fix docs

Signed-off-by: Gilad Barnea <[email protected]>

README.md

@@ -45,23 +45,23 @@ pip install declarai
 
 ### Setup
 ```bash
-export DECLARAI_OPENAI_API_KEY=<your openai token>
+export OPENAI_API_KEY=<your openai token>
 ```
 or pass the token when initializing the declarai object
 ```python
-from declarai import Declarai
+import declarai
 
-declarai = Declarai(provider="openai", model="gpt-3.5-turbo", openai_token="<your-openai-key>")
+gpt_35 = declarai.openai(model="gpt-3.5-turbo", openai_token="<your-openai-key>")
 ```
 
 ## 💡 Basic Usage
 Craft AI-powered functionalities with ease using the `@task` decorator. Just add some type hints and a bit of documentation, and watch Declarai do its magic!
 ```python
 import declarai
 
-openai = declarai.openai(model="gpt-3.5-turbo")
+gpt_35 = declarai.openai(model="gpt-3.5-turbo")
 
-@openai.task
+@gpt_35.task
 def generate_poem(title: str) -> str:
     """
     Write a 4 line poem on the provided title
@@ -96,7 +96,11 @@ The resulting code is readable and easily maintainable.
 
 Python primitives
 ```python
-@openai.task
+import declarai
+
+gpt_35 = declarai.openai(model="gpt-3.5-turbo")
+
+@gpt_35.task
 def rank_by_severity(message: str) -> int:
     """
     Rank the severity of the provided message by it's urgency.
@@ -116,7 +120,11 @@ rank_by_severity(message="How was your weekend?"))
 
 Python complex objects
 ```python
-@openai.task
+import declarai
+
+gpt_35 = declarai.openai(model="gpt-3.5-turbo")
+
+@gpt_35.task
 def datetime_parser(raw_date: str) -> datetime:
     """
     Parse the input into a valid datetime string of the format YYYY-mm-ddThh:mm:ss
@@ -132,13 +140,14 @@ datetime_parser(raw_date="Janury 1st 2020"))
 
 pydantic models
 ```python
+from pydantic import BaseModel
 class Animal(BaseModel):
     name: str
     family: str
     leg_count: int
 
 
-@openai.task
+@gpt_35.task
 def suggest_animals(location: str) -> Dict[int, List[Animal]]:
     """
     Create a list of numbers from 0 to 5
@@ -168,7 +177,7 @@ suggest_animals(location="jungle")
 
 ### Simple Chat interface
 ```python
-@openai.experimental.chat
+@gpt_35.experimental.chat
 class CalculatorBot:
     """
     You a calculator bot,

docs/beginners-guide/controlling-task-behavior.md

@@ -11,20 +11,23 @@ Controlling these parameters is key to achieving the desired results from the mo
 ### Passing parameters to the task :label:
 
 In the following example, we'll create a task that suggests movies to watch based on a given input.
+
 ```python
 import declarai
 
-openai = declarai.openai(model="gpt-3.5-turbo")
+gpt_35 = declarai.openai(model="gpt-3.5-turbo")
+
 
-@openai.task
-def movie_recommender(user_input: str): # (1)!
+@gpt_35.task
+def movie_recommender(user_input: str):  # (1)!
     """
     Recommend a movie to watch based on the user input
     :param user_input: The user's input
-    """ # (2)!
+    """  # (2)!
 ```
 
-1. Notice how providing a type hint for the `user_input` parameter allows declarai to understand the expected input type.
+1. Notice how providing a type hint for the `user_input` parameter allows declarai to understand the expected input
+   type.
 2. Adding the param to the docstring allows declarai to communicate the **meaning** of this parameter to the model.
 
 ```python
@@ -33,71 +36,77 @@ print(movie_recommender(user_input="I want to watch a movie about space"))
 ```
 
 ### Using return types to control the output :gear:
-This is a good start, 
+
+This is a good start,
 but let's say we want to have a selection of movies instead of a single suggestion.
+
 ```python
 from typing import List
 import declarai
 
-openai = declarai.openai(model="gpt-3.5-turbo")
-@openai.task
-def movie_recommender(user_input: str) -> List[str]: # (1)!
+gpt_35 = declarai.openai(model="gpt-3.5-turbo")
+
+
+@gpt_35.task
+def movie_recommender(user_input: str) -> List[str]:  # (1)!
     """
     Recommend a selection of movies to watch based on the user input
     :param user_input: The user's input
     :return: A list of movie recommendations
-    """ # (2)!
+    """  # (2)!
 ```
 
-1. Adding a return type hint allows declarai to parse the output of the llm into the provided type, 
+1. Adding a return type hint allows declarai to parse the output of the llm into the provided type,
    in our case a list of strings.
 2. Explaining the return value aids the model in returning the expected output and avoiding hallucinations.
 
