Update README.md

README.md

@@ -9,8 +9,10 @@ license: mit
 ## 0.1 Supported Tasks
 
 - What tasks can be performed on this dataset?
+  
   Developing robust and reliable multimodal QA systems for ECG interpretation relies on the availability of both high-quality and large quantities of labeled data. Yet, obtaining massive amounts of labeled ECGs from cardiologists is costly, which often results in limited datasets. Traditional supervised learning methods tend to perform well only on data with the same distribution as the training data. In real-world deployment, however, models frequently encounter new tasks and previously unseen populations outside the training distribution, where traditional methods may fail.
 - Are there any code associated with this dataset?
+  
   Electrocardiogram–Language Model for Few-Shot Question Answering with Meta Learning(ICASSP 2025), https://arxiv.org/html/2410.14464v1
 
 ## 0.2 Dataset Highlights
@@ -40,6 +42,7 @@ The dataset is a structured reorganization of the existing ECG-QA dataset, adapt
     "attribute": ["non-diagnostic t abnormalities"]
 }
 ```
+
 ### 1.1.2 Source Datasets Instances distribution
 
 <img src="img/distribution_attr.png" alt="Figure 1: Illustration of class formation and attribute distribution for different question types." width="600"/>
@@ -68,45 +71,102 @@ python load_class.py --paraphrased_path /your/actual/path/to/ecgqa/mimic-iv-ecg/
 | Single-Query    | 30         | attr_1/attr_2/.../attr_n         | 260 (208:52)          | 63,125  | Q: What direction is this ECG deviated to? <br> A: Normal axis/...     |
 | All             | 206        | yes/no/both/none/.../attr_n      | 678 (541:137)         | 144,885 | ...                                                                     |
 
-### Few-shot Build
+## 1.3 Few-shot Build
 
+### 1.3.1 load ECG-QA-FSL dataset
 
 ```python
-train_data = dataset["train"]
-validation_data = dataset["validation"]
-test_data = dataset["test"]
-
-for example in train_data.select(range(3)):
-    print(example)
-```
-
-```bash
-python data_loader.py --paraphrased_path /your/actual/path/to/ecgqa/mimic-iv-ecg/paraphrased --test_dataset mimic
+python data_loader.py
+--model_name /your/actual/path/to/model/download/from/hugging/face
+--paraphrased_path /your/actual/path/to/ecgqa/mimic-iv-ecg/paraphrased
+--test_dataset ptb-xl
 ```
+### 1.3.2 sample of ECG-QA-FSL dataset
 
-<img src="img/FSL_ECG_QAMeta-Learning.png" alt="Few-shot Meta-learning Example" width="600"/>
-
-
-## Dataset Structure
 
-### Data Fields
-
-- `feature1`: a `string` feature representing <!-- description -->
-- `feature2`: a `string` feature representing <!-- description -->
-- `label`: a `int64` classification label, with 0 indicating <!-- meaning --> and 1 indicating <!-- meaning -->
+```python
+episode = {
+    # --- Support Set ---
+    "support_x": [  # ECG feature tensors or preprocessed arrays
+        ecg_sample_1,  # typically a NumPy array or tensor
+        ecg_sample_2,
+        # ...
+    ],
+    "support_y_q": [  # Question token sequences (padded)
+        [12, 45, 78, 0, 0, 0],
+        [23, 67, 89, 90, 0, 0],
+        # ...
+    ],
+    "support_y_a": [  # Answer token sequences (padded)
+        [1, 0, 0, 0, 0, 0],
+        [1, 1, 0, 0, 0, 0],
+        # ...
+    ],
+    "support_y_q_mask": [  # Mask for question tokens (1 = valid, 0 = pad)
+        [1, 1, 1, 0, 0, 0],
+        [1, 1, 1, 1, 0, 0],
+        # ...
+    ],
+    "support_y_a_mask": [  # Mask for answer tokens
+        [1, 0, 0, 0, 0, 0],
+        [1, 1, 0, 0, 0, 0],
+        # ...
+    ],
+    "flatten_support_x": [  # File paths to raw ECG signals
+        "/path/to/support/ecg_1.npy",
+        "/path/to/support/ecg_2.npy",
+        # ...
+    ],
+
+    # --- Query Set ---
+    "query_x": [
+        ecg_query_1,
+        ecg_query_2,
+        # ...
+    ],
+    "query_y_q": [
+        [34, 78, 56, 0, 0, 0],
+        [90, 12, 45, 76, 0, 0],
+        # ...
+    ],
+    "query_y_a": [
+        [0, 1, 1, 0, 0, 0],
+        [1, 0, 0, 0, 0, 0],
+        # ...
+    ],
+    "query_y_q_mask": [
+        [1, 1, 1, 0, 0, 0],
+        [1, 1, 1, 1, 0, 0],
+        # ...
+    ],
+    "query_y_a_mask": [
+        [1, 1, 1, 0, 0, 0],
+        [1, 0, 0, 0, 0, 0],
+        # ...
+    ],
+    "flatten_query_x": [
+        "/path/to/query/ecg_1.npy",
+        "/path/to/query/ecg_2.npy",
+        # ...
+    ],
+}
+```
 
-### Data Splits
+### 1.3.3 Data Splits
 
 - Number of instances in each split (train/test): 8:2
 - Criteria: first split based on template id (no expression overlap between train/test), then random split for support/query set in few-shot tasks.
 
-## Dataset Creation
+### 1.3.4 ECG-QA-FSL dataset for meta-learning
+<img src="img/FSL_ECG_QAMeta-Learning.png" alt="Few-shot Meta-learning Example" width="600"/>
+
+# 2. Summary
 
-### Curation Rationale
+## 2.1 Curation Rationale
 
 Developing robust and reliable multimodal QA systems for ECG interpretation relies on the availability of both high-quality and large quantities of labeled data. Meta-learning, a paradigm focused on “learning to learn”, enables them to acquire transferable knowledge and adapt rapidly to new, unseen tasks with minimal labeled data.
 
-## Citation
+## 2.2 Citation
 
 ```
 @inproceedings{10888594,
@@ -118,39 +178,3 @@ Developing robust and reliable multimodal QA systems for ECG interpretation reli
   doi={10.1109/ICASSP49660.2025.10888594}
 }
 ```
-
-## How to Use
-
-### Example Preprocessing and Training
-
-```python
-from transformers import AutoTokenizer, AutoModelForSequenceClassification, Trainer, TrainingArguments
-
-tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
-
-def tokenize_function(examples):
-    return tokenizer(examples["feature1"], padding="max_length", truncation=True)
-
-tokenized_dataset = dataset.map(tokenize_function, batched=True)
-
-model = AutoModelForSequenceClassification.from_pretrained("bert-base-uncased", num_labels=2)
-
-training_args = TrainingArguments(
-    output_dir="./results",
-    per_device_train_batch_size=16,
-    per_device_eval_batch_size=16,
-    num_train_epochs=3,
-    evaluation_strategy="epoch",
-    save_strategy="epoch",
-    load_best_model_at_end=True,
-)
-
-trainer = Trainer(
-    model=model,
-    args=training_args,
-    train_dataset=tokenized_dataset["train"],
-    eval_dataset=tokenized_dataset["validation"],
-)
-
-trainer.train()
-```