Spaces:
Build error
Build error
File size: 2,770 Bytes
7e3e85d |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 |
from typing import List, Union
class SummModel:
"""
Base model class for SummerTime
"""
# static variables
model_name = "None"
is_extractive = False
is_neural = False
is_query_based = False
is_dialogue_based = False
is_multi_document = False
def __init__(
self,
trained_domain: str = None,
max_input_length: int = None,
max_output_length: int = None,
):
self.trained_domain = trained_domain
self.max_input_length = max_input_length
self.max_output_length = max_output_length
def summarize(
self, corpus: Union[List[str], List[List[str]]], queries: List[str] = None
) -> List[str]:
"""
All summarization models should have this function
:param corpus: each string in the list is a source document to be summarized; if the model is multi-document or
dialogue summarization model, then each instance contains a list of documents/utterances
:param queries: a list of queries if this is a query-based model
:return: a list of generated summaries
"""
raise NotImplementedError(
"The base class for models shouldn't be instantiated!"
)
@classmethod
def assert_summ_input_type(
cls, corpus: Union[List[str], List[List[str]]], queries: Union[List[str], None]
):
"""
Verifies that type of input corpus or queries for summarization align with the model type.
"""
raise NotImplementedError(
"The base class for models shouldn't be instantiated!"
)
@classmethod
def show_capability(cls) -> None:
"""
Use concise language to show the strength and weakness for each model. Try not to use NLP terminologies
"""
raise NotImplementedError(
"The base class for models shouldn't be instantiated!"
)
@classmethod
def generate_basic_description(cls) -> str:
"""
Automatically generate the basic description string based on the attributes
"""
extractive_abstractive = "extractive" if cls.is_extractive else "abstractive"
neural = "neural" if cls.is_neural else "non-neural"
basic_description = (
f"{cls.model_name} is a"
f"{'query-based' if cls.is_query_based else ''} "
f"{extractive_abstractive}, {neural} model for summarization."
)
if cls.is_multi_document or cls.is_dialogue_based:
basic_description += (
f"It can handle {'multi-document' if cls.is_multi_document else ''} "
f"{'dialogue' if cls.is_dialogue_based else ''} textual data."
)
return basic_description
|