Metadata-Version: 2.1
Name: gpt3-simple-primer
Version: 0.0.35
Summary: GPT-3 wrapper for Python
Home-page: https://github.com/happilyeverafter95/gpt-3
Author: happilyeverafter95
Author-email: author@example.com
License: UNKNOWN
Project-URL: Bug Tracker, https://github.com/happilyeverafter95/gpt-3/issues
Description: # gpt3_simple_primer
        
        Simple GPT-3 primer using `openai`.
        
        ## Background
        
        Generative Pre-trained Transformer 3 (GPT-3) is an autoregressive language model that uses deep learning to produce human-like text. For more information, visit https://openai.com/blog/openai-api/.
        
        The [OpenAI Python library](https://github.com/openai/openai-python) is the official Python wrapper for the OpenAI API. The purpose of this library is to simplify the priming process by providing easy to use methods for setting the instructions and adding examples.
        
        ## Priming
        
        Priming is the practice of providing an initial prompt to the language model to improve subsequent model predictions.
        
        GPT-3 generally does very well even with short instructions and a few examples of your intended use case. Examples are typically delimited based on input and output. For instance, GPT-3 can be used to predict food ingredients based on the following prompt:
        
        ```
        Given the name of a food, list the ingredients used to make this meal.
        
        Food: apple pie
        Ingredients: apple, butter, flour, egg, cinnamon, crust, sugar
        
        Food: guacamole
        Ingredients: avocado, tomato, onion, lime, salt
        ```
        
        ## Requirements
        
        You will need an API key from OpenAI to access GPT-3.
        
        ## Usage
        
        `input_text` and `output_text` determines how input and output are delimited in the examples. The default is to use `Input` and `Output`.
        
        ```
        from gpt3_simple_primer import GPT3Generator
        
        key = 'sk-xxxxx'  # openai key
        
        generator = GPT3Generator(input_text='Food',
                                  output_text='Ingredients')
        
        generator.set_key(key)
        generator.set_instructions('List the ingredients for this meal.')
        generator.add_example('apple pie', 'apple, butter, flour, egg, cinnamon, crust, sugar')
        generator.add_example('guacamole', 'avocado, tomato, onion, lime, salt')
        
        # Ingredients: cream, egg yolk, sugar, lime, key lime juice
        generator.generate(starting_text='key lime pie',
                           engine='davinci',
                           max_tokens=20,
                           temperature=0.5,
                           top_p=1)
        ```
        
        To see the prompt used for priming:
        
        ```
        generator.get_prompt()
        ```
        
        To remove an example from the prompt:
        
        ```
        generator.remove_example('apple pie')
        ```
        
        ## Examples
        
        The library includes examples of GPT-3 applications based off of specific prompts.
        
        ```
        from gpt3_simple_primer.examples import idiom_explainer
        
        key = 'sk-xxxxx'  # openai key
        
        idiom_explainer.set_key(key)
        idiom_explainer.generate('hill to die on', max_tokens=15, engine='davinci')
        ```
        
Keywords: gpt-3,text generation,language model,nlp
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.7.1
Description-Content-Type: text/markdown
