Added removing conversations and also retreiving e

Author: codeastra2

main.py

@@ -1,13 +1,104 @@
 import openai
-openai.api_key = "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"
-
-response = openai.ChatCompletion.create(
-    model="gpt-3.5-turbo",
-    messages=[
-            {"role": "system", "content": "You are useful assistant"},
-            {"role": "user", "content": "Can you tell top 10 startups in India"},
-		{"role": "assistant",   "content": "1. Flipkart \n2. Ola Cabs \n3. Paytm \n4. BYJU's \n5. Swiggy \n6. Zomato \n7. Udaan \n8. Freshworks \n9. Nykaa \n10. Policybazaar"},
-		 {"role": "user", "content": "That is good but can you tell top 10 early stage strtups? some of these re late stage...."},
-		
-        ]
-)
+import time
+from functools import wraps
+import requests
+
+
+def retry(tries=4, delay=3, backoff=2):
+    """
+    Retries a function or method until it succeeds or the number of retries is exceeded.
+
+    :param tries: the maximum number of times to retry (default 4).
+    :param delay: the initial delay between retries in seconds (default 3).
+    :param backoff: the backoff multiplier (e.g. value of 2 will double the delay each retry) (default 2).
+    """
+
+    def deco_retry(func):
+        @wraps(func)
+        def f_retry(*args, **kwargs):
+            mtries, mdelay = tries, delay
+            while mtries > 1:
+                try:
+                    return func(*args, **kwargs)
+                except (
+                    openai.error.APIError,
+                    requests.exceptions.RequestException,
+                ) as e:
+                    print(f"Error occurred: {str(e)}")
+                    time.sleep(mdelay)
+                    mtries -= 1
+                    mdelay *= backoff
+            return func(*args, **kwargs)
+
+        return f_retry
+
+    return deco_retry
+
+
+def dot_product(list1, list2):
+    # Check if the lengths of the two lists are equal
+    if len(list1) != len(list2):
+        raise ValueError("Lists must have same length")
+
+    # Calculate the dot product using a loop
+    dot_product = 0
+    for i in range(len(list1)):
+        dot_product += list1[i] * list2[i]
+
+    return dot_product
+
+
+class ChatGptSmartClient(object):
+    """
+    This is a wrapper class for the chatgpt python api,
+    it is meant to provide developers a smooth expereince in
+    developing chatgpt applications of their own without the need
+    for worrying about things like retries, tracking message history
+    or storing messages. This runs on top of the Chat APIs provided by OpenAI
+    read more on that here: https://platform.openai.com/docs/guides/chat .
+    """
+
+    def __init__(self, api_key: str, model: str):
+        openai.api_key = api_key
+
+        self.instruction_msgs = {
+            "role": "system",
+            "content": "You are a  useful assistant",
+        }
+        self.prev_msgs = [self.instruction_msgs]
+        self.model = model
+
+    @retry()
+    def query(self, query: str, w_context=True, add_to_context=True):
+        # TODO: We coud get the embeddings, cache and further use them to speed up the results.
+        # self.get_embeddings(query=query)
+
+        query = {"role": "user", "content": query}
+
+        if w_context:
+            msgs = self.prev_msgs[:]
+            msgs.append(query)
+            response = openai.ChatCompletion.create(model=self.model, messages=msgs)
+        else:
+            msgs = [self.instruction_msgs, query]
+            response = openai.ChatCompletion.create(model=self.model, messages=msgs)
+
+        f_resp = response["choices"][0]["message"]
+
+        if add_to_context:
+            self.prev_msgs.append(f_resp)
+
+        return f_resp
+
+    def erase_history(self):
+        self.prev_msgs = [self.instruction_msgs]
+
+    # This function is used for getting embeddings and hence maybe
+    # used to speedup the system by caching.
+    def get_embeddings(self, query: str):
+        response = openai.Embedding.create(input=query, model="text-embedding-ada-002")
+        embeddings = response["data"][0]["embedding"]
+        print(len(embeddings))
+
+    def rollback_conversation(self):
+        pass