debug added to export video

Author: prishajain1

src/maxdiffusion/pipelines/wan/wan_pipeline_i2v_2p1.py

@@ -237,10 +237,7 @@ def __call__(
       if self.config.expand_timesteps:
          latents = (1 - first_frame_mask) * condition + first_frame_mask * latents
       latents_bcthw = jnp.transpose(latents, (0, 4, 1, 2, 3))
-      max_logging.log(f"[DEBUG CALL] latents shape before denorm: {latents_bcthw.shape}")
-
       latents_denorm_bcthw = self._denormalize_latents(latents_bcthw)
-      max_logging.log(f"[DEBUG CALL] latents shape after denorm: {latents_denorm_bcthw.shape}")
 
 
     if output_type == "latent":

src/maxdiffusion/utils/export_utils.py

@@ -165,6 +165,9 @@ def export_to_video(
   """
   # TODO: Dhruv. Remove by Diffusers release 0.33.0
   # Added to prevent breaking existing code
+  if not video_frames:
+    logger.warning("export_to_video: video_frames list is empty.")
+    return ""
   if not is_imageio_available():
     logger.warning(
         (
@@ -194,7 +197,45 @@ def export_to_video(
     output_video_path = tempfile.NamedTemporaryFile(suffix=".mp4").name
 
   if isinstance(video_frames[0], np.ndarray):
-    video_frames = [(frame * 255).astype(np.uint8) for frame in video_frames]
+    logger.info("Processing np.ndarray frames for video export.")
+    processed_frames = []
+    for i, frame in enumerate(video_frames):
+        if frame.dtype != np.float32:
+            frame = frame.astype(np.float32) # Ensure float32 for checks
+
+        # --- Check for non-finite values ---
+        is_finite = np.isfinite(frame)
+        if not np.all(is_finite):
+            nan_count = np.isnan(frame).sum()
+            inf_count = np.isinf(frame).sum()
+            logger.warning(f"[EXPORT WARN] Frame {i}: Non-finite values detected! "
+                           f"NaNs: {nan_count}, Infs: {inf_count}")
+            # Sanitize: Replace NaNs with 0, Infs with 0 or 1
+            frame = np.nan_to_num(frame, nan=0.0, posinf=1.0, neginf=0.0)
+            logger.info(f"[EXPORT INFO] Frame {i}: Non-finite values replaced.")
+
+        # --- Check for out-of-range values [0.0, 1.0] ---
+        min_val = np.min(frame)
+        max_val = np.max(frame)
+        if min_val < 0.0 or max_val > 1.0:
+            logger.warning(f"[EXPORT WARN] Frame {i}: Values out of [0.0, 1.0] range. "
+                           f"Min={min_val:.4f}, Max={max_val:.4f}")
+            # Clip values to the valid range
+            frame = np.clip(frame, 0.0, 1.0)
+            logger.info(f"[EXPORT INFO] Frame {i}: Values clipped to [0.0, 1.0].")
+
+        # --- Convert to uint8 ---
+        try:
+            frame_uint8 = (frame * 255.0).astype(np.uint8)
+            processed_frames.append(frame_uint8)
+        except Exception as e:
+            logger.error(f"[EXPORT ERROR] Frame {i}: Failed to convert to uint8: {e}")
+            # Fallback: append a black frame
+            if len(video_frames) > 0 and isinstance(video_frames[0], np.ndarray):
+                 processed_frames.append(np.zeros(video_frames[0].shape, dtype=np.uint8))
+
+    video_frames = processed_frames # Use the sanitized and converted frames
+
 
   elif isinstance(video_frames[0], PIL.Image.Image):
     video_frames = [np.array(frame) for frame in video_frames]