# 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. # TODO: Address all TODOs and remove all explanatory comments """TODO: Add a description here.""" import csv import json import os import datasets # TODO: Add BibTeX citation # Find for instance the citation on arxiv or on the dataset repo/website _CITATION = """\ @misc{ li2023what, title={What Does Vision Supervision Bring to Language Models? A Case Study of {CLIP}}, author={Lei Li and Jingjing Xu and Qingxiu Dong and Ce Zheng and Qi Liu and Lingpeng Kong and Xu Sun}, year={2023}, url={https://openreview.net/forum?id=SdBfRJE9SX-} } """ # TODO: Add description of the dataset here # You can copy an official description _DESCRIPTION = """\ Visual and Embodied Concept (VEC) benchmark is designed for evaluating the LLM understanding ability of basic visual (color, shape, size, height and material) and embodied (mass, temperature, hardness) concepts. """ # TODO: Add a link to an official homepage for the dataset here _HOMEPAGE = "" # TODO: Add the licence for the dataset here if you can find it _LICENSE = "Apache 2.0" # TODO: Add link to the official dataset URLs here # The HuggingFace Datasets library doesn't host the datasets but only points to the original files. # This can be an arbitrary nested dict/list of URLs (see below in `_split_generators` method) _URLS = { "color": {"test": "./data/color.json"}, "shape": {"test":"./data/shape.json"}, "size": {"test":"./data/size.json"}, "height": {"test": "./data/height.json"}, "material": {"test":"./data/material.json"}, "hardness": {"test":"./data/hardness.json"}, "temperature": {"test":"./data/temperature.json"}, "mass": {"test":"./data/mass.json"}, } # TODO: Name of the dataset usually matches the script name with CamelCase instead of snake_case class VECDataset(datasets.GeneratorBasedBuilder): """VEC dataset for evaluating visual and embodied concepts""" VERSION = datasets.Version("1.1.0") # This is an example of a dataset with multiple configurations. # If you don't want/need to define several sub-sets in your dataset, # just remove the BUILDER_CONFIG_CLASS and the BUILDER_CONFIGS attributes. # If you need to make complex sub-parts in the datasets with configurable options # You can create your own builder configuration class to store attribute, inheriting from datasets.BuilderConfig # BUILDER_CONFIG_CLASS = MyBuilderConfig # You will be able to load one or the other configurations in the following list with # data = datasets.load_dataset('my_dataset', 'first_domain') # data = datasets.load_dataset('my_dataset', 'second_domain') BUILDER_CONFIGS = [ datasets.BuilderConfig(name="mass", version=VERSION, description="mass dataset"), datasets.BuilderConfig(name="temperature", version=VERSION, description="temperature dataset"), datasets.BuilderConfig(name="hardness", version=VERSION, description="hardness dataset"), datasets.BuilderConfig(name="shape", version=VERSION, description="shape dataset"), datasets.BuilderConfig(name="size", version=VERSION, description="size dataset"), datasets.BuilderConfig(name="material", version=VERSION, description="material dataset"), datasets.BuilderConfig(name="color", version=VERSION, description="color dataset"), datasets.BuilderConfig(name="height", version=VERSION, description="height dataset"), ] DEFAULT_CONFIG_NAME = "hardness" # It's not mandatory to have a default configuration. Just use one if it make sense. def _info(self): # TODO: This method specifies the datasets.DatasetInfo object which contains informations and typings for the dataset if self.config.name in ["color", "shape", "material"]: # This is the name of the configuration selected in BUILDER_CONFIGS above features = datasets.Features( { "obj": datasets.Value("string"), "positive": datasets.Value("string"), "negative": datasets.Value("string"), "relation": datasets.Value("string"), # These are the features of your dataset like images, labels ... } ) else: # for pair comparison features = datasets.Features( { "obj1": datasets.Value("string"), "obj2": datasets.Value("string"), "relation": datasets.Value("string"), "label": datasets.ClassLabel(num_classes=2, names=["<", ">"]) # These are the features of your dataset like images, labels ... } ) return datasets.DatasetInfo( # This is the description that will appear on the datasets page. description=_DESCRIPTION, # This defines the different columns of the dataset and their types features=features, # Here we define them above because they are different between the two configurations # If there's a common (input, target) tuple from the features, uncomment supervised_keys line below and # specify them. They'll be used if as_supervised=True in builder.as_dataset. # supervised_keys=("sentence", "label"), # Homepage of the dataset for documentation homepage=_HOMEPAGE, # License for the dataset if available license=_LICENSE, # Citation for the dataset citation=_CITATION, ) def _split_generators(self, dl_manager): # TODO: This method is tasked with downloading/extracting the data and defining the splits depending on the configuration # If several configurations are possible (listed in BUILDER_CONFIGS), the configuration selected by the user is in self.config.name # dl_manager is a datasets.download.DownloadManager that can be used to download and extract URLS # It can accept any type or nested list/dict and will give back the same structure with the url replaced with path to local files. # By default the archives will be extracted and a path to a cached folder where they are extracted is returned instead of the archive urls = _URLS[self.config.name] data_dir = dl_manager.download_and_extract(urls) return [ datasets.SplitGenerator( name=datasets.Split.TEST, # These kwargs will be passed to _generate_examples gen_kwargs={ "filepath": data_dir["test"], "split": "test" }, ), ] # method parameters are unpacked from `gen_kwargs` as given in `_split_generators` def _generate_examples(self, filepath, split): # TODO: This method handles input defined in _split_generators to yield (key, example) tuples from the dataset. # The `key` is for legacy reasons (tfds) and is not important in itself, but must be unique for each example. with open(filepath, encoding="utf-8") as f: for key, row in enumerate(f): data = json.loads(row) if self.config.name in ['color', 'shape', 'material']: # Yields examples as (key, example) tuples # {"sub": "jacket", "obj": "black", "alt": "purple"} yield f"{self.config.name}-{key}", { "obj": data['sub'].strip(), "positive": data['obj'].strip(), "negative": data['alt'].strip(), "relation": self.config.name # change to prompt template later } elif self.config.name in ["shape", "height"]: # shape #{"text": "An ant and a bird.", "question": "Is an ant taller than a bird?", "obj_a": "ant", "obj_b": "bird", "label": 0, "obj1": "ant", "obj2": "bird"} yield f"{self.config.name}-{key}", { "obj1": data['obj1'].strip(), "obj2": data['obj2'].strip(), "relation": self.config.name, # change to prompt template later "label": data['label'] } else: # hardness, mass, temperature yield f"{self.config.name}-{key}", { "obj1": data['obj1'].strip(), "obj2": data['obj2'].strip(), "relation": self.config.name, "label": data['label'], }