Update LLMasjudge.py

LLMasjudge.py

@@ -0,0 +1,36 @@
+import streamlit as st
+from dotenv import load_dotenv
+
+import phoenix_helpers
+import helpers 
+
+load_dotenv()
+
+if "eval_btn_clicked" not in st.session_state:
+        st.session_state.eval_btn_clicked = False
+
+def callback2():
+    st.session_state.eval_btn_clicked = True
+
+models = helpers.fetch_models()
+if models:
+    st.subheader("Evaluate LLM")
+
+    if "evaluation_result" not in st.session_state:
+        st.session_state.evaluation_result = None
+    if "eval_model" not in st.session_state:
+        st.session_state.eval_model = models[0]
+    st.selectbox(
+        "Choose a model to use for evaluation:", 
+        models, key = 'eval_model',
+        index=models.index(st.session_state.eval_model) if st.session_state.eval_model in models else 0,
+    )
+    if st.button("Evaluate", on_click=callback2):
+        if st.session_state.eval_model:
+            try:
+                st.session_state.evaluation_result = phoenix_helpers.evaluate_model(st.session_state.eval_model)
+                st.dataframe(st.session_state.evaluation_result)
+            except Exception as e:
+                st.write(e)
+
+