Adam Green

My understanding is that this would mean creating a dataset: `AUTH=alan-watts,TEXT=text to train on, `AUTH=bible,TEXT=more text to train on`

Hello @gpt2ent - thank you for the PR, sorry it took me so long. Looks interesting! In `GPT2_with_Javascript_interface_POC.ipynb`, you seem to be connecting to [gpt-2-simple](https://github.com/gpt2ent/gpt-2-simple) which is based on the...

How about SAC?

Hi @adityauser This seems to be a `matplotlib` error? Have you made any progress?

Hi @fokx - thanks for your help :) This was an error on my end - try: ``` aws --no-sign-request --region ap-southeast-2 s3 cp s3://energy-py/public/dataset.zip dataset.zip ```

The data comes from the NEM - it's the 30 min trading price. You can read this for more on the NEM data - https://adgefficiency.com/hackers-aemo/

Hi @bollegijscoding - what platform are you using? I'm on MacOS and can't reproduce your error.

The model is set in a function `load_checkpoint`: https://github.com/ADGEfficiency/creative-writing-with-gpt2/blob/b0e59aa92e90dffa3fe3461eb5447782624774fd/creative/evaluate.py#L22 You may want to look at: https://github.com/ADGEfficiency/creative-writing-with-gpt2/blob/b0e59aa92e90dffa3fe3461eb5447782624774fd/creative/evaluate.py#L38

@andreasslyngstad ## Limit Depth of Discharge > If possible (but not critical), is there any way to set a min/max range of SOC (state of charge)? Lets say state of...

This is done - see https://energypylinear.adgefficiency.com/1.4.0/how-to/custom-constraints/