mrreplikant / avrilai Goto Github PK

Windows 10, Python 3.7.8, AvrilAI 1.5, just installed from scratch today.

Followed the instructions for the install, am getting this error message scrolling off the screen on first running "python launch.py":

Model could not be loaded. Please try another model.

I did run pip install -r requirements .txt and did not get any error messages beforehand.

It also says :09/01/2022 08:50:13 - WARNING - CUDA is not available, you are limited to CPU only. I do have a 3090, does this not support running on NVIDIA cards?

Thanks

Hello! I'm trying to set up your program and the directions were very easy to follow, but I've hit a snag.

When I open the command window and type in the "pip install -r requirements.txt", it keeps giving me an error message saying that such a file doesn't exist in my directory, even though I have the File Directory pulled up and I'm staring at the exact same tile it's saying isn't there?

Do you possibly know what I'm doing wrong?

Thankful for your hard work,

Obsidian

Can it be done using the gpt neo, it is better than gpt 2 and also open source?
Btw great work

There are several mentions of the project's former name, alongside the old prompt. The old prompt (Me: Hello! AI:) appears to no longer work.

Hey there. This seems like a great project, but after doing the installation I've been running into an issue every time I try execute the program. I should also mention I am currently running on manjaro linux, and had to change the module loading in gpt2.py.

original:
from transformers.configuration_gpt2 import GPT2Config
from transformers.modeling_gpt2 import GPT2PreTrainedModel

changed:
from transformers import GPT2Config
from transformers import GPT2PreTrainedModel

not sure if this is correct, but it allows the program to start.

after that It keeps running into a fatal error any time I attempt to use it. Loading a prompt from file, creating a prompt and loading a conversation. here is the error text:

Traceback (most recent call last):
File "/home/userl/Project-Replikant-1.2/play.py", line 740, in
gm.play_story()
File "/home/user/Project-Replikant-1.2/play.py", line 679, in play_story
if not self.init_story(): # Failed init
File "/home/user/Project-Replikant-1.2/play.py", line 427, in init_story
self.story = new_story(self.generator, self.context, self.prompt)
File "/home/user/Project-Replikant-1.2/play.py", line 215, in new_story
story.act(prompt)
File "/home/user/Project-Replikant-1.2/storymanager.py", line 23, in act
result = self.generator.generate(
File "/home/user/Project-Replikant-1.2/gpt2generator.py", line 344, in generate
text = self.generate_raw(
File "/home/user/Project-Replikant-1.2/gpt2generator.py", line 307, in generate_raw
out = self.sample_sequence(
File "/home/user/Project-Replikant-1.2/gpt2generator.py", line 246, in sample_sequence
out = sample_sequence(
File "/home/user/Project-Replikant-1.2/gpt2generator.py", line 131, in sample_sequence
logits, pasts = model(input_ids=input_ids_next, past=pasts)
File "/home/user/.local/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1051, in _call_impl
return forward_call(*input, **kwargs)
TypeError: forward() got an unexpected keyword argument 'past'

hopefully there is an easy fix for this.

I tried to download it like in manual but got this:

So it seems this old version is unavaible now, can't install it from the site. I installed latest version instead (1.11.0) but got this line: ...\Python310\lib\site-packages\torch\nn\parameter.py", line 2, in
from torch._C import _disabled_torch_function_impl
ModuleNotFoundError: No module named 'torch._C'
I don't know what to do with this.

"model can not be found"?? how do I fix this

Hi! I'm running it on a mid 2015 macbook pro with 16GB memory and 500G SSD.
It took ~30s to response in conversations. Is this normal or how it can be improved? What is the optimal system environment to run the program?

Actually I've tried to run it on Google Colab , hoping it would work in the way AIDundeon does. But it turned out that downloading the model from mega.nz took too much time and the server disconnects.

I'm fascinated by AIs, especially this AvrilAI, but I'm having trouble with some of your instructions because I'm new to this. I don't understand the part that reads:

"Download the source code under "Releases", and extract the file after you download it.

Open the file, and click on the file path at the top of Windows explorer. It's the bar to the left of "search", and it contains the file path.

Once you click on it, backspace until the line is blank. Then type "CMD" in the line, and press enter. This should bring everything up in the command prompt".

I'm mostly struggling with that first paragraph. Could you please explain it better to me and what I'm supposed to do? Thanks in advance.

Hey, so I followed the instructions exactly for the Windows install, but I can't get the program to run. I click on play.py and a CMD window opens for a split-second, then closes again.

Any ideas on troubleshooting?

My PC is a Ryzen 3800X, 2080TI, 32GB RAM.