fast-reid/configs/Base-MGN.yml

_BASE_: "Base-SBS.yml"

MODEL:
  META_ARCHITECTURE: 'MGN'

  FREEZE_LAYERS: ["backbone", "b1", "b2", "b3",]

  BACKBONE:
    WITH_NL: False

  HEADS:
    EMBEDDING_DIM: 256

  LOSSES:
    NAME: ("CrossEntropyLoss", "TripletLoss",)
    CE:
      EPSILON: 0.1
      SCALE: 1.0

    TRI:
      MARGIN: 0.0
      HARD_MINING: True
      NORM_FEAT: False
      SCALE: 1.0
v0.3 update Summary: 1. change DPP training in apex way; 2. make warmup scheduler by iter and lr scheduler by epoch; 3. replace random erasing with torchvision implementation; 4. naming modification in config file 2020-12-07 14:19:20 +08:00			`_BASE_: "Base-SBS.yml"`
update all config files 2020-07-06 16:48:14 +08:00
feat: add MGN model support MGN architecture and training config 2020-05-15 11:39:54 +08:00			`MODEL:`
			`META_ARCHITECTURE: 'MGN'`

update all config files 2020-07-06 16:48:14 +08:00			`FREEZE_LAYERS: ["backbone", "b1", "b2", "b3",]`
feat: add MGN model support MGN architecture and training config 2020-05-15 11:39:54 +08:00
			`BACKBONE:`
			`WITH_NL: False`
refactor mgn config 2020-09-02 18:01:07 +08:00
feat: add MGN model support MGN architecture and training config 2020-05-15 11:39:54 +08:00			`HEADS:`
change config file 2020-09-10 10:58:56 +08:00			`EMBEDDING_DIM: 256`
feat: add MGN model support MGN architecture and training config 2020-05-15 11:39:54 +08:00
			`LOSSES:`
			`NAME: ("CrossEntropyLoss", "TripletLoss",)`
			`CE:`
			`EPSILON: 0.1`
update mgn config 2020-09-01 16:16:40 +08:00			`SCALE: 1.0`
feat: add MGN model support MGN architecture and training config 2020-05-15 11:39:54 +08:00
			`TRI:`
			`MARGIN: 0.0`
			`HARD_MINING: True`
			`NORM_FEAT: False`
update mgn config 2020-09-01 16:16:40 +08:00			`SCALE: 1.0`
feat: add MGN model support MGN architecture and training config 2020-05-15 11:39:54 +08:00