Página Principal Explorar Ajuda
Iniciar Sessão
kekezack
/
X_SSL_Net
0
0
Fork 0
Ficheiros Problemas 0 Pull Requests 0 Wiki
Árvore: fbbbfe53c8
Ramos Etiquetas
X_SSL_Net
/
tmp
/
VMamba-main
/
segmentation
/
configs
/
swin
/
metafile.yaml

metafile.yaml 7.2 KB
Histórico Em bruto

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143 
  1. Models:
    2. 

  2. - Name: swin-tiny-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512
    3. 

  3.   In Collection: UPerNet
    4. 

  4.   Results:
    5. 

  5.     Task: Semantic Segmentation
    6. 

  6.     Dataset: ADE20K
    7. 

  7.     Metrics:
    8. 

  8.       mIoU: 44.41
    9. 

  9.       mIoU(ms+flip): 45.79
    10. 

  10.   Config: configs/swin/swin-tiny-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512.py
    11. 

  11.   Metadata:
    12. 

  12.     Training Data: ADE20K
    13. 

  13.     Batch Size: 16
    14. 

  14.     Architecture:
    15. 

  15.     - Swin-T
    16. 

  16.     - UPerNet
    17. 

  17.     Training Resources: 8x V100 GPUS
    18. 

  18.     Memory (GB): 5.02
    19. 

  19.   Weights: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_tiny_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K/upernet_swin_tiny_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K_20210531_112542-e380ad3e.pth
    20. 

  20.   Training log: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_tiny_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K/upernet_swin_tiny_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K_20210531_112542.log.json
    21. 

  21.   Paper:
    22. 

  22.     Title: 'Swin Transformer: Hierarchical Vision Transformer using Shifted Windows'
    23. 

  23.     URL: https://arxiv.org/abs/2103.14030
    24. 

  24.   Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/backbones/swin.py#L524
    25. 

  25.   Framework: PyTorch
    26. 

  26. - Name: swin-small-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512
    27. 

  27.   In Collection: UPerNet
    28. 

  28.   Results:
    29. 

  29.     Task: Semantic Segmentation
    30. 

  30.     Dataset: ADE20K
    31. 

  31.     Metrics:
    32. 

  32.       mIoU: 47.72
    33. 

  33.       mIoU(ms+flip): 49.24
    34. 

  34.   Config: configs/swin/swin-small-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512.py
    35. 

  35.   Metadata:
    36. 

  36.     Training Data: ADE20K
    37. 

  37.     Batch Size: 16
    38. 

  38.     Architecture:
    39. 

  39.     - Swin-S
    40. 

  40.     - UPerNet
    41. 

  41.     Training Resources: 8x V100 GPUS
    42. 

  42.     Memory (GB): 6.17
    43. 

  43.   Weights: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_small_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K/upernet_swin_small_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K_20210526_192015-ee2fff1c.pth
    44. 

  44.   Training log: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_small_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K/upernet_swin_small_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K_20210526_192015.log.json
    45. 

  45.   Paper:
    46. 

  46.     Title: 'Swin Transformer: Hierarchical Vision Transformer using Shifted Windows'
    47. 

  47.     URL: https://arxiv.org/abs/2103.14030
    48. 

  48.   Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/backbones/swin.py#L524
    49. 

  49.   Framework: PyTorch
    50. 

  50. - Name: swin-base-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512
    51. 

  51.   In Collection: UPerNet
    52. 

  52.   Results:
    53. 

  53.     Task: Semantic Segmentation
    54. 

  54.     Dataset: ADE20K
    55. 

  55.     Metrics:
    56. 

  56.       mIoU: 47.99
    57. 

  57.       mIoU(ms+flip): 49.57
    58. 

  58.   Config: configs/swin/swin-base-patch4-window7-in1k-pre_upernet_8xb2-160k_ade20k-512x512.py
    59. 

  59.   Metadata:
    60. 

  60.     Training Data: ADE20K
    61. 

  61.     Batch Size: 16
    62. 

  62.     Architecture:
    63. 

  63.     - Swin-B
    64. 

  64.     - UPerNet
    65. 

  65.     Training Resources: 8x V100 GPUS
    66. 

  66.     Memory (GB): 7.61
    67. 

  67.   Weights: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K_20210526_192340-593b0e13.pth
    68. 

  68.   Training log: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_1K_20210526_192340.log.json
    69. 

  69.   Paper:
    70. 

  70.     Title: 'Swin Transformer: Hierarchical Vision Transformer using Shifted Windows'
    71. 

  71.     URL: https://arxiv.org/abs/2103.14030
    72. 

  72.   Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/backbones/swin.py#L524
    73. 

