itch.io

I can't complete even the first Quest because enemies keep running away out-of-bounds where I can't hit them and then don't come back. Also, there's no feedback at all for being hit by the enemies. The gameplay definitely needs a lot more polishing, especially for a $10 game.

I'm having this issue as well, when I open up the game and go to the main menu everything is fine, but as soon as I start or load a game my processor use skyrockets, all the fans turn on, processor temperature shoots up to 80°C. The game itself works fine but no way in hell I'm going to play a game that makes my processor run that high. And I don't buy that it's the "quantity of assets loaded", because when I press ESC to open up the menu it goes down instantly, and the menu is just an overlay that doesn't unload the game's assets behind it.

My computer is pretty high-end (i7-12700KF + RTX 4080 SUPER) and should definitely not be sweating to run a game like this. After trying out a few things, by capping the game to 30FPS in the NVIDIA Control Panel I was able to make it run without melting my processor. So there is definitely some runaway calculation tied to the FPS running in the background while the game is unpaused that is way more intensive than it probably should be.

Thanks for the tip, changing the version of the transformers library worked like a charm!

I downloaded the latest Windows version (0.0.14) and ran the game, everything seems to have been installed and configured properly using the gpt2-xl model. But when trying to call the AI it fails with the following stacktrace on the console:

Error in connection handler
Traceback (most recent call last):
  File "F:\AI\GodAI\APIs\TransformersAPI\miniconda\envs\godai\lib\site-packages\websockets\server.py", line 191, in handler
    await self.ws_handler(self, path)
  File "F:\AI\GodAI\APIs\TransformersAPI\server.py", line 138, in hello
    replace_tokens = token_encode(["<|endoftext|>"])
  File "F:\AI\GodAI\APIs\TransformersAPI\server.py", line 66, in token_encode
    tokens = tokenizer.encode(text, add_special_tokens = False, add_prefix_space = True)
  File "F:\AI\GodAI\APIs\TransformersAPI\miniconda\envs\godai\lib\site-packages\transformers\tokenization_utils_base.py", line 2091, in encode
    encoded_inputs = self.encode_plus(
  File "F:\AI\GodAI\APIs\TransformersAPI\miniconda\envs\godai\lib\site-packages\transformers\tokenization_utils_base.py", line 2407, in encode_plus
    return self._encode_plus(
  File "F:\AI\GodAI\APIs\TransformersAPI\miniconda\envs\godai\lib\site-packages\transformers\models\gpt2\tokenization_gpt2_fast.py", line 169, in _encode_plus
    return super()._encode_plus(*args, **kwargs)
  File "F:\AI\GodAI\APIs\TransformersAPI\miniconda\envs\godai\lib\site-packages\transformers\tokenization_utils_fast.py", line 446, in _encode_plus
    batched_output = self._batch_encode_plus(
  File "F:\AI\GodAI\APIs\TransformersAPI\miniconda\envs\godai\lib\site-packages\transformers\models\gpt2\tokenization_gpt2_fast.py", line 159, in _batch_encode_plus
    return super()._batch_encode_plus(*args, **kwargs)
TypeError: _batch_encode_plus() got an unexpected keyword argument 'add_prefix_space'
Client Connected
---
Awaiting for input