ML Forge launches no-code PyTorch trainer

ML Forge hits a practical middle ground: faster onboarding for beginners without fully locking advanced users into a no-code abstraction.

• –Auto-filled input shapes plus propagated `in_channels` and `in_features` remove one of the most error-prone parts of early model building.
• –The tabbed workflow mirrors real ML iteration loops, so it works as both a prototyping tool and a learning aid.
• –PyTorch export is the key differentiator, since teams can start visually and then transition to editable code for deeper control.
• –Early community feedback points to the classic no-code tradeoff: convenience vs transparency, especially when debugging complex performance behavior.