Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Convert pydantic objects to regular classes #410

Merged
merged 3 commits into from
Jun 17, 2020
Merged

Convert pydantic objects to regular classes #410

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
7a98b8f
Convert pydantic objects to regular classes
brandenchan Jun 17, 2020
cb436ce
Bugfix squad id assignment
Timoeller Jun 17, 2020
e78aeca
Adjust fields in QA json output
Timoeller Jun 17, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
143 changes: 89 additions & 54 deletions farm/modeling/predictions.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,57 +3,79 @@
from typing import List, Optional, Any
from pydantic import BaseModel

class Pred(BaseModel):
class Pred:
"""
Base class for predictions of every task. Note that it inherits from pydantic.BaseModel which creates an
__init__() with the attributes defined in this class (i.e. id, prediction, context)
"""
id: str
prediction: List[Any]
context: str

def __init__(self,
id: str,
prediction: List[Any],
context: str):
self.id = id
self.prediction = prediction
self.context = context

def to_json(self):
raise NotImplementedError

class QACandidate(BaseModel):
class QACandidate:
"""
A single QA candidate answer. Note that it inherits from pydantic.BaseModel which builds the __init__() method.
See class definition to find list of compulsory and optional arguments and also comments on how they are used.
"""

# self.answer_type can be "is_impossible", "yes", "no" or "span"
answer_type: str
score: float
probability: Optional[float] = None

# If self.answer_type is "span", self.answer is a string answer span
# Otherwise, it is None
answer: Optional[str] = None
offset_answer_start: int
offset_answer_end: int

# If self.answer_type is in ["yes", "no"] then self.answer_support is a text string
# If self.answer is a string answer span or self.answer_type is "is_impossible", answer_support is None
# TODO sample_idx can probably be removed since we have passage_id
answer_support: Optional[str] = None
offset_answer_support_start: Optional[int] = None
offset_answer_support_end: Optional[int] = None
sample_idx: Optional[int] = None

# self.context is the document or passage where the answer is found
context: Optional[str] = None
offset_context_start: Optional[int] = None
offset_context_end: Optional[int] = None

# Offset unit is either "token" or "char"
# Aggregation level is either "doc" or "passage"
offset_unit: str
aggregation_level: str

n_samples_in_doc: Optional[int] = None
document_id: Optional[str] = None
passage_id: Optional[str] = None
def __init__(self,
answer_type: str,
score: str,
offset_answer_start: int,
offset_answer_end: int,
offset_unit: str,
aggregation_level: str,
probability: float=None,
answer: str=None,
answer_support: str=None,
offset_answer_support_start: int=None,
offset_answer_support_end: int=None,
sample_idx: int=None,
context: str=None,
offset_context_start: int=None,
offset_context_end: int=None,
n_samples_in_doc: int=None,
document_id: str=None,
passage_id: str=None
):
# self.answer_type can be "is_impossible", "yes", "no" or "span"
self.answer_type = answer_type
self.score = score
self.probability = probability

# If self.answer_type is "span", self.answer is a string answer span
# Otherwise, it is None
self.answer = answer
self.offset_answer_start = offset_answer_start
self.offset_answer_end = offset_answer_end

# If self.answer_type is in ["yes", "no"] then self.answer_support is a text string
# If self.answer is a string answer span or self.answer_type is "is_impossible", answer_support is None
# TODO sample_idx can probably be removed since we have passage_id
self.answer_support = answer_support
self.offset_answer_support_start = offset_answer_support_start
self.offset_answer_support_end = offset_answer_support_end
self.sample_idx = sample_idx

# self.context is the document or passage where the answer is found
self.context = context
self.offset_context_start = offset_context_start
self.offset_context_end = offset_context_end

# Offset unit is either "token" or "char"
# Aggregation level is either "doc" or "passage"
self.offset_unit = offset_unit
self.aggregation_level = aggregation_level

self.n_samples_in_doc = n_samples_in_doc
self.document_id = document_id
self.passage_id = passage_id


def to_doc_level(self, start, end):
Expand Down Expand Up @@ -81,16 +103,30 @@ class QAPred(Pred):
the attributes are found in the Pred class and not here. Pred in turn inherits from pydantic.BaseModel
which creates an __init__() method. See class definition for required and optional arguments.
"""

question: str
token_offsets: List[int]
context_window_size: int #TODO only needed for to_json() - can we get rid context_window_size, TODO Do we really need this?
aggregation_level: str
question_id: Optional[str]
answer_types: Optional[List[str]] = []
ground_truth_answer: Optional[str] = None
no_answer_gap: Optional[float] = None
n_samples: int = None
def __init__(self,
id: str,
prediction: List[Any],
context: str,
question: str,
token_offsets: List[int],
context_window_size: int,
aggregation_level: str,
answer_types: List[str]=None,
ground_truth_answer: str =None,
no_answer_gap: float =None,
n_samples: int=None,
question_id: int=None
):
super().__init__(id, prediction, context)
self.question = question
self.token_offsets = token_offsets
self.context_window_size = context_window_size #TODO only needed for to_json() - can we get rid context_window_size, TODO Do we really need this?
self.aggregation_level = aggregation_level
self.answer_types = answer_types
self.ground_truth_answer = ground_truth_answer
self.no_answer_gap = no_answer_gap
self.n_samples = n_samples
self.question_id = question_id

def to_json(self, squad=False):
answers = self.answers_to_json(squad)
Expand All @@ -100,7 +136,7 @@ def to_json(self, squad=False):
{
"question": self.question,
"question_id": self.question_id,
"ground_truth": None,
"ground_truth": self.ground_truth_answer,
"answers": answers,
"no_ans_gap": self.no_answer_gap # Add no_ans_gap to current no_ans_boost for switching top prediction
}
Expand All @@ -116,22 +152,21 @@ def answers_to_json(self, squad=False):
string = qa_answer.answer
start_t = qa_answer.offset_answer_start
end_t = qa_answer.offset_answer_end
score = qa_answer.score

_, ans_start_ch, ans_end_ch = span_to_string(start_t, end_t, self.token_offsets, self.context)
context_string, context_start_ch, context_end_ch = self.create_context(ans_start_ch, ans_end_ch, self.context)
if squad:
if string == "is_impossible":
string = ""
curr = {"score": score,
"probability": -1,
curr = {"score": qa_answer.score,
"probability": None,
"answer": string,
"offset_answer_start": ans_start_ch,
"offset_answer_end": ans_end_ch,
"context": context_string,
"offset_context_start": context_start_ch,
"offset_context_end": context_end_ch,
"document_id": self.id}
"document_id": qa_answer.document_id}
ret.append(curr)
return ret

Expand Down