OpenAI GPT 2 project

I need a freelancer that understands and have experience with Python, Deep learning / Machine learning and preferably have a keen interest in GPT-2 / Neural Language & AI in general. The objective is to get a GTP-2 model installed, trained and tweaked so it outputs human readable text. Then develop it so it can take input from a website URL (an article on that URL) and then output a text that is on one hand different from the original ( a minimum of 90% unique when compared), but on the same topic. For example, if you have an article about how to grow tomatoes in your own garden, describing everything from buying the tools and seeds, to how to ferment the ground etc - the new text should still be a guide about how to grow tomatoes in your own garden, just look like it's written by a different person. This is NOT meant to be a text spinner. But rather a machine learning model that can generate great articles. First steps: 1. Install Setup GPT-2 on [login to view URL] or something similar. 2. Train the model. GPT-2 is trained with a simple objective: predict the next word, given all of the previous words within some text we input (see point 3) 3. Write a script that allows the model to read a webpage with an article and then use that as the input. 4. Save the generated text either in a database with proper formatting (line breaks etc) or in a text file. Keep in mind that this is the first part of a long line of steps, but it's crucial that we make this part work. Expect a lot of tweaking. It is imperative that the text generated based on a sample, is on topic, unique and human readable. Some required reading / sources: [login to view URL] [login to view URL] [login to view URL]