Add training data (9750 AI-Human pairs) and updated app.py

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+*.jsonl filter=lfs diff=lfs merge=lfs -text

app.py

@@ -20,9 +20,17 @@ from peft import LoraConfig, get_peft_model, TaskType, PeftModel
 MODEL_NAME = "SamSJackson/paraphrase-dipper-no-ctx"
 DATA_DIR = "/data" if os.path.exists("/data") else "."
 OUTPUT_DIR = os.path.join(DATA_DIR, "dipper-humanizer-lora")
-DATA_FILE = os.path.join(DATA_DIR, "training_pairs.jsonl")
 FINAL_MODEL_DIR = os.path.join(OUTPUT_DIR, "final")
 
+# Copy training data from repo to persistent storage if needed
+REPO_DATA = os.path.join(os.path.dirname(os.path.abspath(__file__)), "training_pairs.jsonl")
+DATA_FILE = os.path.join(DATA_DIR, "training_pairs.jsonl")
+if os.path.exists(REPO_DATA) and not os.path.exists(DATA_FILE):
+    import shutil
+    os.makedirs(DATA_DIR, exist_ok=True)
+    shutil.copy2(REPO_DATA, DATA_FILE)
+    print(f"Copied training data to {DATA_FILE}")
+
 LORA_R = 16
 LORA_ALPHA = 32
 LORA_DROPOUT = 0.05

training_pairs.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94edc7e8afd4c607226b7b3caab2233f99f1b23a489dd2bd7e89b8a3e6fc54b6
+size 45452356