refactor: etc

core/handlers/base.py

@@ -1,10 +1,13 @@
 import os
 import uuid
 from enum import Enum
+from pathlib import Path
 from typing import Dict
 
 import requests
 
+from env import settings
+
 
 class FileType(Enum):
     IMAGE = "image"
@@ -64,6 +67,7 @@ class FileHandler:
         local_filename = os.path.join(
             filetype.value, str(uuid.uuid4())[0:8] + filetype.to_extension()
         )
+        os.makedirs(os.path.dirname(local_filename), exist_ok=True)
         with open(local_filename, "wb") as f:
             size = f.write(data)
         print(f"Inputs: {url} ({size//1000}MB)  => {local_filename}")

core/handlers/dataframe.py

@@ -9,7 +9,7 @@ class CsvToDataframe(BaseHandler):
     def handle(self, filename: str):
         df = pd.read_csv(filename)
         description = (
-            f"Dataframe with {len(df)} rows and {len(df.columns)} columns."
+            f"Dataframe with {len(df)} rows and {len(df.columns)} columns. "
             "Columns are: "
             f"{', '.join(df.columns)}"
         )

core/prompts/file.py

@@ -2,6 +2,9 @@ IMAGE_PROMPT = """
 provide a figure named {filename}. The description is: {description}.
 
 Please understand and answer the image based on this information. The image understanding is complete, so don't try to understand the image again.
+
+USER INPUT
+============
 """
 
 
@@ -9,12 +12,18 @@ AUDIO_PROMPT = """
 provide a audio named {filename}. The description is: {description}.
 
 Please understand and answer the audio based on this information. The audio understanding is complete, so don't try to understand the audio again.
+
+USER INPUT
+============
 """
 
 VIDEO_PROMPT = """
 provide a video named {filename}. The description is: {description}.
 
 Please understand and answer the video based on this information. The video understanding is complete, so don't try to understand the video again.
+
+USER INPUT
+============
 """
 
 DATAFRAME_PROMPT = """
@@ -22,4 +31,7 @@ provide a dataframe named {filename}. The description is: {description}.
 
 You are able to use the dataframe to answer the question.
 You have to act like an data analyst who can do an effective analysis through dataframe.
+
+USER INPUT
+============
 """

core/prompts/input.py

@@ -1,7 +1,7 @@
 EVAL_PREFIX = """{bot_name} can execute any user's request.
 
-{bot_name} has permission to handle one instance and can handle the environment in it at will. File creation is only possible in playground folder, and other folders and files can never be modified under any circumstances. 
-You can code, run, debug, and test yourself. You can correct the code appropriately by looking at the error message. Please do TDD.
+{bot_name} has permission to handle one instance and can handle the environment in it at will.
+You can code, run, debug, and test yourself. You can correct the code appropriately by looking at the error message.
 
 I can understand, process, and create various types of files. Images must be stored in the ./image/, audio in the ./audio/, video in the ./video/, and dataframes must be stored in the ./dataframe/.
 

utils.py

@@ -4,12 +4,6 @@ import uuid
 
 import numpy as np
 
-os.makedirs("image", exist_ok=True)
-os.makedirs("audio", exist_ok=True)
-os.makedirs("video", exist_ok=True)
-os.makedirs("dataframe", exist_ok=True)
-os.makedirs("playground", exist_ok=True)
-
 
 def seed_everything(seed):
     random.seed(seed)
@@ -23,16 +17,6 @@ def seed_everything(seed):
         pass
     return seed
 
-
-def prompts(name, description):
-    def decorator(func):
-        func.name = name
-        func.description = description
-        return func
-
-    return decorator
-
-
 def cut_dialogue_history(history_memory, keep_last_n_words=500):
     tokens = history_memory.split()
     n_tokens = len(tokens)