Update app.py

app.py

@@ -20,11 +20,6 @@ login(token=hf_token)
 
 # SmolAgent initialization
 model = HfApiModel("mistralai/Mixtral-8x7B-Instruct-v0.1", token=hf_token)
-agent = CodeAgent(
-    tools=[],
-    model=model,
-    additional_authorized_imports=["numpy", "pandas", "matplotlib.pyplot", "seaborn", "sklearn"]
-)
 
 df_global = None
 
@@ -48,41 +43,43 @@ def upload_file(file):
 
 import textwrap
 
-def run_agent(_):
-    try:
-        # Check if df_global is not None, meaning a file was uploaded and cleaned
-        if df_global is None:
-            return "Error: No dataset uploaded."
-
-        # Pass the dataset to the agent and generate insights/visualizations
-        prompt = """You are an expert data scientist.
-Please generate 5 data insights and 5 data visualizations for the provided dataset.
-Visualizations should be saved in the current working directory."""
-
-        # Call the agent's run method, passing the dataset as input
-        output = agent.run(
-            prompt,  # Pass the instructions as a string
-            additional_args=dict(dataset=df_global.to_dict(), additional_notes="Additional notes or analysis can go here.")
-        )
-
-        # Print or log the output to inspect it
-        print("SmolAgent Output:", output)
-
-        # Clean and fix indentation issues
-        if isinstance(output, str):
-            cleaned_output = textwrap.dedent(output)  # This will remove leading indentation
-            exec(cleaned_output)
-            return "Insights and visualizations have been generated successfully."
-        else:
-            return f"Unexpected output format: {type(output)}. The output is not a string."
 
-    except Exception as e:
-        return f"SmolAgent Error: {str(e)}"
+# Initialize the agent
+agent = CodeAgent(
+    tools=[],
+    model=model,
+    additional_authorized_imports=["numpy", "pandas", "matplotlib.pyplot", "seaborn"]
+)
+
+# Define the analysis instructions
+analysis_result = agent.run(
+    """
+    You are an expert data analyst. Perform comprehensive analysis including:
+    1. Basic statistics and data quality checks.
+    2. Three insightful analytical questions about relationships in the data.
+    3. Visualization of key patterns and correlations.
+    4. Actionable real-world insights derived from findings.
+    Generate publication-quality visualizations and save them to './figures/'.
+    """,
+    additional_args={
+        "additional_notes": additional_notes,  # Any additional analysis notes
+        "source_file": csv_file  # The path to the input CSV file
+    }
+)
+
+# Optionally, you can inspect the results
+print(analysis_result)
 
 
 def train_model(_):
     wandb.login(key=os.environ.get("WANDB_API_KEY"))
-    wandb_run = wandb.init(project="huggingface-data-analysis", name="Optuna_Run", reinit=True)
+    #wandb_run = wandb.init(project="huggingface-data-analysis", name="Optuna_Run", reinit=True)
+    # At the start of your script
+    run_counter = 1
+
+    # Then when initializing
+    wandb_run = wandb.init(project="huggingface-data-analysis", name=f"Optuna_Run_{run_counter}", reinit=True)
+    run_counter += 1
 
     target = df_global.columns[-1]
     X = df_global.drop(target, axis=1)