[](https://discord.gg/qUtxnK2NMf)
# Simple Mamba
## Install
`pip install simple-mamba`
## Usage
```python
import torch
from simple_mamba import MambaBlock
# Define block parameters
dim = 512
hidden_dim = 128
heads = 8
in_channels = 3
out_channels = 3
kernel_size = 3
# Create an instance of MambaBlock
mamba_block = MambaBlock(
dim, hidden_dim, heads, in_channels, out_channels, kernel_size
)
# Create a sample input tensor
x = torch.randn(1, dim, dim)
# Pass the tensor through the MambaBlock
output = mamba_block(x)
print("Output shape:", output.shape)
```
### `SSM`
```python
import torch
from simple_mamba import SSM
# # Example usage
vocab_size = 10000 # Example vocabulary size
embed_dim = 256 # Example embedding dimension
state_dim = 512 # State dimension
num_layers = 2 # Number of state-space layers
model = SSM(vocab_size, embed_dim, state_dim, num_layers)
# Example input (sequence of word indices)
input_seq = torch.randint(
0, vocab_size, (32, 10)
) # Batch size of 32, sequence length of 10
# Forward pass
logits = model(input_seq)
print(logits.shape) # Should be [32, 10, vocab_size]
```
# License
MIT
# Citation
```bibtex
@misc{gu2023mamba,
title={Mamba: Linear-Time Sequence Modeling with Selective State Spaces},
author={Albert Gu and Tri Dao},
year={2023},
eprint={2312.00752},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
```
Raw data
{
"_id": null,
"home_page": "https://github.com/kyegomez/SimpleMamba ",
"name": "simple-mamba",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.6,<4.0",
"maintainer_email": "",
"keywords": "artificial intelligence,deep learning,optimizers,Prompt Engineering",
"author": "Kye Gomez",
"author_email": "kye@apac.ai",
"download_url": "https://files.pythonhosted.org/packages/a3/01/9616137e14481932a59bac640618d88385b2dab63434066cd333ee05439a/simple_mamba-0.0.4.tar.gz",
"platform": null,
"description": "[](https://discord.gg/qUtxnK2NMf)\n\n# Simple Mamba\n\n## Install\n`pip install simple-mamba`\n\n\n## Usage\n```python\nimport torch\nfrom simple_mamba import MambaBlock\n\n\n# Define block parameters\ndim = 512\nhidden_dim = 128\nheads = 8\nin_channels = 3\nout_channels = 3\nkernel_size = 3\n\n# Create an instance of MambaBlock\nmamba_block = MambaBlock(\n dim, hidden_dim, heads, in_channels, out_channels, kernel_size\n)\n\n# Create a sample input tensor\nx = torch.randn(1, dim, dim)\n\n# Pass the tensor through the MambaBlock\noutput = mamba_block(x)\nprint(\"Output shape:\", output.shape)\n\n\n```\n\n### `SSM`\n```python\nimport torch \nfrom simple_mamba import SSM\n\n\n# # Example usage\nvocab_size = 10000 # Example vocabulary size\nembed_dim = 256 # Example embedding dimension\nstate_dim = 512 # State dimension\nnum_layers = 2 # Number of state-space layers\n\nmodel = SSM(vocab_size, embed_dim, state_dim, num_layers)\n\n# Example input (sequence of word indices)\ninput_seq = torch.randint(\n 0, vocab_size, (32, 10)\n ) # Batch size of 32, sequence length of 10\n\n # Forward pass\nlogits = model(input_seq)\nprint(logits.shape) # Should be [32, 10, vocab_size]\n\n```\n\n\n# License\nMIT\n\n\n# Citation\n```bibtex\n@misc{gu2023mamba,\n title={Mamba: Linear-Time Sequence Modeling with Selective State Spaces}, \n author={Albert Gu and Tri Dao},\n year={2023},\n eprint={2312.00752},\n archivePrefix={arXiv},\n primaryClass={cs.LG}\n}\n\n```",
"bugtrack_url": null,
"license": "MIT",
"summary": "Simple Mambda - Pytorch",
"version": "0.0.4",
"project_urls": {
"Documentation": "https://github.com/kyegomez/SimpleMamba",
"Homepage": "https://github.com/kyegomez/SimpleMamba ",
"Repository": "https://github.com/kyegomez/SimpleMamba"
},
"split_keywords": [
"artificial intelligence",
"deep learning",
"optimizers",
"prompt engineering"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "81ceb6f9b3e62ab44881df43f90580715867ada4ad22832d77a41b671d928e1c",
"md5": "a7cf2f917efeb8c9887ce768c5284c51",
"sha256": "bbe1ce017b7580452e47757121da810c472796ce9f088da6f3683609c7528e2b"
},
"downloads": -1,
"filename": "simple_mamba-0.0.4-py3-none-any.whl",
"has_sig": false,
"md5_digest": "a7cf2f917efeb8c9887ce768c5284c51",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.6,<4.0",
"size": 6064,
"upload_time": "2023-12-16T04:31:56",
"upload_time_iso_8601": "2023-12-16T04:31:56.982872Z",
"url": "https://files.pythonhosted.org/packages/81/ce/b6f9b3e62ab44881df43f90580715867ada4ad22832d77a41b671d928e1c/simple_mamba-0.0.4-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "a3019616137e14481932a59bac640618d88385b2dab63434066cd333ee05439a",
"md5": "12c7bd2f7848e7e4c8c6f4222048cb3b",
"sha256": "d5e02b1cb4d44a1209917b48afb5d33e6a73d2e0489753bca050cbe0e7a8d1c3"
},
"downloads": -1,
"filename": "simple_mamba-0.0.4.tar.gz",
"has_sig": false,
"md5_digest": "12c7bd2f7848e7e4c8c6f4222048cb3b",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.6,<4.0",
"size": 5415,
"upload_time": "2023-12-16T04:31:58",
"upload_time_iso_8601": "2023-12-16T04:31:58.554960Z",
"url": "https://files.pythonhosted.org/packages/a3/01/9616137e14481932a59bac640618d88385b2dab63434066cd333ee05439a/simple_mamba-0.0.4.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-12-16 04:31:58",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "kyegomez",
"github_project": "SimpleMamba ",
"github_not_found": true,
"lcname": "simple-mamba"
}
JSON
