metafile.yml 8.1 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246
  1. Collections:
  2. - Name: Guided Anchoring
  3. Metadata:
  4. Training Data: COCO
  5. Training Techniques:
  6. - SGD with Momentum
  7. - Weight Decay
  8. Training Resources: 8x V100 GPUs
  9. Architecture:
  10. - FPN
  11. - Guided Anchoring
  12. - ResNet
  13. Paper:
  14. URL: https://arxiv.org/abs/1901.03278
  15. Title: 'Region Proposal by Guided Anchoring'
  16. README: configs/guided_anchoring/README.md
  17. Code:
  18. URL: https://github.com/open-mmlab/mmdetection/blob/v2.0.0/mmdet/models/dense_heads/ga_retina_head.py#L10
  19. Version: v2.0.0
  20. Models:
  21. - Name: ga-rpn_r50-caffe_fpn_1x_coco
  22. In Collection: Guided Anchoring
  23. Config: configs/guided_anchoring/ga-rpn_r50-caffe_fpn_1x_coco.py
  24. Metadata:
  25. Training Memory (GB): 5.3
  26. inference time (ms/im):
  27. - value: 63.29
  28. hardware: V100
  29. backend: PyTorch
  30. batch size: 1
  31. mode: FP32
  32. resolution: (800, 1333)
  33. Epochs: 12
  34. Results:
  35. - Task: Region Proposal
  36. Dataset: COCO
  37. Metrics:
  38. AR@1000: 68.4
  39. Weights: https://download.openmmlab.com/mmdetection/v2.0/guided_anchoring/ga_rpn_r50_caffe_fpn_1x_coco/ga_rpn_r50_caffe_fpn_1x_coco_20200531-899008a6.pth
  40. - Name: ga-rpn_r101-caffe_fpn_1x_coco
  41. In Collection: Guided Anchoring
  42. Config: configs/guided_anchoring/ga-rpn_r101-caffe_fpn_1x_coco.py
  43. Metadata:
  44. Training Memory (GB): 7.3
  45. inference time (ms/im):
  46. - value: 76.92
  47. hardware: V100
  48. backend: PyTorch
  49. batch size: 1
  50. mode: FP32
  51. resolution: (800, 1333)
  52. Epochs: 12
  53. Results:
  54. - Task: Region Proposal
  55. Dataset: COCO
  56. Metrics:
  57. AR@1000: 69.5
  58. Weights: https://download.openmmlab.com/mmdetection/v2.0/guided_anchoring/ga_rpn_r101_caffe_fpn_1x_coco/ga_rpn_r101_caffe_fpn_1x_coco_20200531-ca9ba8fb.pth
  59. - Name: ga-rpn_x101-32x4d_fpn_1x_coco
  60. In Collection: Guided Anchoring
  61. Config: configs/guided_anchoring/ga-rpn_x101-32x4d_fpn_1x_coco.py
  62. Metadata:
  63. Training Memory (GB): 8.5
  64. inference time (ms/im):
  65. - value: 100
  66. hardware: V100
  67. backend: PyTorch
  68. batch size: 1
  69. mode: FP32
  70. resolution: (800, 1333)
  71. Epochs: 12
  72. Results:
  73. - Task: Region Proposal
  74. Dataset: COCO
  75. Metrics:
  76. AR@1000: 70.6
  77. Weights: https://download.openmmlab.com/mmdetection/v2.0/guided_anchoring/ga_rpn_x101_32x4d_fpn_1x_coco/ga_rpn_x101_32x4d_fpn_1x_coco_20200220-c28d1b18.pth
  78. - Name: ga-rpn_x101-64x4d_fpn_1x_coco
  79. In Collection: Guided Anchoring
  80. Config: configs/guided_anchoring/ga-rpn_x101-64x4d_fpn_1x_coco.py
  81. Metadata:
  82. Training Memory (GB): 7.1
  83. inference time (ms/im):
  84. - value: 133.33
  85. hardware: V100
  86. backend: PyTorch
  87. batch size: 1
  88. mode: FP32
  89. resolution: (800, 1333)
  90. Epochs: 12
  91. Results:
  92. - Task: Region Proposal
  93. Dataset: COCO
  94. Metrics:
  95. AR@1000: 70.6
  96. Weights: https://download.openmmlab.com/mmdetection/v2.0/guided_anchoring/ga_rpn_x101_64x4d_fpn_1x_coco/ga_rpn_x101_64x4d_fpn_1x_coco_20200225-3c6e1aa2.pth
  97. - Name: ga-faster-rcnn_r50-caffe_fpn_1x_coco
  98. In Collection: Guided Anchoring
  99. Config: configs/guided_anchoring/ga-faster-rcnn_r50-caffe_fpn_1x_coco.py
  100. Metadata:
  101. Training Memory (GB): 5.5
  102. Epochs: 12
  103. Results:
  104. - Task: Object Detection
  105. Dataset: COCO
  106. Metrics:
  107. box AP: 39.6
  108. Weights: https://download.openmmlab.com/mmdetection/v2.0/guided_anchoring/ga_faster_r50_caffe_fpn_1x_coco/ga_faster_r50_caffe_fpn_1x_coco_20200702_000718-a11ccfe6.pth
  109. - Name: ga-faster-rcnn_r101-caffe_fpn_1x_coco
  110. In Collection: Guided Anchoring
  111. Config: configs/guided_anchoring/ga-faster-rcnn_r101-caffe_fpn_1x_coco.py
  112. Metadata:
  113. Training Memory (GB): 7.5
  114. Epochs: 12
  115. Results:
  116. - Task: Object Detection
  117. Dataset: COCO
  118. Metrics:
  119. box AP: 41.5
  120. Weights: https://download.openmmlab.com/mmdetection/v2.0/guided_anchoring/ga_faster_r101_caffe_fpn_1x_coco/ga_faster_r101_caffe_fpn_1x_coco_bbox_mAP-0.415_20200505_115528-fb82e499.pth
