# Builder script # coding=utf-8 # Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. # MIT License # Copyright (c) PlotQA. # Permission is hereby granted, free of charge, to any person obtaining a copy # of this software and associated documentation files (the "Software"), to deal # in the Software without restriction, including without limitation the rights # to use, copy, modify, merge, publish, distribute, sublicense, and/or sell # copies of the Software, and to permit persons to whom the Software is # furnished to do so, subject to the following conditions: # The above copyright notice and this permission notice shall be included in all # copies or substantial portions of the Software. # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR # IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, # FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE # AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER # LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE # SOFTWARE """PlotQA dataset""" import copy import json import os import pandas as pd # importing the "tarfile" module import tarfile import datasets from datasets import load_dataset _CITATION = """\ @inproceedings{methani2020plotqa, title={Plotqa: Reasoning over scientific plots}, author={Methani, Nitesh and Ganguly, Pritha and Khapra, Mitesh M and Kumar, Pratyush}, booktitle={Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision}, pages={1527--1536}, year={2020} } """ _DESCRIPTION = """\ PlotQA dataset Chart images, tables, image annotations, questions, answers """ _LICENSE = "CC-BY-4.0 license" _SPLITS = ["train", "val", "test"] _URL = "https://huggingface.co/datasets/Dodon/PlotQA_dataset/resolve/main/PlotQA.zip" class PlotQA(datasets.GeneratorBasedBuilder): def _info(self): features = datasets.Features( { "imgname": datasets.Value("string"), #"image": datasets.Image(), #"version": datasets.Value("string"), "query": datasets.Sequence(datasets.Value("string")), "query_token": datasets.Sequence(datasets.Value("string")), "label": datasets.Sequence(datasets.Value("string")), "img_ann": datasets.Value("string"), ### "image_index": datasets.Value("string"), #"question_string": datasets.Value("string"), #"answer": datasets.Value("string"), #"answer_bbox": datasets.Value("string"), #"template": datasets.Value("string"), #"type": datasets.Value("string"), #This format required change ## "table_name": datasets.Value("string"), ## "table": datasets.Value("string"), #"table": datasets.table.Table(), } ) return datasets.DatasetInfo( description=_DESCRIPTION, features=features, supervised_keys=None, license=_LICENSE, ) def _split_generators(self, dl_manager): downloaded_file = dl_manager.download_and_extract(_URL) + "/PlotQA" return [ datasets.SplitGenerator( name=datasets.Split.TRAIN, gen_kwargs={ "annotation_v1_path": downloaded_file + "/train/qa_pairs_V1.json", "annotation_v2_path": downloaded_file + "/train/qa_pairs_V2.json", "images_path": downloaded_file + "/train/png.tar.gz", "img_anno_path": downloaded_file + "/train/annotations.json", ##"table_path": downloaded_file + "/train/tables", }, ), datasets.SplitGenerator( name=datasets.Split.VALIDATION, gen_kwargs={ "annotation_v1_path": downloaded_file + "/validation/qa_pairs_V1.json", "annotation_v2_path": downloaded_file + "/validation/qa_pairs_V2.json", "images_path": downloaded_file + "/validation/png.tar.gz", "img_anno_path": downloaded_file + "/validation/annotations.json", ##"table_path": downloaded_file + "/train/tables", }, ), datasets.SplitGenerator( name=datasets.Split.TEST, gen_kwargs={ "annotation_v1_path": downloaded_file + "/test/qa_pairs_V1.json", "annotation_v2_path": downloaded_file + "/test/qa_pairs_V2.json", "images_path": downloaded_file + "/test/png.tar.gz", "img_anno_path": downloaded_file + "/test/annotations.json", ##"table_path": downloaded_file + "/train/tables", }, ), ] # def find_qa(self, annotation_v1_path:str, annotation_v2_path:str, img_idx:int): # with open(annotation_v1_path, "r", encoding="utf-8") as v1: # data_v1 = json.load(v1) # with open(annotation_v2_path, "r", encoding="utf-8") as v2: # data_v2 = json.load(v2) # #1. # version = 1 # _temp_item = [] # for ele in data_v1: # if ele['image_index'] == img_idx: # _temp_item.append(ele) # for ele in data_v2: # if ele['image_index'] == img_idx: # _temp_item.append(ele) # return _temp_item def _generate_examples(self, annotation_v1_path:str, annotation_v2_path:str, img_anno_path:str ,images_path: str): #Load image folder # open file ## file = tarfile.open(images_path) # extracting file ## file.extractall('./imgs') ## file.close() _multi_anno = [annotation_v1_path, annotation_v2_path] with open(_multi_anno, "r", encoding="utf-8") as v: data = json.load(v) idx = 0 with open(img_anno_path, "r", encoding="utf-8") as a: img_data = json.load(a) for ele in img_data: item = {} item["img_ann"] = ele item["imgname"] = str(ele['image_index'])+'.png' # item['image'] = os.path.join('./imgs',item['imgname']) item["query_token"] = [] _temp_item = [] for element in data_v1: if element['image_index'] == ele['image_index']: _temp_item.append(element) #qa_returns = find_qa(annotation_v1_path,annotation_v2_path, item['image_index']) _question = [] _label = [] for pair in qa_returns: _question.append(pair['question_string']) _label.append(pair["answer"]) # file = os.path.splitext(file_name) # if file == "test_augmented": # item["human"] = False # else: # item["human"] = True # img_anot_file = os.path.splitext(item["imgname"])[0]+'.json' # img_anot = os.path.join(img_anno_path, img_anot_file) # with open(img_anot) as f: # item["img_ann"] = json.load(f) """ item['table'] = os.path.join(images_path,item["imgname"]) # annotation item["img_anno"] = load json file... t_path = os.path.join(table_path,item["table_name"]) table_data = load_dataset("csv", data_files=[t_path]) yield table_data """ yield idx, item idx += 1