Akshan Krithick

Results 2 issues of Akshan Krithick

Hey, I have been trying to run the qwen_image_edit finetune code, but the training doesn't stop. It keeps on running, the max_training_steps in the config doesn't work as intended

# What does this PR do? Enables robust batch inference for `QwenImageEditPlusPipeline` by normalizing input tensor shapes, implementing a "resize" strategy, and handling variable-length prompt embeddings. Adds a new batch...