Fix: Separate train and test configs to handle schema differences

README.md

@@ -20,69 +20,81 @@ configs:
       - split: java_test
         path: java_zeroshot_test.jsonl
 
-  # A/B format (instruction + input + output) - for instruction tuning
-  - config_name: ab
+  # A/B format - Training data only (instruction + input + output)
+  - config_name: ab_train
     data_files:
       - split: cpp_train
         path: cpp_ab_train.jsonl
       - split: cpp_val
         path: cpp_ab_val.jsonl
-      - split: cpp_test
-        path: cpp_ab_test.jsonl
       - split: python_train
         path: python_ab_train.jsonl
       - split: python_val
         path: python_ab_val.jsonl
-      - split: python_test
-        path: python_ab_test.jsonl
       - split: java_train
         path: java_ab_train.jsonl
       - split: java_val
         path: java_ab_val.jsonl
+
+  # A/B format - Test data (has extra columns: speedup, log2_speedup)
+  - config_name: ab_test
+    data_files:
+      - split: cpp_test
+        path: cpp_ab_test.jsonl
+      - split: python_test
+        path: python_ab_test.jsonl
       - split: java_test
         path: java_ab_test.jsonl
 
-  # A/B with diff format
-  - config_name: abdiff
+  # A/B with diff format - Training data only
+  - config_name: abdiff_train
     data_files:
       - split: cpp_train
         path: cpp_abdiff_train.jsonl
       - split: cpp_val
         path: cpp_abdiff_val.jsonl
-      - split: cpp_test
-        path: cpp_abdiff_test.jsonl
       - split: python_train
         path: python_abdiff_train.jsonl
       - split: python_val
         path: python_abdiff_val.jsonl
-      - split: python_test
-        path: python_abdiff_test.jsonl
       - split: java_train
         path: java_abdiff_train.jsonl
       - split: java_val
         path: java_abdiff_val.jsonl
+
+  # A/B with diff format - Test data
+  - config_name: abdiff_test
+    data_files:
+      - split: cpp_test
+        path: cpp_abdiff_test.jsonl
+      - split: python_test
+        path: python_abdiff_test.jsonl
       - split: java_test
         path: java_abdiff_test.jsonl
 
-  # A with diff format
-  - config_name: adiff
+  # A with diff format - Training data only
+  - config_name: adiff_train
     data_files:
       - split: cpp_train
         path: cpp_adiff_train.jsonl
       - split: cpp_val
         path: cpp_adiff_val.jsonl
-      - split: cpp_test
-        path: cpp_adiff_test.jsonl
       - split: python_train
         path: python_adiff_train.jsonl
       - split: python_val
         path: python_adiff_val.jsonl
-      - split: python_test
-        path: python_adiff_test.jsonl
       - split: java_train
         path: java_adiff_train.jsonl
       - split: java_val
         path: java_adiff_val.jsonl
+
+  # A with diff format - Test data
+  - config_name: adiff_test
+    data_files:
+      - split: cpp_test
+        path: cpp_adiff_test.jsonl
+      - split: python_test
+        path: python_adiff_test.jsonl
       - split: java_test
         path: java_adiff_test.jsonl
 
@@ -116,13 +128,16 @@ This dataset contains pairs of functionally equivalent code snippets with differ
 
 ## 📁 Configurations
 
-| Config | Description | Splits | Schema |
-|--------|-------------|--------|--------|
-| `benchmark` | Main benchmark (raw code pairs) | cpp, python, java | `slow_code`, `fast_code`, `speedup`, `difficulty` |
-| `zeroshot` | Zero-shot LLM evaluation | {lang}_test | `system_prompt`, `user_prompt`, `code_a`, `code_b`, `expected` |
-| `ab` | A/B comparison (instruction tuning) | {lang}_train/val/test | `instruction`, `input`, `output` |
-| `abdiff` | A/B with code diff | {lang}_train/val/test | `instruction`, `input`, `output` |
-| `adiff` | Single code with diff | {lang}_train/val/test | `instruction`, `input`, `output` |
+| Config | Description | Splits |
+|--------|-------------|--------|
+| `benchmark` | Main benchmark (raw code pairs with timing) | cpp, python, java |
+| `zeroshot` | Zero-shot LLM evaluation format | {lang}_test |
+| `ab_train` | A/B comparison - training/validation | {lang}_train, {lang}_val |
+| `ab_test` | A/B comparison - test set | {lang}_test |
+| `abdiff_train` | A/B with diff - training/validation | {lang}_train, {lang}_val |
+| `abdiff_test` | A/B with diff - test set | {lang}_test |
+| `adiff_train` | Single code with diff - training/validation | {lang}_train, {lang}_val |
+| `adiff_test` | Single code with diff - test set | {lang}_test |
 
 ## 🚀 Usage
 
@@ -138,11 +153,12 @@ ds_zs = load_dataset("JinNian0072/efficodebench-dataset", "zeroshot")
 print(ds_zs["python_test"][0])
 
 # Load A/B format for training
-ds_ab = load_dataset("JinNian0072/efficodebench-dataset", "ab")
-print(ds_ab["cpp_train"][0])
+ds_train = load_dataset("JinNian0072/efficodebench-dataset", "ab_train")
+print(ds_train["cpp_train"][0])
 
-# Load specific language and split
-ds_java = load_dataset("JinNian0072/efficodebench-dataset", "abdiff", split="java_test")
+# Load A/B format for testing
+ds_test = load_dataset("JinNian0072/efficodebench-dataset", "ab_test")
+print(ds_test["java_test"][0])
 ```
 
 ## 📋 Data Format