Add FAQ in README

README.md

@@ -192,6 +192,14 @@ sh slurm_test.sh {JOB_NAME} {NUM_GPU} {TRAIN_OUTPUT_DIR} {CKPT_ID}
 - NUM_GPU = 1 is recommended for testing
 - Logs and results  will be saved to `SMPLer-X/output/test_{JOB_NAME}_ep{CKPT_ID}_{TEST_DATSET}`
 
+## FAQ
+- `RuntimeError: Subtraction, the '-' operator, with a bool tensor is not supported. If you are trying to invert a mask, use the '~' or 'logical_not()' operator instead.`
+  
+  Follow [this post](https://github.com/mks0601/I2L-MeshNet_RELEASE/issues/6#issuecomment-675152527) and modify `torchgeometry`
+
+- `KeyError: 'SinePositionalEncoding is already registered in position encoding'` or any other similar KeyErrors due to duplicate module registration.
+
+  Maually add `force=True` to respective module registration under `main/transformer_utils/mmpose/models/utils`, e.g. `@POSITIONAL_ENCODING.register_module(force=True)` in [this file](main/transformer_utils/mmpose/models/utils/positional_encoding.py)
 
 ## References
 - [Hand4Whole](https://github.com/mks0601/Hand4Whole_RELEASE)

main/transformer_utils/mmpose/models/utils/positional_encoding.py

@@ -6,7 +6,7 @@ from mmcv.cnn.bricks.transformer import POSITIONAL_ENCODING
 from mmcv.runner import BaseModule
 
 
-@POSITIONAL_ENCODING.register_module()
+@POSITIONAL_ENCODING.register_module(force=True)
 class SinePositionalEncoding(BaseModule):
     """Position encoding with sine and cosine functions.
     See `End-to-End Object Detection with Transformers
@@ -98,7 +98,7 @@ class SinePositionalEncoding(BaseModule):
         return repr_str
 
 
-@POSITIONAL_ENCODING.register_module()
+@POSITIONAL_ENCODING.register_module(force=True)
 class LearnedPositionalEncoding(BaseModule):
     """Position embedding with learnable embedding weights.
     Args: