Detectron2 throwing "Non-existent config key" error on CPU even though the key is present in config.yaml

[rudyoactiv]

I have trained Detectron2 for Page Layout Detection on a PC with GPU and it is giving satisfactory results.

If I try to use the trained model on non-CUDA Detectron2 environment on same PC, or a non-GPU PC I get the error:

raise KeyError("Non-existent config key: {}".format(full_key))
KeyError: 'Non-existent config key: MODEL.ROI_BOX_HEAD.FED_LOSS_FREQ_WEIGHT_POWER'

I have checked the config.yaml and it has:

MODEL:
  ROI_BOX_HEAD:
    FED_LOSS_FREQ_WEIGHT_POWER: 0.5

I have tried using pre-trained models with the non-GPU PC and they perform fine, only my own model raises the error specifically on non-GPU systems. I need to be able to use the trained model on a non-GPU work laptop.

[github-actions]

You've chosen to report an unexpected problem or bug. Unless you already know the root cause of it, please include details about it by filling the issue template.
The following information is missing: "Instructions To Reproduce the Issue and Full Logs";

[rudyoactiv]

Instructions To Reproduce the Issue:

Trained model on a GPU enabled PC, then use trained model on CPU-only environment or non-GPU PC.
Load config.yaml and model_final.pth for testing trained model.

