Upload update added scheduled logging

Test commit to update app file with scheduled commit functions

app.py

@@ -1,22 +1,35 @@
+from sentence_transformers import SentenceTransformer
+from huggingface_hub import CommitScheduler
+from datasets import Dataset
 import gradio as gr
-
 import pandas as pd
-from datasets import Dataset
-from sentence_transformers import SentenceTransformer
+
+
 from parameter_extractor import ParameterExtractor
 from DNAseq import DNAseq
-from helper import list_at_index_0, list_at_index_1, logger
-
+from helper import list_at_index_0, list_at_index_1
+from logger import cts_log_file_create, logger, cts_logger
+
+# Create csv file for data logging
+log_file_path = cts_log_file_create("flagged")
+
+# Initialise CommitScheduler
+scheduler = CommitScheduler(
+    repo_id="kevkev05/CTS-performance-metrics",
+    repo_type="dataset",
+    folder_path=log_file_path.parent,
+    path_in_repo="data",
+    every=5,
+    private=True,
+)
 
 def chat_to_sequence(sequence, user_query):
     
-    # Log information to a CSV file
-    log_filename = "CTS_user_log.csv"
 
     # Sequence to be analysed/queried
     input_sequence = sequence
 
-    # Set ParameterExtractor class expected variable
+    # Set DNAseq class expected variable
     dna = input_sequence
 
     # Model
@@ -113,23 +126,15 @@ def chat_to_sequence(sequence, user_query):
         proximal_upper_threshold,
     ]
 
-    # Logger Testing Code
-    user_log = pd.read_csv(log_filename)
-    print(user_log.tail(3))
-
-    # Flagged folder test code
-    flagged_log = pd.read_csv("flagged/log.csv")
-    print(flagged_log.tail(2))
-    
     # Check the query score against threshold values
     if query_score >= proximal_upper_threshold:
         response = threshold_exceeded_message
-        logger(log_filename, log_data, response)
+        cts_logger(scheduler, log_file_path, log_data, response)
         print(threshold_exceeded_message)
 
     elif proximal_lower_threshold < query_score < proximal_upper_threshold:
-        response = threshold_approximate_message + "/n" + ref_question
-        logger(log_filename, log_data, response)
+        response = threshold_approximate_message + "\n" + ref_question
+        cts_logger(scheduler, log_file_path, log_data, response)
         print(threshold_approximate_message, ref_question)
     else:
         print("Execute query")
@@ -144,11 +149,11 @@ def chat_to_sequence(sequence, user_query):
             function = matching_row.iloc[0]["function"]
             response = str(eval(function))
             code_descript_message = query_code_description.title()
-            logger(log_filename, log_data, response)
+            cts_logger(scheduler, log_file_path, log_data, response)
         else:
             response = "Error processing query"
             query_code = "No Match Error"
-            logger(log_filename, log_data, response)
+            cts_logger(scheduler, log_file_path, log_data, response)
             print("No matching code found for the function:", code)
 
         return response, code_descript_message
@@ -160,7 +165,7 @@ ChatToSequence = gr.Interface(
     inputs=[gr.Textbox(label="Sequence", placeholder="Input DNA Sequence..."),
             gr.Textbox(label="Query", placeholder="Input Query...")],
     outputs=[gr.Textbox(label="Response"), gr.Textbox(label="Action Executed")],
-    allow_flagging="auto",
+    allow_flagging="never",
     title="Chat-To-Sequence",
     description="This Demo App Allows You To Explore Your DNA Sequence Using Natural Language",
     theme=gr.themes.Soft(),

helper.py

@@ -1,5 +1,3 @@
-import csv
-import datetime
 
 def list_at_index(extracted_list, index):
     value_at_index = extracted_list[index]
@@ -17,18 +15,3 @@ def list_at_index_1(extracted_list):
     value_at_index = extracted_list[1]
 
     return value_at_index
-
-def logger(log_filename, log_data, response):
-    with open(log_filename, mode='a', newline='') as log_file:
-        log_writer = csv.writer(log_file, delimiter=',', quotechar='"', quoting=csv.QUOTE_MINIMAL)
-
-        # Get the current date and time
-        current_datetime = datetime.datetime.now()
-        date_str = current_datetime.strftime("%Y-%m-%d")
-        time_str = current_datetime.strftime("%H:%M:%S")
-        log_data.append(date_str)
-        log_data.append(time_str)
-        log_data.append(response)
-        # Write the log data to the CSV file
-        log_writer.writerow(log_data)
-

logger.py

@@ -0,0 +1,54 @@
+import csv
+import datetime
+import uuid
+from pathlib import Path
+
+
+def cts_logger(scheduler, log_filename, log_data, response) -> None:
+    with scheduler.lock:
+        logger(log_filename, log_data, response)
+
+
+def logger(log_filename, log_data, response):
+    with open(log_filename, mode='a', newline='') as log_file:
+        log_writer = csv.writer(log_file, delimiter=',', quotechar='"', quoting=csv.QUOTE_MINIMAL)
+
+        # Get the current date and time
+        current_datetime = datetime.datetime.now()
+        date_str = current_datetime.strftime("%Y-%m-%d")
+        time_str = current_datetime.strftime("%H:%M:%S")
+        log_data.append(date_str)
+        log_data.append(time_str)
+        log_data.append(response)
+
+        # Write the log data to the CSV file
+        log_writer.writerow(log_data)
+
+
+def cts_log_file_create(folder_name):
+    file_path = Path(folder_name + "/") / f"CTS_user_log{uuid.uuid4()}.csv"
+    with open(file_path, mode='a', newline='') as log_file:
+        log_writer = csv.writer(log_file, delimiter=',', quotechar='"', quoting=csv.QUOTE_MINIMAL)
+
+        headers = [
+            "user_query",
+            "ref_question",
+            "query_score",
+            "code_executed",
+            "ref_question_2",
+            "query_score_2",
+            "ref_question_3",
+            "query_score_3",
+            "similarity_metric",
+            "model_used_for_embeddings",
+            "lower_threshold",
+            "upper_threshold",
+            "date",
+            "time",
+            "response",
+        ]
+
+        # Write headers to the CSV file
+        log_writer.writerow(headers)
+
+        return file_path