Add training CSV (3,063 instruction-code pairs) and document it

README.md

@@ -22,6 +22,10 @@ configs:
     data_files:
       - split: train
         path: code4ml_annotated.csv
+  - config_name: training
+    data_files:
+      - split: train
+        path: code4ml_annotated_training.csv
 ---
 
 # Code4ML Annotated Subset with SonarQube and MLScent Ratings
@@ -137,6 +141,30 @@ No rows were removed or modified; duplicates and short snippets are kept as in t
 3. The data was split into two halves, each rated by its own SonarQube + MLScent instance
    pair, then merged back by `id`.
 
+## Training file
+
+`code4ml_annotated_training.csv` holds 3,063 instruction–code pairs built from
+`code4ml_annotated.csv` for fine-tuning:
+
+| Step | Rule | Rows dropped |
+|---|---|---:|
+| 1 | Drop task classes without ML logic: `Debug`, `Environment`, `Other` | 2,259 |
+| 2 | Drop `combined_rating` = N/A | 2,502 |
+| 2 | Drop `combined_rating` = D or E (quality floor) | 121 |
+| 3 | Drop duplicate `code_block` | 0 |
+| 4 | Turn the subclass label into an instruction, e.g. `train_model` → "Write Python code to train the model." | — |
+| 5 | Drop snippets whose 8-grams overlap a DS-1000 reference solution by ≥ 50% | 2 |
+| 6 | Split 80/10/10 per task class with seed 42 | — |
+
+| Column | Description |
+|---|---|
+| `id` | row `id` in `code4ml_annotated.csv` |
+| `split` | `train` (2,451), `validation` (308), or `test` (304) |
+| `graph_vertex_class`, `graph_vertex_subclass` | Code4ML task label |
+| `instruction` | instruction sentence built from the subclass label |
+| `code` | the snippet |
+| `combined_rating` | `A` (2,122), `B` (868), or `C` (73) |
+
 ## Usage
 
 ```python
@@ -144,6 +172,9 @@ from datasets import load_dataset
 
 ds = load_dataset("dimsmauls/code4ml-annotated-rated", split="train")
 print(ds[0]["graph_vertex_class"], ds[0]["combined_rating"], ds[0]["code_block"])
+
+train = load_dataset("dimsmauls/code4ml-annotated-rated", "training", split="train")
+train_set = train.filter(lambda row: row["split"] == "train")
 ```
 
 ## Provenance & attribution

code4ml_annotated_training.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a363f17fe6e28bd2d3a06ea99021f2a63b9c571f2ec83a22d617be1fbe92fa03
+size 959034