Posts

Post not yet marked as solved
0 Replies
178 Views
Hello, Apple just terminated my Developer account and revoked my certificate. The email mentioned : "Apple has good reason to believe, or has reasonable suspicions, that your Application contains malware or malicious, suspicious or harmful code or components." I have just 2 published applications: an iOS app on the AppStore that has not been updated for 9 years a macOS app the does screen recording to send to Syphon published 2 years ago. It uses macOS security settings to allow/disallow screen access. I recently helped 2 clients solve their notarization issues: one were using an old Electron framework that were causing notarization Reject. Upgrading the framework solved the issue. second one had a problematic source code (calling GoogleChrome via NSTask). I have shown the client that without this problematic code, the notarization was OK. I just wanted to know which "Application" is causing my account termination? Did I notarize good code but with a bundleID that was flagged as suspicious ? I have been developing as a freelance for macOS/iOS for 16 years. The lesson here is to ALWAYS use the certificate/app-specific-password of your client. The problem here is that I could NO LONGER work and this is putting my freelance activity at risk.
Posted
by r2d3.
Last updated
.
Post marked as solved
1 Replies
389 Views
I am trying to coremltools.converters.convert a traced PyTorch model and I got an error: PyTorch convert function for op 'intimplicit' not implemented I am trying to convert a RVC model from github. I traced the model with torch.jit.trace and it fails. So I traced down the problematic part to the ** layer : https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/blob/main/infer/lib/infer_pack/modules.py#L188 import torch import coremltools as ct from infer.lib.infer_pack.modules import ** model = **(192, 5, dilation_rate=1, n_layers=16, ***_channels=256, p_dropout=0) model.remove_weight_norm() model.eval() test_x = torch.rand(1, 192, 200) test_x_mask = torch.rand(1, 1, 200) test_g = torch.rand(1, 256, 1) traced_model = torch.jit.trace(model, (test_x, test_x_mask, test_g), check_trace = True) x = ct.TensorType(name='x', shape=test_x.shape) x_mask = ct.TensorType(name='x_mask', shape=test_x_mask.shape) g = ct.TensorType(name='g', shape=test_g.shape) mlmodel = ct.converters.convert(traced_model, inputs=[x, x_mask, g]) I got an error RuntimeError: PyTorch convert function for op 'intimplicit' not implemented. How could I modify the **::forward so it does not generate an intimplicit operator ? Thanks David
Posted
by r2d3.
Last updated
.
Post not yet marked as solved
3 Replies
1.3k Views
I am using Xcode 12.5.1 on macOS BigSur (11.6.1). I have upgraded my iPhone6 to iOS 12.5.5 and since then the Metal replayer is quitting before the end of the replay. It was working fine with iOS 12.5.4. I have tried Xcode 13 (and Xcode 13 beta) and I have the same issue. Is there a way to debug/profile Metal code on these older devices ? Regards David
Posted
by r2d3.
Last updated
.
Post not yet marked as solved
0 Replies
470 Views
Hello,I would like to use CCTag https://github.com/alicevision/CCTag in order to get very precise subpixel 2D point in image.I am wondering if there is a way to give ARKit the 2D position of a marker in each frame and get ARKit to use these to keep its 3D position. Each CCTag will just have a position (no rotation) in space and should help the whole 3D reconstruction.RegardsDavid
Posted
by r2d3.
Last updated
.