fixed bug for updating dashboard and added preliminary logs

app.py

@@ -17,6 +17,7 @@ import plotly.graph_objects as go
 import boto3
 import json
 from plotly.subplots import make_subplots
+from logger import logger
 
 fs = fsspec.filesystem(
         's3',
@@ -37,6 +38,42 @@ EMAIL_PATH = f"s3://{os.getenv('AWS_BUCKET_NAME')}/{os.getenv('EMAILS_KEY')}"
 TEMP_DIR = f"s3://{os.getenv('AWS_BUCKET_NAME')}/{os.getenv('AUDIOS_KEY')}"
 CREATE_TASK_URL = os.getenv("CREATE_TASK_URL")
 
+def create_files():
+    if not fs.exists(SAVE_PATH):
+        logger.info("Creating save file")
+        with fs.open(SAVE_PATH, 'wb') as f:
+            headers = [
+            'email',
+            'path',
+            'Ori Apex_score', 'Ori Apex XT_score', 'deepgram_score', 'Ori Swift_score', 'Ori Prime_score',
+            'Ori Apex_appearance', 'Ori Apex XT_appearance', 'deepgram_appearance', 'Ori Swift_appearance', 'Ori Prime_appearance',
+            'Ori Apex_duration', 'Ori Apex XT_duration', 'deepgram_duration', 'Ori Swift_duration', 'Ori Prime_duration','azure_score','azure_appearance','azure_duration'
+        ]
+            df = pd.DataFrame(columns=headers)
+            df.to_csv(f, index=False)
+
+    if not fs.exists(ELO_JSON_PATH):
+        logger.info("Creating Elo json file")
+        with fs.open(ELO_JSON_PATH, 'w') as f:
+            models = ['Ori Apex', 'Ori Apex XT', 'deepgram', 'Ori Swift', 'Ori Prime', 'azure']
+            models = {model: 1000 for model in models}
+            json.dump(models, f)
+
+    if not fs.exists(ELO_CSV_PATH):
+        logger.info("Creating Elo csv file")
+        with fs.open(ELO_CSV_PATH, 'wb') as f:
+            models = ['Ori Apex', 'Ori Apex XT', 'deepgram', 'Ori Swift', 'Ori Prime', 'azure']
+            models = {k:1000 for k in models}
+            df = pd.DataFrame(models,index=[0])
+            df.to_csv(f, index=False)
+
+    if not fs.exists(EMAIL_PATH):
+        logger.info("Creating email file")
+        with fs.open(EMAIL_PATH, 'wb') as f:
+            existing_content = ''
+            new_content = existing_content
+            with fs.open(EMAIL_PATH, 'w') as f:
+                f.write(new_content.encode('utf-8'))
 
 def write_email(email):
         if fs.exists(EMAIL_PATH):
@@ -96,8 +133,6 @@ class ResultWriter:
 
         send_task(payload)
 
-
-
 def decode_audio_array(base64_string):
     bytes_data = base64.b64decode(base64_string)
 
@@ -113,7 +148,9 @@ def send_task(payload):
     response = requests.post(CREATE_TASK_URL,json=payload,headers=header)
     try:
         response = response.json()
-    except Exception:
+    except Exception as e:
+        logger.error("Error while sending task %s",e)
+        logger.debug("Payload which caused the error %s",payload)
         return "error please try again"
 
     if payload["task"] == "transcribe_with_fastapi":
@@ -160,21 +197,21 @@ def call_function(model_name):
     return transcript
 
 def transcribe_audio():
+    with st.spinner("🎯 Transcribing audio... this may take up to 30 seconds"):
+        models_list = ["Ori Apex", "Ori Apex XT", "deepgram", "Ori Swift", "Ori Prime","azure"]
+        model1_name, model2_name = random.sample(models_list, 2)
 
-    models_list = ["Ori Apex", "Ori Apex XT", "deepgram", "Ori Swift", "Ori Prime","azure"]
-    model1_name, model2_name = random.sample(models_list, 2)
-
-    st.session_state.option_1_model_name = model1_name
-    st.session_state.option_2_model_name = model2_name
+        st.session_state.option_1_model_name = model1_name
+        st.session_state.option_2_model_name = model2_name
 
-    time_1 = time.time()
-    transcript1 = call_function(model1_name)
-    time_2 = time.time()
-    transcript2 = call_function(model2_name)
-    time_3 = time.time()
+        time_1 = time.time()
+        transcript1 = call_function(model1_name)
+        time_2 = time.time()
+        transcript2 = call_function(model2_name)
+        time_3 = time.time()
 
