adding some prints

cogvideo_controlnet_pcd.py

@@ -172,8 +172,9 @@ class CogVideoXControlnetPCD(ModelMixin, ConfigMixin, PeftAdapterMixin):
     ):
         print("hidden_states.shape =", hidden_states.shape)
         print("controlnet_states.shape =", controlnet_states.shape)
-        print("image_rotary_emb.shape =", image_rotary_emb.shape)
-        
+        for i, emb in enumerate(image_rotary_emb):
+            print(f"image_rotary_emb[{i}].shape =", emb.shape)
+
         hidden_states = torch.cat([hidden_states, controlnet_states], dim=2)
         
         # controlnet_states = self.controlnext_encoder(controlnet_states, timestep=timestep)

scripts/inference.sh

@@ -28,7 +28,7 @@ CUDA_VISIBLE_DEVICES=0 python inference/cli_demo_camera_i2v_pcd.py \
     --controlnet_transformer_out_proj_dim_zero_init \
     --vae_channels 16 \
     --num_frames 49 \
-    --controlnet_transformer_num_layers 8 \
+    --controlnet_transformer_num_layers 8 \ # 4
     --infer_with_mask \
     --pool_style 'max' \
     --seed 1