  73.   Framework: PyTorch
    74. 

  74. - Name: swin-base-patch4-window7-in22k-pre_upernet_8xb2-160k_ade20k-512x512
    75. 

  75.   In Collection: UPerNet
    76. 

  76.   Results:
    77. 

  77.     Task: Semantic Segmentation
    78. 

  78.     Dataset: ADE20K
    79. 

  79.     Metrics:
    80. 

  80.       mIoU: 50.13
    81. 

  81.       mIoU(ms+flip): 51.9
    82. 

  82.   Config: configs/swin/swin-base-patch4-window7-in22k-pre_upernet_8xb2-160k_ade20k-512x512.py
    83. 

  83.   Metadata:
    84. 

  84.     Training Data: ADE20K
    85. 

  85.     Batch Size: 16
    86. 

  86.     Architecture:
    87. 

  87.     - Swin-B
    88. 

  88.     - UPerNet
    89. 

  89.     Training Resources: 8x V100 GPUS
    90. 

  90.   Weights: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_22K/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_22K_20210526_211650-762e2178.pth
    91. 

  91.   Training log: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_22K/upernet_swin_base_patch4_window7_512x512_160k_ade20k_pretrain_224x224_22K_20210526_211650.log.json
    92. 

  92.   Paper:
    93. 

  93.     Title: 'Swin Transformer: Hierarchical Vision Transformer using Shifted Windows'
    94. 

  94.     URL: https://arxiv.org/abs/2103.14030
    95. 

  95.   Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/backbones/swin.py#L524
    96. 

  96.   Framework: PyTorch
    97. 

  97. - Name: swin-base-patch4-window12-in1k-384x384-pre_upernet_8xb2-160k_ade20k-512x512
    98. 

  98.   In Collection: UPerNet
    99. 

  99.   Results:
    100. 

  100.     Task: Semantic Segmentation
    101. 

  101.     Dataset: ADE20K
    102. 

  102.     Metrics:
    103. 

  103.       mIoU: 48.35
    104. 

  104.       mIoU(ms+flip): 49.65
    105. 

  105.   Config: configs/swin/swin-base-patch4-window12-in1k-384x384-pre_upernet_8xb2-160k_ade20k-512x512.py
    106. 

  106.   Metadata:
    107. 

  107.     Training Data: ADE20K
    108. 

  108.     Batch Size: 16
    109. 

  109.     Architecture:
    110. 

  110.     - Swin-B
    111. 

  111.     - UPerNet
    112. 

  112.     Training Resources: 8x V100 GPUS
    113. 

  113.     Memory (GB): 8.52
    114. 

  114.   Weights: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_1K/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_1K_20210531_132020-05b22ea4.pth
    115. 

  115.   Training log: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_1K/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_1K_20210531_132020.log.json
    116. 

  116.   Paper:
    117. 

  117.     Title: 'Swin Transformer: Hierarchical Vision Transformer using Shifted Windows'
    118. 

  118.     URL: https://arxiv.org/abs/2103.14030
    119. 

  119.   Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/backbones/swin.py#L524
    120. 

  120.   Framework: PyTorch
    121. 

  121. - Name: swin-base-patch4-window12-in22k-384x384-pre_upernet_8xb2-160k_ade20k-512x512
    122. 

  122.   In Collection: UPerNet
    123. 

  123.   Results:
    124. 

  124.     Task: Semantic Segmentation
    125. 

  125.     Dataset: ADE20K
    126. 

  126.     Metrics:
    127. 

  127.       mIoU: 50.76
    128. 

  128.       mIoU(ms+flip): 52.4
    129. 

  129.   Config: configs/swin/swin-base-patch4-window12-in22k-384x384-pre_upernet_8xb2-160k_ade20k-512x512.py
    130. 

  130.   Metadata:
    131. 

  131.     Training Data: ADE20K
    132. 

  132.     Batch Size: 16
    133. 

  133.     Architecture:
    134. 

  134.     - Swin-B
    135. 

  135.     - UPerNet
    136. 

  136.     Training Resources: 8x V100 GPUS
    137. 

  137.   Weights: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_22K/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_22K_20210531_125459-429057bf.pth
    138. 

  138.   Training log: https://download.openmmlab.com/mmsegmentation/v0.5/swin/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_22K/upernet_swin_base_patch4_window12_512x512_160k_ade20k_pretrain_384x384_22K_20210531_125459.log.json
    139. 

  139.   Paper:
    140. 

  140.     Title: 'Swin Transformer: Hierarchical Vision Transformer using Shifted Windows'
    141. 

  141.     URL: https://arxiv.org/abs/2103.14030
    142. 

  142.   Code: https://github.com/open-mmlab/mmsegmentation/blob/v0.17.0/mmseg/models/backbones/swin.py#L524
    143. 

  143.   Framework: PyTorch
    144.