Datasets:
metadata
license: cc-by-sa-4.0
language:
- vi
size_categories:
- 100K<n<1M
configs:
- config_name: products
data_files:
- split: test
path:
- products.parquet
features:
- name: product_id
dtype: string
- name: shop_id
dtype: string
- name: brand
dtype: string
- name: image_path
dtype: string
- name: price
dtype: int64
- name: processed_description
dtype: string
- name: product_name
dtype: string
- name: translated_brand
dtype: string
- name: translated_description
dtype: string
- name: translated_name
dtype: string
- config_name: reviews
data_files:
- split: test
path: reviews.parquet
tags:
- tabular
Script to download and use the dataset:
from datasets import load_dataset
from huggingface_hub import hf_hub_download
import os
from zipfile import ZipFile
from PIL import Image
# 1. Load products and reviews datasets
products = load_dataset("allietran/CAMEO", "products", split="test")
reviews = load_dataset("allietran/CAMEO", "reviews", split="test")
# 2. Download the images
zip_path = hf_hub_download(repo_id="allietran/CAMEO", filename="images.zip", repo_type="dataset")
# 3. Extract images to a local folder
extract_path = "./images"
with ZipFile(zip_path, 'r') as zip_ref:
zip_ref.extractall(extract_path)
# 4. Access a row
sample = products[0]
relative_path = sample["image_path"]
# 5. Open the image manually
full_path = os.path.join(extract_path, relative_path)
if os.path.exists(full_path):
img = Image.open(full_path)
img.show()
else:
print(f"File not found at: {full_path}")