Unduhan equiformer pytorch - pengunduhan kode sumber equiformer pytorch

Equiformer - Pytorch (wip)

Implementasi Equiformer, jaringan perhatian ekivalen SE3/E3 yang mencapai SOTA baru, dan diadopsi untuk digunakan oleh EquiFold (Prescient Design) untuk pelipatan protein

Desainnya sepertinya dibuat dari SE3 Transformers, dengan perhatian produk titik diganti dengan Perhatian MLP dan pesan non-linier disampaikan dari GATv2. Ia juga melakukan produk tensor yang mendalam untuk efisiensi yang lebih tinggi. Jika menurut Anda saya salah, silakan kirim email kepada saya.

Pembaruan: Telah ada perkembangan baru yang membuat penskalaan jumlah derajat untuk jaringan ekivalen SE3 jauh lebih baik! Makalah ini pertama kali mencatat bahwa dengan menyelaraskan representasi sepanjang sumbu z (atau sumbu y menurut konvensi lain), harmonik bola menjadi jarang. Ini menghilangkan dimensi _mf dari persamaan. Makalah tindak lanjut dari Passaro dkk. mencatat matriks Clebsch Gordan juga menjadi jarang, menyebabkan penghapusan m _i dan l _f . Mereka juga menyimpulkan bahwa masalahnya telah berkurang dari SO(3) menjadi SO(2) setelah menyelaraskan repetisi ke satu sumbu. Equiformer v2 (Repositori resmi) memanfaatkan ini dalam kerangka kerja seperti transformator untuk mencapai SOTA baru.

Pasti akan melakukan lebih banyak pekerjaan/eksplorasi dalam hal ini. Untuk saat ini, saya telah memasukkan trik dari dua makalah pertama untuk Equiformer v1, simpan untuk konversi lengkap ke SO(2).

Pembaruan 2: Tampaknya ada SOTA baru tanpa interaksi apa pun antara perwakilan tingkat yang lebih tinggi (dengan kata lain, semua produk tensor/matematika clebsch gordan hilang). GotenNet, yang sepertinya merupakan versi trafo dari HEGNN

Memasang

$ pip install equiformer-pytorch

Penggunaan

 import torch
from equiformer_pytorch import Equiformer

model = Equiformer (
    num_tokens = 24 ,
    dim = ( 4 , 4 , 2 ),               # dimensions per type, ascending, length must match number of degrees (num_degrees)
    dim_head = ( 4 , 4 , 4 ),          # dimension per attention head
    heads = ( 2 , 2 , 2 ),             # number of attention heads
    num_linear_attn_heads = 0 ,     # number of global linear attention heads, can see all the neighbors
    num_degrees = 3 ,               # number of degrees
    depth = 4 ,                     # depth of equivariant transformer
    attend_self = True ,            # attending to self or not
    reduce_dim_out = True ,         # whether to reduce out to dimension of 1, say for predicting new coordinates for type 1 features
    l2_dist_attention = False      # set to False to try out MLP attention
). cuda ()

feats = torch . randint ( 0 , 24 , ( 1 , 128 )). cuda ()
coors = torch . randn ( 1 , 128 , 3 ). cuda ()
mask  = torch . ones ( 1 , 128 ). bool (). cuda ()

out = model ( feats , coors , mask ) # (1, 128)

out . type0 # invariant type 0    - (1, 128)
out . type1 # equivariant type 1  - (1, 128, 3)

Repositori ini juga mencakup cara untuk memisahkan penggunaan memori dari kedalaman menggunakan jaringan yang dapat dibalik. Dengan kata lain, jika Anda meningkatkan kedalaman, biaya memori akan tetap konstan dengan penggunaan satu blok transformator equiformer (perhatian dan umpan maju).

