feat: joblib cache (#749)

* cache function

* fix test

* bin cache

* fix test

* fix test

* fix test

* cache for different source

* cache for localenv

* remove unnecessary log

* reformat

* remove unrelated modify

rdagent/components/coder/data_science/raw_data_loader/eval_tests/data_loader_test.txt

@@ -9,12 +9,22 @@ from load_data import load_data
 
 import sys
 import reprlib
+from joblib.memory import MemorizedFunc
+
+
+def get_original_code(func):
+    if isinstance(func, MemorizedFunc):
+        return func.func.__code__
+    return func.__code__
+
+
 def debug_info_print(func):
     aRepr = reprlib.Repr()
     aRepr.maxother=300
     def wrapper(*args, **kwargs):
+        original_code = get_original_code(func)
         def local_trace(frame, event, arg):
-            if event == "return" and frame.f_code == func.__code__:
+            if event == "return" and frame.f_code == original_code:
                 print("\n" + "="*20 + "Running data_load code, local variable values:" + "="*20)
                 for k, v in frame.f_locals.items():
                     printed = aRepr.repr(v)

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

@@ -334,6 +334,11 @@ data_loader_coder:
     ## Guidelines
     1. Ensure that the dataset is loaded strictly from `/kaggle/input/`, following the exact folder structure described in the **Data Folder Description**, and do not attempt to load data from the current directory (`./`).
     2. You should avoid using logging module to output information in your generated code, and instead use the print() function.
+    3. You should use the following cache decorator to cache the results of the function:
+    ```python
+    from joblib import Memory
+    memory = Memory(location='/tmp/cache', verbose=0)
+    @memory.cache```
     
     ## Exploratory Data Analysis (EDA) part(Required):
     - Before returning the data, you should always add an EDA part describing the data to help the following steps understand the data better.