res101_mx_3k.yml

# --------------------------------------------------------------
# R-FCN-3000@30FPS: Decoupling Classification and Detection
# Licensed under The Apache-2.0 License [see LICENSE for details]
# by Mahyar Najibi, Bharat Singh and Hengduo Li
# --------------------------------------------------------------
---
MXNET_VERSION: "mxnet"
output_path: "./output/chips_resnet101_3k"
symbol: resnet_mx_101_e2e_3k
#gpus: '0,1,2,3,4,5,6,7'
gpus: '0'
CLASS_AGNOSTIC: true
default:
  frequent: 100
  kvstore: device
network:
  pretrained: "./data/pretrained_model/soft"
  pretrained_epoch: 0
  PIXEL_MEANS:
  - 103.939
  - 116.779
  - 123.68
  RPN_FEAT_STRIDE: 16
  FIXED_PARAMS:
  - conv0
  - bn0
  - stage1

  ANCHOR_RATIOS:
  - 0.5
  - 1
  - 2
  ANCHOR_SCALES:
  - 4
  - 8
  - 16
  - 24
  NUM_ANCHORS: 12
dataset:
  NUM_CLASSES: 3130
  dataset: imagenet
  dataset_path: "./data/imagenet"
  image_set: fall11_whole
  root_path: "./data"
  test_image_set: ILSVRC2013_DET_val
  proposal: rpn
TRAIN:
  # Whether to use C++ or python code for chip generation
  CPP_CHIPS: false
  # Whether to extract negative chips during training
  USE_NEG_CHIPS: true

  # Multi-processing params
  NUM_PROCESS: 64
  NUM_THREAD: 8

  # Whether to train with segmentation mask
  WITH_MASK: false

  # Training scales(high res scale, medium res scale, smallest size in the pyramid)
  SCALES:
  - 512.0

  # Valid ranges in each scale
  VALID_RANGES:
  - !!python/tuple [0,-1]

  visualize: false
  # Path to save the visualization images
  visualization_path: 'debug/visualization'
  # Number of epochs between each visualization
  visualization_freq: 100
  

  lr: 0.015 #0.002 #0.0005
  lr_step: '5.33'
  warmup: true
  fp16: true
  warmup_lr: 0.0005 #0.00005
  wd: 0.0001
  scale: 100.0
  warmup_step: 1000 #4000 #1000
  begin_epoch: 0
  end_epoch: 7 #9
  model_prefix: 'rcnn'

  # whether flip image
  FLIP: true
  # whether shuffle image
  SHUFFLE: true
  # whether use OHEM
  ENABLE_OHEM: true
  # size of images for each device, 2 for rcnn, 1 for rpn and e2e
  BATCH_IMAGES: 16
  # e2e changes behavior of anchor loader and metric
  END2END: true
  # R-CNN
  # rcnn rois batch size
  BATCH_ROIS: -1
  BATCH_ROIS_OHEM: 256
  # rcnn rois sampling params
  FG_FRACTION: 0.25
  FG_THRESH: 0.5
  BG_THRESH_HI: 0.5
  BG_THRESH_LO: 0.0
  # rcnn bounding box regression params
  BBOX_REGRESSION_THRESH: 0.5
  BBOX_WEIGHTS:
  - 1.0
  - 1.0
  - 1.0
  - 1.0

  # RPN anchor loader
  # rpn anchors batch size
  RPN_BATCH_SIZE: 256
  # rpn anchors sampling params
  RPN_FG_FRACTION: 0.5
  RPN_POSITIVE_OVERLAP: 0.5
  RPN_NEGATIVE_OVERLAP: 0.4
  RPN_CLOBBER_POSITIVES: false
  # rpn bounding box regression params
  RPN_BBOX_WEIGHTS:
  - 1.0
  - 1.0
  - 1.0
  - 1.0
  RPN_POSITIVE_WEIGHT: -1.0
  # used for end2end training
  # RPN proposal
  CXX_PROPOSAL: false
  RPN_NMS_THRESH: 0.7
  RPN_PRE_NMS_TOP_N: 6000
  RPN_POST_NMS_TOP_N: 300
  RPN_MIN_SIZE: 0
  # approximate bounding box regression
  BBOX_NORMALIZATION_PRECOMPUTED: true
  BBOX_MEANS:
  - 0.0
  - 0.0
  - 0.0
  - 0.0
  BBOX_STDS:
  - 0.1
  - 0.1
  - 0.2
  - 0.2
TEST:
  # Maximum number of detections per image
  # Set to -1 to disable
  MAX_PER_IMAGE: 200

  # Whether to do multi-scale inference
  SCALES:
  - !!python/tuple [512, 512]

  # Number of images per gpu for each scale
  BATCH_IMAGES:
  - 6

  # Number of concurrent jobs used for inference
  # if greater than 1, the roidb is distributed over
  # concurrent jobs to increase throughput
  CONCURRENT_JOBS: 3

  # Ranges to specify valid proposal length
  # in each of the test scale, square area
  # would be computed based on the lengths
  # to invalidate, -1 means unbounded, use
  # -1 everywhere if you want to have all proposals
  VALID_RANGES:
  - !!python/tuple [0,-1]

  # Use rpn to generate proposal
  HAS_RPN: true

  # RPN proposal
  RPN_NMS_THRESH: 0.7
  RPN_PRE_NMS_TOP_N: 6000
  RPN_POST_NMS_TOP_N: 300
  RPN_MIN_SIZE: 0
  PROPOSAL_NMS_THRESH: 0.7
  PROPOSAL_PRE_NMS_TOP_N: 20000
  PROPOSAL_POST_NMS_TOP_N: 2000
  PROPOSAL_MIN_SIZE: 0

  # RCNN nms
  NMS: -1 #0.45
  NMS_SIGMA: 0.55
  TEST_EPOCH: 7

  # VISUALIZATION CONFIG
  VISUALIZATION_PATH: './debug/visualization'
  # Whether to visualize all intermediate scales
  # before aggregation (when doing multi-scale inference)
  # If False, only final detections are saved to 
  # VISUALIZATION_PATH
  VISUALIZE_INTERMEDIATE_SCALES: True

  # PROPOSAL EXTRACTION FLAGS
  # If true only would extract proposals
  EXTRACT_PROPOSALS: false

  # The folder path to be used for saving proposals
  PROPOSAL_SAVE_PATH: 'output/proposals'

  # Number of proposals extracted per scale
  # SCALES and BATCH_IMAGES above would be used to
  # Specify scales and number of images per batch for
  # each scale, no valid ranges would be applied for
  # aggregating proposals
  N_PROPOSAL_PER_SCALE: 300