[](https://discord.gg/qUtxnK2NMf)
# MambaFormer
Implementation of MambaFormer in Pytorch ++ Zeta from the paper: "Can Mamba Learn How to Learn? A Comparative Study on In-Context Learning Tasks"
## install
`pip3 install mamba-former`
## usage
```python
import torch
from mamba_former.main import MambaFormer
# Forward pass example
x = torch.randint(1, 1000, (1, 100)) # Token
# Tokens are integers representing input data
# Model
model = MambaFormer(
dim=512, # Dimension of the model
num_tokens=1000, # Number of unique tokens in the input data
depth=6, # Number of transformer layers
d_state=512, # Dimension of the transformer state
d_conv=128, # Dimension of the convolutional layer
heads=8, # Number of attention heads
dim_head=64, # Dimension of each attention head
return_tokens=True, # Whether to return the tokens in the output
)
# Forward pass
out = model(x) # Perform a forward pass through the model
# If training
# out = model(x, return_loss=True) # Perform a forward pass and calculate the loss
# Print the output
print(out) # Print the output tensor
print(out.shape) # Print the shape of the output tensor
```
# License
MIT
Raw data
{
"_id": null,
"home_page": "https://github.com/kyegomez/MambaFormer",
"name": "mamba-former",
"maintainer": null,
"docs_url": null,
"requires_python": "<4.0,>=3.9",
"maintainer_email": null,
"keywords": "artificial intelligence, deep learning, optimizers, Prompt Engineering",
"author": "Kye Gomez",
"author_email": "kye@apac.ai",
"download_url": "https://files.pythonhosted.org/packages/c0/0a/5fcc07d8485c4f2c03af84a68c00b5c1c0f208cc713f0f19bef389f53538/mamba_former-0.0.3.tar.gz",
"platform": null,
"description": "[](https://discord.gg/qUtxnK2NMf)\n\n# MambaFormer\nImplementation of MambaFormer in Pytorch ++ Zeta from the paper: \"Can Mamba Learn How to Learn? A Comparative Study on In-Context Learning Tasks\"\n\n## install\n`pip3 install mamba-former`\n\n## usage\n```python\nimport torch\nfrom mamba_former.main import MambaFormer\n\n# Forward pass example\nx = torch.randint(1, 1000, (1, 100)) # Token\n# Tokens are integers representing input data\n\n# Model\nmodel = MambaFormer(\n dim=512, # Dimension of the model\n num_tokens=1000, # Number of unique tokens in the input data\n depth=6, # Number of transformer layers\n d_state=512, # Dimension of the transformer state\n d_conv=128, # Dimension of the convolutional layer\n heads=8, # Number of attention heads\n dim_head=64, # Dimension of each attention head\n return_tokens=True, # Whether to return the tokens in the output\n)\n\n# Forward pass\nout = model(x) # Perform a forward pass through the model\n\n# If training\n# out = model(x, return_loss=True) # Perform a forward pass and calculate the loss\n\n# Print the output\nprint(out) # Print the output tensor\nprint(out.shape) # Print the shape of the output tensor\n\n```\n\n\n# License\nMIT\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "Paper - Pytorch",
"version": "0.0.3",
"project_urls": {
"Documentation": "https://github.com/kyegomez/MambaFormer",
"Homepage": "https://github.com/kyegomez/MambaFormer",
"Repository": "https://github.com/kyegomez/MambaFormer"
},
"split_keywords": [
"artificial intelligence",
" deep learning",
" optimizers",
" prompt engineering"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "3713bfc8e5b02af8bd22677d085771d85e1773dca56ddce87e5879eb39127be4",
"md5": "01b32ef4a1577295d34c701b8aa8c6a3",
"sha256": "b83622fd9dcbf6977a39d2056f1769e13a69c8f2569a05563774c37d00bc2a7d"
},
"downloads": -1,
"filename": "mamba_former-0.0.3-py3-none-any.whl",
"has_sig": false,
"md5_digest": "01b32ef4a1577295d34c701b8aa8c6a3",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": "<4.0,>=3.9",
"size": 3788,
"upload_time": "2024-04-05T00:54:45",
"upload_time_iso_8601": "2024-04-05T00:54:45.505045Z",
"url": "https://files.pythonhosted.org/packages/37/13/bfc8e5b02af8bd22677d085771d85e1773dca56ddce87e5879eb39127be4/mamba_former-0.0.3-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "c00a5fcc07d8485c4f2c03af84a68c00b5c1c0f208cc713f0f19bef389f53538",
"md5": "8dcd9882368c3b79f0e125855a437330",
"sha256": "0b54d02e36848c3a9ade42c397d19f5b98e34df77664fb29213d77d775137bbd"
},
"downloads": -1,
"filename": "mamba_former-0.0.3.tar.gz",
"has_sig": false,
"md5_digest": "8dcd9882368c3b79f0e125855a437330",
"packagetype": "sdist",
"python_version": "source",
"requires_python": "<4.0,>=3.9",
"size": 3905,
"upload_time": "2024-04-05T00:54:46",
"upload_time_iso_8601": "2024-04-05T00:54:46.529112Z",
"url": "https://files.pythonhosted.org/packages/c0/0a/5fcc07d8485c4f2c03af84a68c00b5c1c0f208cc713f0f19bef389f53538/mamba_former-0.0.3.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-04-05 00:54:46",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "kyegomez",
"github_project": "MambaFormer",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"requirements": [],
"lcname": "mamba-former"
}
JSON