 import torch
from equiformer_pytorch import Equiformer

model = Equiformer (
    num_tokens = 24 ,
    dim = ( 4 , 4 , 2 ),
    dim_head = ( 4 , 4 , 4 ),
    heads = ( 2 , 2 , 2 ),
    num_degrees = 3 ,
    depth = 48 ,          # depth of 48 - just to show that it runs - in reality, seems to be quite unstable at higher depths, so architecture stil needs more work
    reversible = True ,   # just set this to True to use https://arxiv.org/abs/1707.04585
). cuda ()

feats = torch . randint ( 0 , 24 , ( 1 , 128 )). cuda ()
coors = torch . randn ( 1 , 128 , 3 ). cuda ()
mask  = torch . ones ( 1 , 128 ). bool (). cuda ()

out = model ( feats , coors , mask )

out . type0 . sum (). backward ()

Tepian

dengan tepi, mis. ikatan atom

 import torch
from equiformer_pytorch import Equiformer

model = Equiformer (
    num_tokens = 28 ,
    dim = 64 ,
    num_edge_tokens = 4 ,       # number of edge type, say 4 bond types
    edge_dim = 16 ,             # dimension of edge embedding
    depth = 2 ,
    input_degrees = 1 ,
    num_degrees = 3 ,
    reduce_dim_out = True
)

atoms = torch . randint ( 0 , 28 , ( 2 , 32 ))
bonds = torch . randint ( 0 , 4 , ( 2 , 32 , 32 ))
coors = torch . randn ( 2 , 32 , 3 )
mask  = torch . ones ( 2 , 32 ). bool ()

out = model ( atoms , coors , mask , edges = bonds )

out . type0 # (2, 32)
out . type1 # (2, 32, 3)

dengan matriks ketetanggaan

 import torch
from equiformer_pytorch import Equiformer

model = Equiformer (
    dim = 32 ,
    heads = 8 ,
    depth = 1 ,
    dim_head = 64 ,
    num_degrees = 2 ,
    valid_radius = 10 ,
    reduce_dim_out = True ,
    attend_sparse_neighbors = True ,  # this must be set to true, in which case it will assert that you pass in the adjacency matrix
    num_neighbors = 0 ,               # if you set this to 0, it will only consider the connected neighbors as defined by the adjacency matrix. but if you set a value greater than 0, it will continue to fetch the closest points up to this many, excluding the ones already specified by the adjacency matrix
    num_adj_degrees_embed = 2 ,       # this will derive the second degree connections and embed it correctly
    max_sparse_neighbors = 8         # you can cap the number of neighbors, sampled from within your sparse set of neighbors as defined by the adjacency matrix, if specified
)

feats = torch . randn ( 1 , 128 , 32 )
coors = torch . randn ( 1 , 128 , 3 )
mask  = torch . ones ( 1 , 128 ). bool ()

# placeholder adjacency matrix
# naively assuming the sequence is one long chain (128, 128)

i = torch . arange ( 128 )
adj_mat = ( i [:, None ] <= ( i [ None , :] + 1 )) & ( i [:, None ] >= ( i [ None , :] - 1 ))

out = model ( feats , coors , mask , adj_mat = adj_mat )

out . type0 # (1, 128)
out . type1 # (1, 128, 3)

Apresiasi

StabilityAI atas sponsor yang murah hati, serta sponsor saya yang lain di luar sana

Pengujian

Tes untuk kesetaraan dll

$ python setup.py test

Contoh

Pertama instal sidechainnet

$ pip install sidechainnet

Kemudian jalankan tugas penolakan tulang punggung protein

$ python denoise.py

Semua yang harus dilakukan

Kutipan

 @article { Liao2022EquiformerEG ,
    title   = { Equiformer: Equivariant Graph Attention Transformer for 3D Atomistic Graphs } ,
    author  = { Yi Liao and Tess E. Smidt } ,
    journal = { ArXiv } ,
    year    = { 2022 } ,
    volume  = { abs/2206.11990 }
}

 @article { Lee2022.10.07.511322 ,
    author  = { Lee, Jae Hyeon and Yadollahpour, Payman and Watkins, Andrew and Frey, Nathan C. and Leaver-Fay, Andrew and Ra, Stephen and Cho, Kyunghyun and Gligorijevic, Vladimir and Regev, Aviv and Bonneau, Richard } ,
    title   = { EquiFold: Protein Structure Prediction with a Novel Coarse-Grained Structure Representation } ,
    elocation-id = { 2022.10.07.511322 } ,
    year    = { 2022 } ,
    doi     = { 10.1101/2022.10.07.511322 } ,
    publisher = { Cold Spring Harbor Laboratory } ,
    URL     = { https://www.biorxiv.org/content/early/2022/10/08/2022.10.07.511322 } ,
    eprint  = { https://www.biorxiv.org/content/early/2022/10/08/2022.10.07.511322.full.pdf } ,
    journal = { bioRxiv }
}

 @article { Shazeer2019FastTD ,
    title   = { Fast Transformer Decoding: One Write-Head is All You Need } ,
    author  = { Noam M. Shazeer } ,
    journal = { ArXiv } ,
    year    = { 2019 } ,
    volume  = { abs/1911.02150 }
}

 @misc { ding2021cogview ,
    title   = { CogView: Mastering Text-to-Image Generation via Transformers } ,
    author  = { Ming Ding and Zhuoyi Yang and Wenyi Hong and Wendi Zheng and Chang Zhou and Da Yin and Junyang Lin and Xu Zou and Zhou Shao and Hongxia Yang and Jie Tang } ,
    year    = { 2021 } ,
    eprint  = { 2105.13290 } ,
    archivePrefix = { arXiv } ,
    primaryClass = { cs.CV }
}

 @inproceedings { Kim2020TheLC ,
    title   = { The Lipschitz Constant of Self-Attention } ,
    author  = { Hyunjik Kim and George Papamakarios and Andriy Mnih } ,
    booktitle = { International Conference on Machine Learning } ,
    year    = { 2020 }
}

 @article { Zitnick2022SphericalCF ,
    title   = { Spherical Channels for Modeling Atomic Interactions } ,
    author  = { C. Lawrence Zitnick and Abhishek Das and Adeesh Kolluru and Janice Lan and Muhammed Shuaibi and Anuroop Sriram and Zachary W. Ulissi and Brandon C. Wood } ,
    journal = { ArXiv } ,
    year    = { 2022 } ,
    volume  = { abs/2206.14331 }
}

 @article { Passaro2023ReducingSC ,
  title     = { Reducing SO(3) Convolutions to SO(2) for Efficient Equivariant GNNs } ,
  author    = { Saro Passaro and C. Lawrence Zitnick } ,
  journal   = { ArXiv } ,
  year      = { 2023 } ,
  volume    = { abs/2302.03655 }
}

 @inproceedings { Gomez2017TheRR ,
    title   = { The Reversible Residual Network: Backpropagation Without Storing Activations } ,
    author  = { Aidan N. Gomez and Mengye Ren and Raquel Urtasun and Roger Baker Grosse } ,
    booktitle = { NIPS } ,
    year    = { 2017 }
}

 @article { Bondarenko2023QuantizableTR ,
    title   = { Quantizable Transformers: Removing Outliers by Helping Attention Heads Do Nothing } ,
    author  = { Yelysei Bondarenko and Markus Nagel and Tijmen Blankevoort } ,
    journal = { ArXiv } ,
    year    = { 2023 } ,
    volume  = { abs/2306.12929 } ,
    url     = { https://api.semanticscholar.org/CorpusID:259224568 }
}

 @inproceedings { Arora2023ZoologyMA ,
  title   = { Zoology: Measuring and Improving Recall in Efficient Language Models } ,
  author  = { Simran Arora and Sabri Eyuboglu and Aman Timalsina and Isys Johnson and Michael Poli and James Zou and Atri Rudra and Christopher R'e } ,
  year    = { 2023 } ,
  url     = { https://api.semanticscholar.org/CorpusID:266149332 }
}