nero-dv
commented
6 months ago This can be resolved by modifying the my_open function in the "block_requests.py" file under the modules directory.
# Kindly provided by our friend WizardLM-30B
def my_open(*args, **kwargs):
filename = str(args[0])
if filename.endswith("index.html"):
with original_open(*args, **kwargs) as f:
file_contents = f.read()
file_contents = file_contents.replace("cdnjs.cloudflare.com", "127.0.0.1")
file_contents = file_contents.replace(
"</head>",
'\n <script src="file/js/katex/katex.min.js"></script>'
'\n <script src="file/js/katex/auto-render.min.js"></script>'
'\n <script src="file/js/highlightjs/highlight.min.js"></script>'
'\n <script src="file/js/highlightjs/highlightjs-copy.min.js"></script>'
"\n <script>hljs.addPlugin(new CopyButtonPlugin());</script>"
"\n </head>",
)
return io.StringIO(file_contents)
else:
return original_open(*args, **kwargs)though depending on how old installation is, you would omit the newer replace
# Kindly provided by our friend WizardLM-30B
def my_open(*args, **kwargs):
filename = str(args[0])
if filename.endswith("index.html"):
with original_open(*args, **kwargs) as f:
file_contents = f.read()
file_contents = file_contents.replace("cdnjs.cloudflare.com", "127.0.0.1")
return io.StringIO(file_contents)
else:
return original_open(*args, **kwargs)I had to do this after updating to the latest versions of pytorch, flash_attention 2 and gradio.
Bedoshady
qnnnnez
maxPietsch42
Accurio
Describe the bug
I tried installing server and running it which worked at first, but couldnt load my llama 3 models, I typed this command
conda install pytorch torchvision torchaudio pytorch-cuda=11.7 -c pytorch -c nvidiathen it stopped working, I tried different solutions but it didnt workIs there an existing issue for this?
Reproduction
Install web text ui normally via miniconda then do this command
conda install pytorch torchvision torchaudio pytorch-cuda=11.7 -c pytorch -c nvidiaScreenshot
No response
Logs
System Info