error importing sam2/_C.so when running the video_predictor_example.ipynb

[rishabh297]

I am getting the following error when I run the line of code that says:

labels = np.array([1], np.int32)
_, out_obj_ids, out_mask_logits = predictor.add_new_points(
    inference_state=inference_state,
    frame_idx=ann_frame_idx,
    obj_id=ann_obj_id,
    points=points,
    labels=labels,
)

The main import error message is:

ImportError: /content/segment-anything-2/sam2/_C.so: undefined symbol: _ZN3c1015SmallVectorBaseIjE8grow_podEPKvmm

This is the entire error:

---

ImportError Traceback (most recent call last)
in <cell line: 8>()
6 # for labels, 1 means positive click and 0 means negative click
7 labels = np.array([1], np.int32)
----> 8 _, out_obj_ids, out_mask_logits = predictor.add_new_points(
9 inference_state=inference_state,
10 frame_idx=ann_frame_idx,

4 frames
/usr/local/lib/python3.10/dist-packages/torch/utils/_contextlib.py in decorate_context(*args, **kwargs)
113 def decorate_context(*args, **kwargs):
114 with ctx_factory():
--> 115 return func(*args, **kwargs)
116
117 return decorate_context

/content/segment-anything-2/sam2/sam2_video_predictor.py in add_new_points(self, inference_state, frame_idx, obj_id, points, labels, clear_old_points, normalize_coords)
219 # Clamp the scale of prev_sam_mask_logits to avoid rare numerical issues.
220 prev_sam_mask_logits = torch.clamp(prev_sam_mask_logits, -32.0, 32.0)
--> 221 current_out, _ = self._run_single_frame_inference(
222 inference_state=inference_state,
223 output_dict=obj_output_dict, # run on the slice of a single object

/content/segment-anything-2/sam2/sam2_video_predictor.py in _run_single_frame_inference(self, inference_state, output_dict, frame_idx, batch_size, is_init_cond_frame, point_inputs, mask_inputs, reverse, run_mem_encoder, prev_sam_mask_logits)
808 # potentially fill holes in the predicted masks
809 if self.fill_hole_area > 0:
--> 810 pred_masks_gpu = fill_holes_in_mask_scores(
811 pred_masks_gpu, self.fill_hole_area
812 )

/content/segment-anything-2/sam2/utils/misc.py in fill_holes_in_mask_scores(mask, max_area)
221 # (background regions are those with mask scores <= 0)
222 assert max_area > 0, "max_area must be positive"
--> 223 labels, areas = get_connected_components(mask <= 0)
224 is_hole = (labels > 0) & (areas <= max_area)
225 # We fill holes with a small positive mask score (0.1) to change them to foreground.

/content/segment-anything-2/sam2/utils/misc.py in get_connected_components(mask)
59 components for foreground pixels and 0 for background pixels.
60 """
---> 61 from sam2 import _C
62
63 return _C.get_connected_componnets(mask.to(torch.uint8).contiguous())

ImportError: /content/segment-anything-2/sam2/_C.so: undefined symbol: _ZN3c1015SmallVectorBaseIjE8grow_podEPKvmm

Can you please help out with this issue. It would enable me to proceed with the project. Not sure how to fix it.

[vanamagautam24]

!python setup.py build_ext --inplace worked for me

run the above command at the beginning of the jupyter / colab notebook.

[iAsakiT3T]

I'm having the same issue.Pls have you solved it？

[topshu]

@vanamagautam24 thanks! it works for me!

[ronghanghu]

Hi, we have added INSTALL.md as an installation FAQ.

Regarding this error:

This is usually because you haven't run the pip install -e ".[demo]" step above or the installation failed. Please install SAM 2 first, and see the other issues if your installation fails. (You may also try python setup.py build_ext --inplace in the repo root as others suggested in https://github.com/facebookresearch/segment-anything-2/issues/77)

[rishabh297]

Thank you, What worked for me was running these commands in my Colab Notebook:

%cd /content/segment-anything-2
!python setup.py build_ext --inplace

Run the above commands after building the relevant sam2 model. This should help build the _C module appropriately.