-    st.session_state.option_2_response_time = round(time_3 - time_2,3)
-    st.session_state.option_1_response_time = round(time_2 - time_1,3)
+        st.session_state.option_2_response_time = round(time_3 - time_2,3)
+        st.session_state.option_1_response_time = round(time_2 - time_1,3)
 
 
     return transcript1, transcript2
@@ -429,20 +466,23 @@ def create_metric_container(label, value, full_name=None):
             st.markdown(f"<h3 style='margin-top: 0;'>{value}</h3>", unsafe_allow_html=True)
 
 def on_refresh_click():
-    with fs.open(SAVE_PATH, 'rb') as f:
-        st.session_state.df = pd.read_csv(f)
+    with st.spinner("Refreshing data... please wait"):
+        with fs.open(SAVE_PATH, 'rb') as f:
+            st.session_state.df = pd.read_csv(f)
 
-    try:
-        with fs.open(ELO_JSON_PATH,'r') as f:
-            st.session_state.elo_json = json.load(f)
-    except Exception:
-        st.session_state.elo_json = None
+        try:
+            with fs.open(ELO_JSON_PATH,'r') as f:
+                st.session_state.elo_json = json.load(f)
+        except Exception as e:
+            logger.error("Error while reading elo json file %s",e)
+            st.session_state.elo_json = None
 
-    try:
-        with fs.open(ELO_CSV_PATH,'rb') as f:
-            st.session_state.elo_df = pd.read_csv(f)
-    except Exception:
-        st.session_state.elo_df = None
+        try:
+            with fs.open(ELO_CSV_PATH,'rb') as f:
+                st.session_state.elo_df = pd.read_csv(f)
+        except Exception as e:
+            logger.error("Error while reading elo csv file %s",e)
+            st.session_state.elo_df = None
 
 def dashboard():
     st.title('Model Arena Scoreboard')
@@ -451,17 +491,13 @@ def dashboard():
         with fs.open(SAVE_PATH, 'rb') as f:
             st.session_state.df = pd.read_csv(f)
     if "elo_json" not in st.session_state:
-        try:
-            with fs.open(ELO_JSON_PATH,'r') as f:
-                st.session_state.elo_json = json.load(f)
-        except Exception:
-            st.session_state.elo_json = None
+        with fs.open(ELO_JSON_PATH,'r') as f:
+            elo_json = json.load(f)
+            st.session_state.elo_json = elo_json
     if "elo_df" not in st.session_state:
-        try:
-            with fs.open(ELO_CSV_PATH,'rb') as f:
-                st.session_state.elo_df = pd.read_csv(f)
-        except Exception:
-            st.session_state.elo_df = None
+        with fs.open(ELO_CSV_PATH,'rb') as f:
+            elo_df = pd.read_csv(f)
+            st.session_state.elo_df = elo_df
 
     st.button("Refresh",on_click=on_refresh_click)
 
@@ -527,7 +563,7 @@ def dashboard():
         appearance_chart = create_appearance_chart(metrics)
         st.plotly_chart(appearance_chart, use_container_width=True)
 
-        if st.session_state.elo_json and st.session_state.elo_df:
+        if st.session_state.elo_json is not None and st.session_state.elo_df is not None:
             st.header('Elo Ratings')
             st.dataframe(pd.DataFrame(st.session_state.elo_json,index=[0]),use_container_width=True)
             elo_progression_chart = create_elo_chart(st.session_state.elo_df)
@@ -714,7 +750,7 @@ def main():
             st.button("Prefer None",on_click=on_option_none_click)
 
     with scoreboard:
-        if st.session_state.logged_in:
+        if st.session_state.logged_in or os.getenv("IS_TEST"):
             dashboard()
         else:
             with st.form("contact_us_form"):
@@ -741,4 +777,5 @@ def main():
     with about_tab:
         about()
 
+create_files()
 main()

logger.py

@@ -0,0 +1,14 @@
+import logging
+import os
+
+loglevel = os.getenv("LOGLEVEL", "INFO")
+
+
+logger = logging.getLogger("App")
+logger.setLevel(loglevel)
+
+console_formatter = logging.Formatter("%(name)s: %(levelname)s -> [%(filename)s:%(lineno)s - %(funcName)s] %(message)s")
+
+console_handler = logging.StreamHandler()
+console_handler.setFormatter(console_formatter)
+logger.addHandler(console_handler)