Update app.py

app.py

@@ -69,29 +69,29 @@ def analyze_data(data, visualization_type, class_size=10):
     # Combine numerical and non-numerical summaries
     numeric_data = data.select_dtypes(include=[np.number])
 
-
-    # Visualization logic
-    if visualization_type == "Heatmap" and not numeric_data.empty:
-        st.subheader("Correlation Heatmap")
-        fig, ax = plt.subplots(figsize=(8, 6))
-        sns.heatmap(numeric_data.corr(), annot=True, ax=ax, cmap="coolwarm", fmt=".2f")
-        st.pyplot(fig)
-
-    elif visualization_type == "Bar Chart" and not numeric_data.empty:
+    if visualization_type == "Bar Chart" and not numeric_data.empty:
         st.subheader("Bar Chart")
-        x_col = st.selectbox("Select the X-axis column for the Bar Chart:", data.columns)
-        y_col = st.selectbox("Select the Y-axis column for the Bar Chart:", data.columns)
+        x_col = st.selectbox("Select the X-axis column for the Bar Chart (Non-Numeric):", data.columns)
+        y_col = st.selectbox("Select the Y-axis column for the Bar Chart (Numeric):", data.columns)
         
         fig, ax = plt.subplots(figsize=(8, 6))
         data.groupby(x_col)[y_col].sum().plot(kind='bar', ax=ax)
         ax.set_xlabel(x_col)
         ax.set_ylabel(y_col)
         st.pyplot(fig)
+    # Visualization logic
+    elif visualization_type == "Heatmap" and not numeric_data.empty:
+        st.subheader("Correlation Heatmap")
+        fig, ax = plt.subplots(figsize=(8, 6))
+        sns.heatmap(numeric_data.corr(), annot=True, ax=ax, cmap="coolwarm", fmt=".2f")
+        st.pyplot(fig)
+
+
 
     elif visualization_type == "Line Graph" and not numeric_data.empty:
         st.subheader("Line Graph")
-        x_col = st.selectbox("Select the X-axis column for the Line Graph:", numeric_data.columns)
-        y_col = st.selectbox("Select the Y-axis column for the Line Graph:", numeric_data.columns)
+        x_col = st.selectbox("Select the X-axis column for the Line Graph (Non-Numeric):", numeric_data.columns)
+        y_col = st.selectbox("Select the Y-axis column for the Line Graph (Numeric):", numeric_data.columns)
 
         fig, ax = plt.subplots(figsize=(8, 6))
         ax.plot(data[x_col], data[y_col])
@@ -99,25 +99,8 @@ def analyze_data(data, visualization_type, class_size=10):
         ax.set_ylabel(y_col)
         st.pyplot(fig)
 
-    elif visualization_type == "Scatter Plot" and not numeric_data.empty:
-        st.subheader("Scatter Plot")
-        x_col = st.selectbox("Select the X-axis column for the Scatter Plot:", numeric_data.columns)
-        y_col = st.selectbox("Select the Y-axis column for the Scatter Plot:", numeric_data.columns)
 
-        fig, ax = plt.subplots(figsize=(8, 6))
-        ax.scatter(data[x_col], data[y_col])
-        ax.set_xlabel(x_col)
-        ax.set_ylabel(y_col)
-        st.pyplot(fig)
 
-    elif visualization_type == "Histogram" and not numeric_data.empty:
-        st.subheader("Histogram")
-        column = st.selectbox("Select a column for the Histogram:", numeric_data.columns)
-        fig, ax = plt.subplots(figsize=(8, 6))
-        data[column].plot(kind='hist', bins=class_size, ax=ax)
-        ax.set_xlabel(column)
-        ax.set_ylabel("Frequency")
-        st.pyplot(fig)
 
     elif visualization_type == "Area Chart" and not numeric_data.empty:
         st.subheader("Area Chart")
@@ -137,11 +120,12 @@ def analyze_data(data, visualization_type, class_size=10):
 
 # Function to generate a prompt based on the data analysis
 def generate_groq_prompt(data, visualization_type, class_size):
-    # Convert DataFrame to a string without the index
-    data_without_index = data.to_string(index=False)
-    
+    # Convert a small sample of the DataFrame to a string without the index
+    sample_data = data.head(10)  # Include only the first 10 rows for clarity
+    data_without_index = sample_data.to_string(index=False)
+
     prompt = f"""
-    Here is the summary statistics for the dataset:
+    Here is the summary statistics for the dataset (showing a sample of 10 rows for clarity):
     {data_without_index}
 
     The user has selected the '{visualization_type}' visualization type with a class size of {class_size}.
@@ -162,7 +146,7 @@ if uploaded_file is not None:
         if data is not None:
             data = preprocess_dataframe(data)  # Fix serialization issues
             st.subheader("Uploaded Data")
-            st.write(data.head())
+            st.dataframe(data)  # Show raw data in tabular form
 
             # Visualization Selection
             visualization_type = st.selectbox(
@@ -175,7 +159,10 @@ if uploaded_file is not None:
 
             # Perform Analysis and Visualization
             prompt = analyze_data(data, visualization_type, class_size)
-            st.text(f"Prompt sent to Groq:\n{prompt}")
+            
+            # Display prompt sent to Groq with preserved formatting
+            st.subheader("Prompt Sent to Groq")
+            st.text_area("Preview of Prompt", prompt, height=200)
 
             # Chat with Groq Section
             st.subheader("Chat with Groq")