write into one file

app.py

@@ -17,36 +17,32 @@ with open("system_prompt.txt", "r") as f:
 client = InferenceClient(MODEL_NAME)
 api = HfApi()
 
-# ---- Fixed generation parameters ----
-MAX_TOKENS = 512
-TEMPERATURE = 0.7
-TOP_P = 0.95
+# ---- Session ID and local file setup ----
+SESSION_ID = str(uuid.uuid4())
+LOCAL_LOG_PATH = f"chatlog_{SESSION_ID}.jsonl"
+REMOTE_LOG_PATH = f"sessions/{SESSION_ID}.jsonl"  # stored in a folder in the dataset
 
-# ---- Logging function ----
-def log_to_dataset(user_message, assistant_message):
+# ---- Logging per session ----
+def append_to_session_log(user_message, assistant_message):
     row = {
         "timestamp": datetime.now().isoformat(),
-        "session_id": str(uuid.uuid4()),
         "user": user_message,
         "assistant": assistant_message,
         "system_prompt": SYSTEM_PROMPT,
+        "session_id": SESSION_ID
     }
-
-    # Save as a JSONL file and upload to the dataset repo
-    filename = f"log_{datetime.now().strftime('%Y%m%d_%H%M%S')}_{uuid.uuid4().hex[:8]}.jsonl"
-    with open(filename, "w", encoding="utf-8") as f:
+    with open(LOCAL_LOG_PATH, "a", encoding="utf-8") as f:
         f.write(json.dumps(row) + "\n")
 
+    # Push to HF dataset
     api.upload_file(
-        path_or_fileobj=filename,
-        path_in_repo=filename,
+        path_or_fileobj=LOCAL_LOG_PATH,
+        path_in_repo=REMOTE_LOG_PATH,
         repo_id=DATASET_REPO,
         repo_type="dataset",
         token=HF_TOKEN
     )
 
-    os.remove(filename)
-
 # ---- Chatbot function ----
 def respond(message, history):
     messages = [{"role": "system", "content": SYSTEM_PROMPT}]
@@ -62,18 +58,18 @@ def respond(message, history):
 
     for chunk in client.chat_completion(
         messages,
-        max_tokens=MAX_TOKENS,
+        max_tokens=512,
         stream=True,
-        temperature=TEMPERATURE,
-        top_p=TOP_P,
+        temperature=0.7,
+        top_p=0.95,
     ):
         token = chunk.choices[0].delta.content
         if token:
             response += token
             yield response
 
-    # Log after full response is generated
-    log_to_dataset(message, response)
+    # Save after each message pair
+    append_to_session_log(message, response)
 
 # ---- Gradio Interface ----
 demo = gr.ChatInterface(