Update app.py

app.py

@@ -188,8 +188,8 @@ class LTX23DistilledA2VPipeline(DistilledPipeline):
         stage_1_output_shape = VideoPixelShape(
             batch=1,
             frames=num_frames,
-            width=width // 2,
-            height=height // 2,
+            width=width,
+            height=height,
             fps=frame_rate,
         )
         stage_1_conditionings = combined_image_conditionings(
@@ -213,39 +213,6 @@ class LTX23DistilledA2VPipeline(DistilledPipeline):
             initial_audio_latent=encoded_audio_latent,
         )
 
-        torch.cuda.synchronize()
-        cleanup_memory()
-
-        upscaled_video_latent = upsample_video(
-            latent=video_state.latent[:1],
-            video_encoder=video_encoder,
-            upsampler=self.model_ledger.spatial_upsampler(),
-        )
-        stage_2_sigmas = torch.tensor(STAGE_2_DISTILLED_SIGMA_VALUES, device=self.device)
-        stage_2_output_shape = VideoPixelShape(batch=1, frames=num_frames, width=width, height=height, fps=frame_rate)
-        stage_2_conditionings = combined_image_conditionings(
-            images=images,
-            height=stage_2_output_shape.height,
-            width=stage_2_output_shape.width,
-            video_encoder=video_encoder,
-            dtype=dtype,
-            device=self.device,
-        )
-        video_state = denoise_video_only(
-            output_shape=stage_2_output_shape,
-            conditionings=stage_2_conditionings,
-            noiser=noiser,
-            sigmas=stage_2_sigmas,
-            stepper=stepper,
-            denoising_loop_fn=denoising_loop,
-            components=self.pipeline_components,
-            dtype=dtype,
-            device=self.device,
-            noise_scale=stage_2_sigmas[0],
-            initial_video_latent=upscaled_video_latent,
-            initial_audio_latent=encoded_audio_latent,
-        )
-
         torch.cuda.synchronize()
         del transformer
         del video_encoder