apply suggestions from review

Author: a-r-r-o-w

src/diffusers/pipelines/cogvideo/pipeline_cogvideox_image2video.py

@@ -362,16 +362,9 @@ def prepare_latents(
                 f" size of {batch_size}. Make sure the batch size matches the length of the generators."
             )
 
-        assert image.ndim == 4
         image = image.unsqueeze(2)  # [B, C, F, H, W]
 
         if isinstance(generator, list):
-            if len(generator) != batch_size:
-                raise ValueError(
-                    f"You have passed a list of generators of length {len(generator)}, but requested an effective batch"
-                    f" size of {batch_size}. Make sure the batch size matches the length of the generators."
-                )
-
             image_latents = [
                 retrieve_latents(self.vae.encode(image[i].unsqueeze(0)), generator[i]) for i in range(batch_size)
             ]

tests/pipelines/cogvideo/test_cogvideox_image2video.py

@@ -54,6 +54,7 @@ class CogVideoXPipelineFastTests(PipelineTesterMixin, unittest.TestCase):
             "callback_on_step_end_tensor_inputs",
         ]
     )
+    test_xformers_attention = False
 
     def get_dummy_components(self):
         torch.manual_seed(0)
@@ -283,10 +284,6 @@ def test_vae_tiling(self, expected_diff_max: float = 0.3):
             "VAE tiling should not affect the inference results",
         )
 
-    @unittest.skip("xformers attention processor does not exist for CogVideoX")
-    def test_xformers_attention_forwardGenerator_pass(self):
-        pass
-
     def test_fused_qkv_projections(self):
         device = "cpu"  # ensure determinism for the device-dependent torch.Generator
         components = self.get_dummy_components()