 ```python
 print(movie_recommender(user_input="I want to watch a movie about space"))
-> ['Interstellar', 'Gravity', 'The Martian', 'Apollo 13', '2001: A Space Odyssey', 'Moon', 'Sunshine', 'Contact', 'The Right Stuff', 'Hidden Figures']
+> ['Interstellar', 'Gravity', 'The Martian', 'Apollo 13', '2001: A Space Odyssey', 'Moon', 'Sunshine', 'Contact',
+   'The Right Stuff', 'Hidden Figures']
 ```
 
-
 !!! info
 
     Notice How the text in our documentation has changed from singular to plural form. 
     Maintaining consistency between the task's description and the return type is important for the model to understand the expected output.<br>
     For more best-practices, see [here](../../best-practices). 
 
-
-Awesome! 
+Awesome!
 
 Now we have a list of movies to choose from!
 
 But what if we want to go even further :thinking:? <br>
 Let's say we want the model to also provide a short description of each movie.
+
 ```python
 from typing import Dict
 import declarai
 
-openai = declarai.openai(model="gpt-3.5-turbo")
-@openai.task
-def movie_recommender(user_input: str) -> Dict[str, str]: # (1)!
+gpt_35 = declarai.openai(model="gpt-3.5-turbo")
+
+
+@gpt_35.task
+def movie_recommender(user_input: str) -> Dict[str, str]:  # (1)!
     """
     Recommend a selection of movies to watch based on the user input
     For each movie provide a short description as well
     :param user_input: The user's input
     :return: A dictionary of movie names and descriptions
-    """ # (2)!
+    """  # (2)!
 ```
 
 1. We've updated the return value to allow for the creation of a dictionary of movie names and descriptions.
 2. We re-enforce the description of the return value to ensure the model understands the expected output.
 
 ```python
 print(movie_recommender(user_input="I want to watch a movie about space"))
->{
-   'Interstellar': "A team of explorers travel through a wormhole in space in an attempt to ensure humanity's survival.", 
-   'Gravity': 'Two astronauts work together to survive after an accident leaves them stranded in space.', 
-   'The Martian': 'An astronaut is left behind on Mars after his team assumes he is dead and must find a way to survive and signal for rescue.', 
-   'Apollo 13': 'The true story of the Apollo 13 mission, where an explosion in space jeopardizes the lives of the crew and their safe return to Earth.', 
-   '2001: A Space Odyssey': "A journey through human evolution and the discovery of a mysterious black monolith that may hold the key to humanity's future."
+> {
+    'Interstellar': "A team of explorers travel through a wormhole in space in an attempt to ensure humanity's survival.",
+    'Gravity': 'Two astronauts work together to survive after an accident leaves them stranded in space.',
+    'The Martian': 'An astronaut is left behind on Mars after his team assumes he is dead and must find a way to survive and signal for rescue.',
+    'Apollo 13': 'The true story of the Apollo 13 mission, where an explosion in space jeopardizes the lives of the crew and their safe return to Earth.',
+    '2001: A Space Odyssey': "A journey through human evolution and the discovery of a mysterious black monolith that may hold the key to humanity's future."
 }
 ```
 
@@ -111,10 +120,6 @@ print(movie_recommender(user_input="I want to watch a movie about space"))
     Try experimenting with various descriptions and see how far you can push the model's understanding!
     who knows what you'll find :open_mouth:!
 
-
-
-
-
 <div style="display: flex; justify-content: space-between;">
     <a href="../simple-task" class="md-button">
         Previous <i class="fas fa-arrow-left"></i>

docs/beginners-guide/debugging-tasks.md

@@ -17,9 +17,9 @@ Let's take the last task from the previous section and add a call to the `compil
 from typing import Dict
 import declarai
 
-openai = declarai.openai(model="gpt-3.5-turbo")
+gpt_35 = declarai.openai(model="gpt-3.5-turbo")
 
-@openai.task
+@gpt_35.task
 def movie_recommender(user_input: str) -> Dict[str, str]:
     """
     Recommend a selection of movies to watch based on the user input

docs/beginners-guide/index.md

@@ -31,7 +31,7 @@ Once completed, the rest of the examples in this module should be as simple as c
 ```python title="declarai_tutorial.py"
 import declarai
 
-openai = declarai.openai(model="gpt-3.5-turbo", openai_token="<your-openai-token>")
+gpt_35 = declarai.openai(model="gpt-3.5-turbo", openai_token="<your-openai-token>")
 ```
 
 

docs/beginners-guide/simple-task.md

@@ -9,9 +9,9 @@ The simplest Declarai usage is a function decorated with `@task`:
 ```py
 import declarai
 
-openai = declarai.openai(model="gpt-3.5-turbo")
+gpt_35 = declarai.openai(model="gpt-3.5-turbo")
 
-@openai.task
+@gpt_35.task
 def say_something() -> str:
     """
     Say something short to the world

docs/best-practices/index.md

@@ -22,8 +22,8 @@ Reviewing the movie recommender example from the beginner's guide, we can see a
 from typing import Dict
 import declarai
 
