dataset = prompter(dataset)
print(dataset.to_pandas())

without passing in a key, it halts...

(bespokelabs-curator-py3.12) (base) ➜  bella git:(dev) ✗ python examples/big.py 
2024-12-10 09:57:46,384 - bespokelabs.curator.prompter.prompter - INFO - Requesting text output from gpt-4o-mini, using OpenAI backend
2024-12-10 09:57:46,384 - bespokelabs.curator.request_processor.base_request_processor - DEBUG - Adjusting file descriptor limit from 1048576 to 1048576 (hard limit: 1048576)
2024-12-10 09:57:46,385 - bespokelabs.curator.prompter.prompter - DEBUG - Curator Cache Fingerprint String: 20127c6e62fd7298_e2caa4a2ecc553dc_gpt-4o-mini_text_False_openai
2024-12-10 09:57:46,385 - bespokelabs.curator.prompter.prompter - DEBUG - Curator Cache Fingerprint: 4d0c2e9c9ff55370
2024-12-10 09:57:46,389 - bespokelabs.curator.request_processor.base_online_request_processor - INFO - Running OpenAIOnlineRequestProcessor completions with model: gpt-4o-mini
2024-12-10 09:57:46,390 - bespokelabs.curator.request_processor.base_request_processor - INFO - Using cached requests. If you want to regenerate the dataset, disable or delete the cache.
2024-12-10 09:57:46,390 - bespokelabs.curator.request_processor.base_request_processor - DEBUG - There are 100 existing requests in /home/charlieji/.cache/curator/4d0c2e9c9ff55370/requests_0.jsonl
Example request in /home/charlieji/.cache/curator/4d0c2e9c9ff55370/requests_0.jsonl:
{
  "model": "gpt-4o-mini",
  "messages": [
    {
      "role": "user",
      "content": "write me a poem"
    }
  ],
  "response_format": null,
  "original_row": {
    "prompt": "write me a poem"
  },
  "original_row_idx": 0
}
2024-12-10 09:57:46,485 - bespokelabs.curator.request_processor.base_online_request_processor - WARNING - No manual max_tokens_per_minute set, and headers based detection failed, using default value of 1000
2024-12-10 09:57:46,486 - bespokelabs.curator.request_processor.base_online_request_processor - WARNING - No manual max_requests_per_minute set, and headers based detection failed, using default value of 10
Processing OpenAIOnlineRequestProcessor requests:   0%|                                                                                                              | 0/100 [00:00<?, ?it/s]

[CharlieJCJ]

The reason of halt is that after key isn't passed in, it uses default values that are too low.

The "too low" default values is resolved in #245

[CharlieJCJ]

The key not passed in, got a better raise exception in #237