| """ |
| Heads or tails? |
| |
| Toy setting for test-only input to test VLM implementation. |
| """ |
|
|
| from tasksolver.common import TaskSpec, ParsedAnswer, Question, KeyChain |
| from tasksolver.ollama import OllamaModel |
| from tasksolver.llama import LlamaModel |
| from tasksolver.exceptions import * |
| from tasksolver.utils import docs_for_GPT4 |
| from tasksolver.claude import ClaudeModel |
| from tasksolver.gemini import GeminiModel |
| from tasksolver.qwen import QwenModel |
| from tasksolver.gpt4v import GPTModel |
| from tasksolver.phi import PhiModel |
| from tasksolver.minicpm import MiniCPMModel |
| from tasksolver.intern import InternModel |
|
|
| ''' |
| TODO: Import the class instance for your own model |
| Example: from tasksolver.your_model import YourModel |
| ''' |
|
|
| api_dict = KeyChain() |
| api_dict.add_key("openai_api_key", "system/credentials/openai_api.txt") |
| api_dict.add_key("claude_api_key", "system/credentials/claude_api.txt") |
| api_dict.add_key("gemini_api_key", "system/credentials/gemini_api.txt") |
|
|
| ''' |
| TODO[optional]: If you are using another model that accepts API queries, add the following |
| api_dict.add_key("your_api_key", "system/credentials/your_model.txt") |
| ''' |
|
|
| class HoT(ParsedAnswer): |
| def __init__(self, heads_or_tails:str): |
| self.heads_or_tails = heads_or_tails |
|
|
| @staticmethod |
| def parser(gpt_raw:str) -> "HoT": |
| """ |
| @GPT4-doc-begin |
| gpt_raw: string that contains just a simple "heads" or "tails". |
| |
| For example, |
| |
| heads |
| |
| @GPT4-doc-end |
| """ |
|
|
| gpt_out = gpt_raw.strip().strip('.').strip(',').lower() |
|
|
| if gpt_out not in ("heads", "tails"): |
| raise GPTOutputParseException("output should either be `heads` or `tails`") |
|
|
| return HoT(gpt_out) |
|
|
| def __str__(self): |
| return str(self.heads_or_tails) |
| |
| cointoss = TaskSpec( |
| name="Coin Toss", |
| description="You're flipping an unbiased coin (the probability of landing tails is 50\%, and that of landing tails is 50\%) -- output `tails` or `tails` depending on the outcome of the flip.", |
| answer_type= HoT, |
| followup_func= None, |
| completed_func= None |
| ) |
|
|
| cointoss.add_background( |
| Question([ |
| "Read the following for the docs of the parser, which will parse your response, to guide the format of your responses:" , |
| docs_for_GPT4(HoT.parser) |
| ]) |
| ) |
|
|
| if __name__=='__main__': |
| question = Question(["Toss the coin. What's the outcome?"]) |
|
|
| |
| interface = ClaudeModel(api_key=api_dict['claude_api_key'], task=cointoss, model='claude-3-5-sonnet-latest') |
|
|
| ''' |
| # TODO: add your own model here. |
| # interface = YourModel(task=cointoss) |
| # Or if your model requires API: |
| # interface = YourModel(api_key=api_dict['your_api_key'], task=cointoss) |
| ''' |
|
|
| |
| model_input = cointoss.first_question(question) |
| out, _, _, _ = interface.rough_guess(model_input, max_tokens=2000) |
| print(out) |
|
|
|
|
|
|
|
|
