chore: sync training sample inputs

Author: jortsupetterson

.gitignore

@@ -28,3 +28,6 @@ tesseract-artifacts/
 
 # Partial download artifacts from receipt dataset acquisition
 models/data/downloaded_*/**/archives/*.part-*
+
+# GitHub cannot store this archive because the repo/account LFS object cap is 2 GiB.
+models/.data/eng/images/huggingface.co/datasets/mahmoud2019/ReceiptQA/data.zip

models/.data/eng/images/huggingface.co/datasets/mahmoud2019/ReceiptQA/README.md

@@ -0,0 +1,123 @@
+---
+license: mit
+task_categories:
+- question-answering
+language:
+- en
+tags:
+- finance
+size_categories:
+- 100K<n<1M
+---
+# ReceiptQA: A Comprehensive Dataset for Receipt Understanding and Question Answering
+
+ReceiptQA is a large-scale dataset specifically designed to support and advance research in receipt understanding through question-answering (QA) tasks. This dataset offers a wide range of questions derived from real-world receipt images, addressing diverse challenges such as text extraction, layout understanding, and numerical reasoning. ReceiptQA provides a benchmark for evaluating and improving models for receipt-based QA tasks.
+
+
+
+## Dataset Overview
+ReceiptQA consists of 3,500 receipt images paired with 171,000 question-answer pairs, constructed using two complementary approaches:
+
+1. **LLM-Generated Subset:** 70,000 QA pairs generated by GPT-4o, validated by human annotators to ensure accuracy and relevance.
+2. **Human-Created Subset:** 101,000 QA pairs crafted manually, including both answerable and unanswerable questions for diverse evaluation.
+
+### Key Features:
+- Covers five domains: Retail, Food Services, Supermarkets, Fashion, and Medical.
+- Includes both straightforward and complex questions.
+- Offers a comprehensive benchmark for receipt-specific QA tasks.
+
+### Dataset Statistics
+| Domain          | Receipts | Human QA Pairs | LLM QA Pairs |
+|-----------------|----------|----------------|--------------|
+| Retail          | 800      | 23,200         | 16,000       |
+| Food Services   | 700      | 20,300         | 14,000       |
+| Supermarkets    | 700      | 20,300         | 14,000       |
+| Fashion         | 650      | 18,850         | 13,000       |
+| coffe shop        | 650      | 18,850         | 13,000       |
+| **Total**       | **3,500**| **101,935**    | **70,000**   |
+
+### Example of Data
+
+Here is a sample of the data structure used in the ReceiptQA dataset:
+
+```json
+{
+  "question": "What is the total amount for this receipt?",
+  "answer": "559.99 L.E"
+},
+{
+  "question": "What is the name of item 1?",
+  "answer": "Pullover PU-SOK1175"
+},
+{
+  "question": "What is the transaction number?",
+  "answer": "29786"
+},
+{
+  "question": "How many items were purchased?",
+  "answer": "2"
+}
+```
+## Requirements
+```bash
+# Install required libraries for inference
+pip install torch==1.10.0
+pip install transformers==4.5.0
+pip install datasets==2.3.0
+pip install Pillow
+```
+
+
+
+## Download Links
+
+### Full Dataset
+- **Train Set :** [Images](https://huggingface.co/datasets/mahmoud2019/ReceiptQA/resolve/main/train_images.zip?download=true) | [Labels](https://huggingface.co/datasets/mahmoud2019/ReceiptQA/resolve/main/train_label.zip?download=true) 
+- **Validation Set :** [Images](https://huggingface.co/datasets/mahmoud2019/ReceiptQA/resolve/main/validation_images.zip?download=true) | [Labels](https://huggingface.co/datasets/mahmoud2019/ReceiptQA/resolve/main/validation_label.zip?download=true) 
+- **Test Set :** [Images](https://huggingface.co/datasets/mahmoud2019/ReceiptQA/resolve/main/test_images.zip?download=true) | [Labels](https://huggingface.co/datasets/mahmoud2019/ReceiptQA/resolve/main/test_label.zip?download=true) 
+
+
+## Using ReceiptQA
+To use ReceiptQA for training or evaluation, follow these steps:
+
+### Step 1: Clone the Repository
+```bash
+git clone https://github.com/your-repo/ReceiptQA](https://github.com/MahmoudElsayedMahmoud/ReceiptQA-A-Comprehensive-Dataset-for-Receipt-Understanding-and-Question-Answering
+cd ReceiptQA
+```
+
+### Step 2: Download the Dataset
+Download the dataset using the links provided above and place it in the `data/` directory.
+
+
+## Evaluation Metrics
+ReceiptQA provides the following metrics for evaluating QA models:
+- **Exact Match (EM):** Measures if the predicted answer exactly matches the ground truth.
+- **F1 Score:** Evaluates the overlap between the predicted and ground truth answers.
+- **Precision:** Measures the accuracy of the predictions.
+- **Recall:** Measures the ability to retrieve relevant answers.
+- **Answer Containment:** Checks if the ground truth answer is included in the predicted response.
+
+## Models Benchmarked
+ReceiptQA has been used to evaluate state-of-the-art models, including:
+- **GPT-4**
+- **Llama3.2 (11B)**
+- **Gemni 2.0**
+- **Phi 3.5 Vision**
+- **InternVL2 (4B/8B)**
+- **LLaVA 7B**
+
+
+
+## Citation
+If you use ReceiptQA in your research, please cite our paper:
+```
+Will be publish soon !!
+```
+
+
+
+## Contact
+For questions or feedback, please contact:
+- Mahmoud Abdalla: [mahmoudelsayed@chungbuk.ac.kr](mailto:mahmoudelsayed@chungbuk.ac.kr)
+- GitHub Issues: [Submit an issue](https://github.com/your-repo/ReceiptQA/issues)

models/.data/eng/images/huggingface.co/datasets/mahmoud2019/ReceiptQA/mit.md

@@ -0,0 +1,50 @@
+---
+title: MIT License
+spdx-id: MIT
+featured: true
+hidden: false
+
+description: A short and simple permissive license with conditions only requiring preservation of copyright and license notices. Licensed works, modifications, and larger works may be distributed under different terms and without source code.
+
+how: Create a text file (typically named LICENSE or LICENSE.txt) in the root of your source code and copy the text of the license into the file. Replace [year] with the current year and [fullname] with the name (or names) of the copyright holders.
+
+using:
+  Babel: https://github.com/babel/babel/blob/master/LICENSE
+  .NET: https://github.com/dotnet/runtime/blob/main/LICENSE.TXT
+  Rails: https://github.com/rails/rails/blob/master/MIT-LICENSE
+
+permissions:
+  - commercial-use
+  - modifications
+  - distribution
+  - private-use
+
+conditions:
+  - include-copyright
+
+limitations:
+  - liability
+  - warranty
+
+---
+MIT License
+
+Copyright (c) [year] [fullname]
+
+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.