Felix

Is there any solution?

@9woods123 did i solve it somehow? I also have to problem and i urgently need a solution

Any update for Windows? still "Could not build wheels for pymesh2, which is required to install pyproject.toml-based projects"

Yes, I solved it with nn.BCEWithLogitsLoss and the performance improved also a bit

In train.py change line 96: `if device=="cuda":` to `if torch.cuda.device_count() > 1`. It should only be called if you have more than one card bc net doesn't have a module...

I have the same problem with frida and also with objection no response I am using an rooted physical device: https://www.devicespecifications.com/de/model-cpu/066948b6 Frida: C:\Users\###\Downloads\platform-tools-latest-windows\platform-tools>frida -U -f tech.httptoolkit.pinning_demo -l frida_multiple_unpinning.js ____ /...

Update: I could now inject a script via frida-inject Example: ps -A | grep "tech.httptoolkit.pinning_demo" # Get PID_RUNNING_APP su ./frida -p PID_RUNNING_APP -s frida_multiple_unpinning.js

Its the Utils.getCrsf() function just replace it with ``` protected static String getCrsf() throws IOException { Request request = new Request.Builder() .url("https://www.instagram.com/") .addHeader("user-agent", Constants.WEB_USER_AGENT) .addHeader("accept", "*/*") .addHeader("accept-language", "en-US,en;q=0.9") .addHeader("x-requested-with", "XMLHttpRequest")...