  121. - Name: ga-faster-rcnn_x101-32x4d_fpn_1x_coco
  122. In Collection: Guided Anchoring
  123. Config: configs/guided_anchoring/ga-faster-rcnn_x101-32x4d_fpn_1x_coco.py
  124. Metadata:
  125. Training Memory (GB): 8.7
  126. inference time (ms/im):
  127. - value: 103.09
  128. hardware: V100
  129. backend: PyTorch
  130. batch size: 1
  131. mode: FP32
  132. resolution: (800, 1333)
  133. Epochs: 12
  134. Results:
  135. - Task: Object Detection
  136. Dataset: COCO
  137. Metrics:
  138. box AP: 43.0
  139. Weights: https://download.openmmlab.com/mmdetection/v2.0/guided_anchoring/ga_faster_x101_32x4d_fpn_1x_coco/ga_faster_x101_32x4d_fpn_1x_coco_20200215-1ded9da3.pth
  140. - Name: ga-faster-rcnn_x101-64x4d_fpn_1x_coco
  141. In Collection: Guided Anchoring
  142. Config: configs/guided_anchoring/ga-faster-rcnn_x101-64x4d_fpn_1x_coco.py
  143. Metadata:
  144. Training Memory (GB): 11.8
  145. inference time (ms/im):
  146. - value: 136.99
  147. hardware: V100
  148. backend: PyTorch
  149. batch size: 1
  150. mode: FP32
  151. resolution: (800, 1333)
  152. Epochs: 12
  153. Results:
  154. - Task: Object Detection
  155. Dataset: COCO
  156. Metrics:
  157. box AP: 43.9
  158. Weights: https://download.openmmlab.com/mmdetection/v2.0/guided_anchoring/ga_faster_x101_64x4d_fpn_1x_coco/ga_faster_x101_64x4d_fpn_1x_coco_20200215-0fa7bde7.pth
  159. - Name: ga-retinanet_r50-caffe_fpn_1x_coco
  160. In Collection: Guided Anchoring
  161. Config: configs/guided_anchoring/ga-retinanet_r50-caffe_fpn_1x_coco.py
  162. Metadata:
  163. Training Memory (GB): 3.5
  164. inference time (ms/im):
  165. - value: 59.52
  166. hardware: V100
  167. backend: PyTorch
  168. batch size: 1
  169. mode: FP32
  170. resolution: (800, 1333)
  171. Epochs: 12
  172. Results:
  173. - Task: Object Detection
  174. Dataset: COCO
  175. Metrics:
  176. box AP: 36.9
  177. Weights: https://download.openmmlab.com/mmdetection/v2.0/guided_anchoring/ga_retinanet_r50_caffe_fpn_1x_coco/ga_retinanet_r50_caffe_fpn_1x_coco_20201020-39581c6f.pth
  178. - Name: ga-retinanet_r101-caffe_fpn_1x_coco
  179. In Collection: Guided Anchoring
  180. Config: configs/guided_anchoring/ga-retinanet_r101-caffe_fpn_1x_coco.py
  181. Metadata:
  182. Training Memory (GB): 5.5
  183. inference time (ms/im):
  184. - value: 77.52
  185. hardware: V100
  186. backend: PyTorch
  187. batch size: 1
  188. mode: FP32
  189. resolution: (800, 1333)
  190. Epochs: 12
  191. Results:
  192. - Task: Object Detection
  193. Dataset: COCO
  194. Metrics:
  195. box AP: 39.0
  196. Weights: https://download.openmmlab.com/mmdetection/v2.0/guided_anchoring/ga_retinanet_r101_caffe_fpn_1x_coco/ga_retinanet_r101_caffe_fpn_1x_coco_20200531-6266453c.pth
  197. - Name: ga-retinanet_x101-32x4d_fpn_1x_coco
  198. In Collection: Guided Anchoring
  199. Config: configs/guided_anchoring/ga-retinanet_x101-32x4d_fpn_1x_coco.py
  200. Metadata:
  201. Training Memory (GB): 6.9
  202. inference time (ms/im):
  203. - value: 94.34
  204. hardware: V100
  205. backend: PyTorch
  206. batch size: 1
  207. mode: FP32
  208. resolution: (800, 1333)
  209. Epochs: 12
  210. Results:
  211. - Task: Object Detection
  212. Dataset: COCO
  213. Metrics:
  214. box AP: 40.5
  215. Weights: https://download.openmmlab.com/mmdetection/v2.0/guided_anchoring/ga_retinanet_x101_32x4d_fpn_1x_coco/ga_retinanet_x101_32x4d_fpn_1x_coco_20200219-40c56caa.pth
  216. - Name: ga-retinanet_x101-64x4d_fpn_1x_coco
  217. In Collection: Guided Anchoring
  218. Config: configs/guided_anchoring/ga-retinanet_x101-64x4d_fpn_1x_coco.py
  219. Metadata:
  220. Training Memory (GB): 9.9
  221. inference time (ms/im):
  222. - value: 129.87
  223. hardware: V100
  224. backend: PyTorch
  225. batch size: 1
  226. mode: FP32
  227. resolution: (800, 1333)
  228. Epochs: 12
  229. Results:
  230. - Task: Object Detection
  231. Dataset: COCO
  232. Metrics:
  233. box AP: 41.3
  234. Weights: https://download.openmmlab.com/mmdetection/v2.0/guided_anchoring/ga_retinanet_x101_64x4d_fpn_1x_coco/ga_retinanet_x101_64x4d_fpn_1x_coco_20200226-ef9f7f1f.pth