-openai = declarai.openai(model="gpt-3.5-turbo")
-@openai.task
+gpt_35 = declarai.openai(model="gpt-3.5-turbo")
+@gpt_35.task
 def movie_recommender(user_input: str) -> Dict[str, str]:
     """
     Recommend a selection of movies to watch based on the user input
@@ -58,9 +58,9 @@ For example, in the following, the prompt is written in single form, while the r
 from typing import List
 import declarai
 
-openai = declarai.openai(model="gpt-3.5-turbo")
+gpt_35 = declarai.openai(model="gpt-3.5-turbo")
 
-@openai.task
+@gpt_35.task
 def movie_recommender(user_input: str) -> List[str]:
     """
     Recommend a movie to watch based on the user input
@@ -74,8 +74,8 @@ Instead, we could write the prompt as follows:
 from typing import List
 import declarai
 
-openai = declarai.openai(model="gpt-3.5-turbo")
-@openai.task
+gpt_35 = declarai.openai(model="gpt-3.5-turbo")
+@gpt_35.task
 def movie_recommender(user_input: str) -> List[str]:
     """
     Recommend a selection of movies to watch based on the user input
@@ -94,9 +94,9 @@ For example in this implementation of a calculator bot, the bot usually returns
 from typing import Union
 import declarai
 
-openai = declarai.openai(model="gpt-3.5-turbo")
+gpt_35 = declarai.openai(model="gpt-3.5-turbo")
 
-@openai.experimental.chat
+@gpt_35.experimental.chat
 class CalculatorBot:
     """
     You a calculator bot,

docs/examples/deployments/index.md

@@ -13,10 +13,10 @@ from fastapi import FastAPI, APIRouter
 import declarai
 app = FastAPI()
 router = APIRouter()
-openai = declarai.openai(model="gpt-3.5-turbo")
+gpt_35 = declarai.openai(model="gpt-3.5-turbo")
 
 
-@openai.task
+@gpt_35.task
 def movie_recommender(user_input: str) -> Dict[str, str]:
     """
     Recommend a selection of real movies to watch based on the user input

docs/features/chat/advanced-initialization.md

@@ -11,8 +11,8 @@ Let's see how we can initialize a chatbot by passing the `system` and `greeting`
 ```py
 import declarai
 
-openai = declarai.openai(model="gpt-3.5-turbo")
-@openai.experimental.chat
+gpt_35 = declarai.openai(model="gpt-3.5-turbo")
+@gpt_35.experimental.chat
 class SQLBot:
     ...
 

docs/features/chat/chat-memory/file-memory.md

@@ -5,11 +5,11 @@ For chat that requires a persistent message history, you can use a file to store
 ## Set file memory
 
 ```py
-from declarai import Declarai
+import declarai
 from declarai.memory import FileMessageHistory
-declarai = Declarai(provider="openai", model="gpt-3.5-turbo")
+gpt_35 = declarai.openai(model="gpt-3.5-turbo")
 
-@declarai.experimental.chat(chat_history=FileMessageHistory("sql_bot_history.txt")) # (1)!
+@gpt_35.experimental.chat(chat_history=FileMessageHistory("sql_bot_history.txt")) # (1)!
 class SQLBot:
     """
     You are a sql assistant. You help with SQL related questions with one-line answers.
@@ -28,10 +28,11 @@ We can also initialize the `FileMessageHistory` class with a custom file path.
 In case you want to set the file memory at runtime, you can use the `set_memory` method.
 
 ```py
-from declarai import Declarai
+import declarai
 from declarai.memory import FileMessageHistory
+gpt_35 = declarai.openai(model="gpt-3.5-turbo")
 
-@declarai.experimental.chat
+@gpt_35.experimental.chat
 class SQLBot:
     """
     You are a sql assistant. You help with SQL related questions with one-line answers.

docs/features/chat/chat-memory/index.md

@@ -61,12 +61,12 @@ If you prefer to have a persistent history, you can use the `FileMessageHistory`
 Setting up a memory is done by passing `chat_history` as a keyword argument to the `declarai.experimental.chat` decorator.
 
 ```py
-from declarai import Declarai
+import declarai
 from declarai.memory import FileMessageHistory
 
-declarai = Declarai(provider="openai", model="gpt-3.5-turbo")
+gpt_35 = declarai.openai(model="gpt-3.5-turbo")
 
-@declarai.experimental.chat(chat_history=FileMessageHistory("sql_bot_history.txt")) # (1)!
+@gpt_35.experimental.chat(chat_history=FileMessageHistory("sql_bot_history.txt")) # (1)!
 class SQLBot:
     """
     You are a sql assistant. You help with SQL related questions with one-line answers.
@@ -78,12 +78,12 @@ class SQLBot:
 We can also initialize the chat_history at runtime
 
 ```py
-from declarai import Declarai
+import declarai
 from declarai.memory import FileMessageHistory
 
-declarai = Declarai(provider="openai", model="gpt-3.5-turbo")
+gpt_35 = declarai.openai(model="gpt-3.5-turbo")
 
-@declarai.experimental.chat
+@gpt_35.experimental.chat
 class SQLBot:
     """
     You are a sql assistant. You help with SQL related questions with one-line answers.