Upload app_public_version.py

app_public_version.py

@@ -18,6 +18,22 @@ from contextual import ContextualAI
 from openai import AzureOpenAI
 from datetime import datetime
 import sys
+from datetime import datetime
+from pathlib import Path
+from uuid import uuid4
+from huggingface_hub import CommitScheduler
+
+
+JSON_DATASET_DIR = Path("json_dataset")
+JSON_DATASET_DIR.mkdir(parents=True, exist_ok=True)
+JSON_DATASET_PATH = JSON_DATASET_DIR / f"train-{uuid4()}.json"
+
+scheduler = CommitScheduler(
+    repo_id="ai-law-society-lab/NJ-caselaw-queries",
+    repo_type="dataset",
+    folder_path=JSON_DATASET_DIR,
+    path_in_repo="data", token=os.getenv('hf_token')
+)
 
 
 def format_metadata_as_str(metadata):
@@ -123,7 +139,11 @@ def register_review(history, additional_feedback, response_json):
         "timestamp": datetime.now().strftime('%Y-%m-%d %H:%M:%S'), 
         "additional_feedback": additional_feedback
     }
-    print (res)
+
+    with scheduler.lock:
+        with JSON_DATASET_PATH.open("a") as f:
+            json.dump(res, f)
+            f.write("\n")
 
 
 # load search functionality here
@@ -202,7 +222,7 @@ def run_retrieval(query):
             item["meta_data"] = format_metadata_as_str(metadata[item["opinion_idx"]])
         else:
             item["meta_data"] = ""        
-        out_dict.append(tmp)
+        out_dict.append(item)
     return out_dict
 
 
@@ -234,7 +254,7 @@ if "NV" in model_name:
     model = load_embeddings_model(model_name=model_name)
 
 
-examples = ["Can officers always order a passenger out of a car?","Find me briefs about credential searches", "Can police search an impounded car without a warrant?", "State is arguing State v. Carty is not good law"]
+examples = ["Can officers always order a passenger out of a car?"]
 
 css = """
 .svelte-i3tvor {visibility: hidden}