Fixed error in type hints

[Souryadipstan]

Description

I encountered this error while trying to import OpenAIInstructionParser, TaskType. Then, I solved that issue and further checked that the same issue is present in other files too, so solved them. also, pre-commit was throwing FI58 error while merging the new changes to added a parameter in pre-commit config to ignore the particular error.

TypeError Traceback (most recent call last)
Cell In[1], line 1
----> 1 from prompt2model.prompt_parser import OpenAIInstructionParser, TaskType
3 prompt_spec = OpenAIInstructionParser(task_type=TaskType.TEXT_GENERATION)
4 prompt_spec.parse_from_prompt(prompt)

File ~/.pyenv/versions/3.8.17/lib/python3.8/site-packages/prompt2model/prompt_parser/init.py:3
1 """Import PromptSpec classes."""
2 from prompt2model.prompt_parser.base import PromptSpec, TaskType
----> 3 from prompt2model.prompt_parser.instr_parser import OpenAIInstructionParser
4 from prompt2model.prompt_parser.mock import MockPromptSpec
6 all = (
7 "PromptSpec",
8 "TaskType",
9 "MockPromptSpec",
10 "OpenAIInstructionParser",
11 )

File ~/.pyenv/versions/3.8.17/lib/python3.8/site-packages/prompt2model/prompt_parser/instr_parser.py:15
10 from prompt2model.prompt_parser.base import PromptSpec, TaskType
12 from prompt2model.prompt_parser.instr_parser_prompt import ( # isort: split
13 construct_prompt_for_instruction_parsing,
14 )
---> 15 from prompt2model.utils import (
16 OPENAI_ERRORS,
17 ChatGPTAgent,
18 get_formatted_logger,
19 handle_openai_error,
20 )
22 logger = get_formatted_logger("PromptParser")
24 os.environ["TOKENIZERS_PARALLELISM"] = "false"

File ~/.pyenv/versions/3.8.17/lib/python3.8/site-packages/prompt2model/utils/init.py:10
3 from prompt2model.utils.openai_tools import (
4 OPENAI_ERRORS,
5 ChatGPTAgent,
6 count_tokens_from_string,
7 handle_openai_error,
8 )
9 from prompt2model.utils.rng import seed_generator
---> 10 from prompt2model.utils.tevatron_utils import encode_text, retrieve_objects
12 all = ( # noqa: F401
13 "ChatGPTAgent",
14 "encode_text",
(...)
20 "get_formatted_logger",
21 )

File ~/.pyenv/versions/3.8.17/lib/python3.8/site-packages/prompt2model/utils/tevatron_utils/init.py:3
1 """Import Tevatron utility functions."""
2 from prompt2model.utils.tevatron_utils.encode import encode_text
----> 3 from prompt2model.utils.tevatron_utils.retrieve import retrieve_objects
5 all = ("encode_text", "retrieve_objects")

File ~/.pyenv/versions/3.8.17/lib/python3.8/site-packages/prompt2model/utils/tevatron_utils/retrieve.py:11
4 import numpy as np
5 from tevatron.faiss_retriever import BaseFaissIPRetriever
8 def retrieve_objects(
9 query_vector: np.ndarray,
10 encoded_datasets_path: str,
---> 11 document_names: list[str],
12 depth: int,
13 ) -> list[tuple[str, float]]:
14 """Return a ranked list of object indices and their scores.
15
16 Args:
(...)
22 Ranked list of object names and their inner product similarity to the query.
23 """
24 assert query_vector.shape[0] == 1, "Only a single query vector is expected."

TypeError: 'type' object is not subscriptable

References

Below mentioned issue is solved by this PR

• TypeError: 'type' object is not subscriptable #325

[neubig]

Looks great, and also fixes #323