several bug fix (#621)

rdagent/components/coder/data_science/feature/eval_tests/feature_test.txt

@@ -63,13 +63,15 @@ if isinstance(X, pd.DataFrame) and isinstance(X_test, pd.DataFrame):
 if isinstance(X, pd.DataFrame):
     X_dtypes_unique_sorted = sorted([str(dt) for dt in X.dtypes.unique()])
     X_loaded_dtypes_unique_sorted = sorted([str(dt) for dt in X_loaded.dtypes.unique()])
+
+    X_dtypes_unique_sorted_new = [
+        dt for dt in X_dtypes_unique_sorted if dt not in X_loaded_dtypes_unique_sorted and dt != "object"
+    ]
     assert (
-        len(X_loaded_dtypes_unique_sorted) == 1
-        and X_loaded_dtypes_unique_sorted[0] in {np.float64, np.float32}
-    ) or (
-        X_dtypes_unique_sorted == X_loaded_dtypes_unique_sorted
+        np.dtypes.ObjectDType in X_loaded_dtypes_unique_sorted or len(X_dtypes_unique_sorted_new) == 0
     ), f"feature engineering has produced new data types which is not allowed, data loader data types are {X_loaded_dtypes_unique_sorted} and feature engineering data types are {X_dtypes_unique_sorted}"
 
+
 print(
     "Feature Engineering test passed successfully. All checks including length, width, and data types have been validated."
 )

rdagent/components/coder/data_science/model/prompts.yaml

@@ -37,6 +37,7 @@ model_coder:
     {{ feature_code }}
     2. You should avoid using logging module to output information in your generated code, and instead use the print() function.
     3. You can decide whether to use AutoML based on the characteristics of the task.
+    4. If the model can both be implemented by PyTorch and Tensorflow, please use pytorch for broader compatibility.
 
     ## Output Format
     {% if out_spec %}