mamba
mamba copied to clipboard
state-spaces
Thanks for your excellent work! Based on this repo, we build VMamba for vision tasks: https://github.com/MzeroMiko/VMamba. VMamba exhibits global effective reception fields with linear complexity. Welcome to use VMamba!
When using `mamba`, I found the following RuntimeError: ``` File "/data2/user/workspace/D-Mamba/d_mamba/models/mamba.py", line 171, in forward h = self.mamba(h) File "/home/user/miniconda3/envs/py39/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1518, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File "/home/user/miniconda3/envs/py39/lib/python3.9/site-packages/torch/nn/modules/module.py", line...
Hello, I'm trying to fine-tune the mamba model with a huggingface trainer but I'm facing an issue : `AttributeError: 'MambaConfig' object has no attribute 'to_json_string'` This is due to the...
I have tried to train the algorithm using different order of tokens, and I have realized anytime the token order is random, it generates NaN in the embedding layer. I'm...
Hi, I am trying to build the mamba from the source with `pip install --no-build-isolation -e .` and it fails bc it can't find files in mamba/build/. I have no...
**Reproduce:** Clone `https://github.com/havenhq/mamba-chat` and run training example in README with current version of `mamba_ssm`. **Solution:** Derive MambdaConfig from PretrainedConfig base class
Greetings, I would like to train a complex-valued Mamba model. As far as I can tell from the repository, the SSM op and kernel both support complex valued inputs, but...
I am stuck on this "Building wheel for mamba-ssm (setup.py) ... error" error for a long time now, and I haven't found a solution, it's too crashy~ have any of...
Any plan or interest to use OpenChat algorithm (https://github.com/imoneoi/openchat) to make mamba-chat? I tried OpenChat, and it seems very good.