Config.yaml

CUDNN_BENCHMARK: false
DATALOADER:
  ASPECT_RATIO_GROUPING: true
  FILTER_EMPTY_ANNOTATIONS: true
  NUM_WORKERS: 0
  REPEAT_THRESHOLD: 0.0
  SAMPLER_TRAIN: TrainingSampler
DATASETS:
  PRECOMPUTED_PROPOSAL_TOPK_TEST: 1000
  PRECOMPUTED_PROPOSAL_TOPK_TRAIN: 2000
  PROPOSAL_FILES_TEST: []
  PROPOSAL_FILES_TRAIN: []
  TEST: []
  TRAIN: []
GLOBAL:
  HACK: 1.0
INPUT:
  CROP:
    ENABLED: false
    SIZE:
    - 0.9
    - 0.9
    TYPE: relative_range
  FORMAT: BGR
  RANDOM_FLIP: horizontal
  MASK_FORMAT: polygon
  MAX_SIZE_TEST: 1333
  MAX_SIZE_TRAIN: 1333
  MIN_SIZE_TEST: 800
  MIN_SIZE_TRAIN:
  - 640
  - 672
  - 704
  - 736
  - 768
  - 800
  MIN_SIZE_TRAIN_SAMPLING: choice
MODEL:
  ANCHOR_GENERATOR:
    ANGLES:
    - - -90
      - 0
      - 90
    ASPECT_RATIOS:
    - - 0.5
      - 1.0
      - 2.0
    NAME: DefaultAnchorGenerator
    OFFSET: 0.0
    SIZES:
    - - 32
    - - 64
    - - 128
    - - 256
    - - 512
  BACKBONE:
    FREEZE_AT: 2
    NAME: build_resnet_fpn_backbone
  DEVICE: cuda
  FPN:
    FUSE_TYPE: sum
    IN_FEATURES:
    - res2
    - res3
    - res4
    - res5
    NORM: ''
    OUT_CHANNELS: 256
  KEYPOINT_ON: false
  LOAD_PROPOSALS: false
  MASK_ON: false
  META_ARCHITECTURE: GeneralizedRCNN
  PANOPTIC_FPN:
    COMBINE:
      ENABLED: true
      INSTANCES_CONFIDENCE_THRESH: 0.5
      OVERLAP_THRESH: 0.5
      STUFF_AREA_LIMIT: 4096
    INSTANCE_LOSS_WEIGHT: 1.0
  PIXEL_MEAN:
  - 103.53
  - 116.28
  - 123.675
  PIXEL_STD:
  - 1.0
  - 1.0
  - 1.0
  PROPOSAL_GENERATOR:
    MIN_SIZE: 0
    NAME: RPN
  RESNETS:
    DEFORM_MODULATED: false
    DEFORM_NUM_GROUPS: 1
    DEFORM_ON_PER_STAGE:
    - false
    - false
    - false
    - false
    DEPTH: 50
    NORM: FrozenBN
    NUM_GROUPS: 1
    OUT_FEATURES:
    - res2
    - res3
    - res4
    - res5
    RES2_OUT_CHANNELS: 256
    RES5_DILATION: 1
    STEM_OUT_CHANNELS: 64
    STRIDE_IN_1X1: true
    WIDTH_PER_GROUP: 64
  RETINANET:
    BBOX_REG_WEIGHTS:
    - 1.0
    - 1.0
    - 1.0
    - 1.0
    FOCAL_LOSS_ALPHA: 0.25
    FOCAL_LOSS_GAMMA: 2.0
    IN_FEATURES:
    - p3
    - p4
    - p5
    - p6
    - p7
    IOU_LABELS:
    - 0
    - -1
    - 1
    IOU_THRESHOLDS:
    - 0.4
    - 0.5
    NMS_THRESH_TEST: 0.5
    NUM_CLASSES: 2
    NUM_CONVS: 4
    PRIOR_PROB: 0.01
    SCORE_THRESH_TEST: 0.05
    SMOOTH_L1_LOSS_BETA: 0.1
    TOPK_CANDIDATES_TEST: 1000
  ROI_BOX_CASCADE_HEAD:
    BBOX_REG_WEIGHTS:
    - - 10.0
      - 10.0
      - 5.0
      - 5.0
    - - 20.0
      - 20.0
      - 10.0
      - 10.0
    - - 30.0
      - 30.0
      - 15.0
      - 15.0
    IOUS:
    - 0.5
    - 0.6
    - 0.7
  ROI_BOX_HEAD:
    BBOX_REG_WEIGHTS:
    - 10.0
    - 10.0
    - 5.0
    - 5.0
    CLS_AGNOSTIC_BBOX_REG: true
    CONV_DIM: 256
    FC_DIM: 1024
    NAME: FastRCNNConvFCHead
    NORM: ''
    NUM_CONV: 0
    NUM_FC: 2
    POOLER_RESOLUTION: 7
    POOLER_SAMPLING_RATIO: 0
    POOLER_TYPE: ROIAlignV2
    SMOOTH_L1_BETA: 0.0
    TRAIN_ON_PRED_BOXES: false
  ROI_HEADS:
    BATCH_SIZE_PER_IMAGE: 512
    IN_FEATURES:
    - p2
    - p3
    - p4
    - p5
    IOU_LABELS:
    - 0
    - 1
    IOU_THRESHOLDS:
    - 0.5
    NAME: StandardROIHeads
    NMS_THRESH_TEST: 0.5
    NUM_CLASSES: 2
    POSITIVE_FRACTION: 0.25
    PROPOSAL_APPEND_GT: true
    SCORE_THRESH_TEST: 0.05
  ROI_KEYPOINT_HEAD:
    CONV_DIMS:
    - 512
    - 512
    - 512
    - 512
    - 512
    - 512
    - 512
    - 512
    LOSS_WEIGHT: 1.0
    MIN_KEYPOINTS_PER_IMAGE: 1
    NAME: KRCNNConvDeconvUpsampleHead
    NORMALIZE_LOSS_BY_VISIBLE_KEYPOINTS: true
    NUM_KEYPOINTS: 17
    POOLER_RESOLUTION: 14
    POOLER_SAMPLING_RATIO: 0
    POOLER_TYPE: ROIAlignV2
  ROI_MASK_HEAD:
    CLS_AGNOSTIC_MASK: false
    CONV_DIM: 256
    NAME: MaskRCNNConvUpsampleHead
    NORM: ''
    NUM_CONV: 4
    POOLER_RESOLUTION: 14
    POOLER_SAMPLING_RATIO: 0
    POOLER_TYPE: ROIAlignV2
  RPN:
    BATCH_SIZE_PER_IMAGE: 256
    BBOX_REG_WEIGHTS:
    - 1.0
    - 1.0
    - 1.0
    - 1.0
    BOUNDARY_THRESH: -1
    HEAD_NAME: StandardRPNHead
    IN_FEATURES:
    - p2
    - p3
    - p4
    - p5
    - p6
    IOU_LABELS:
    - 0
    - -1
    - 1
    IOU_THRESHOLDS:
    - 0.3
    - 0.7
    LOSS_WEIGHT: 1.0
    NMS_THRESH: 0.7
    POSITIVE_FRACTION: 0.5
    POST_NMS_TOPK_TEST: 1000
    POST_NMS_TOPK_TRAIN: 1000
    PRE_NMS_TOPK_TEST: 1000
    PRE_NMS_TOPK_TRAIN: 2000
    SMOOTH_L1_BETA: 0.0
  SEM_SEG_HEAD:
    COMMON_STRIDE: 4
    CONVS_DIM: 128
    IGNORE_VALUE: 255
    IN_FEATURES:
    - p2
    - p3
    - p4
    - p5
    LOSS_WEIGHT: 1.0
    NAME: SemSegFPNHead
    NORM: GN
    NUM_CLASSES: 2
  WEIGHTS: detectron2://ImageNetPretrained/MSRA/R-50.pkl
OUTPUT_DIR: ./output
SEED: -1
SOLVER:
  BASE_LR: 0.0025
  BIAS_LR_FACTOR: 1.0
  CHECKPOINT_PERIOD: 20000
  GAMMA: 0.1
  IMS_PER_BATCH: 8
  LR_SCHEDULER_NAME: WarmupMultiStepLR
  MAX_ITER: 2000
  MOMENTUM: 0.9
  STEPS:
  - 1500
  - 1750
  WARMUP_FACTOR: 0.001
  WARMUP_ITERS: 400
  WARMUP_METHOD: linear
  WEIGHT_DECAY: 0.0001
  WEIGHT_DECAY_BIAS: 0.0001
  WEIGHT_DECAY_NORM: 0.0
TEST:
  AUG:
    ENABLED: false
    FLIP: true
    MAX_SIZE: 4000
    MIN_SIZES:
    - 400
    - 500
    - 600
    - 700
    - 800
    - 900
    - 1000
    - 1100
    - 1200
  DETECTIONS_PER_IMAGE: 100
  EVAL_PERIOD: 0
  EXPECTED_RESULTS: []
  KEYPOINT_OKS_SIGMAS: []
  PRECISE_BN:
    ENABLED: false
    NUM_ITER: 200
VERSION: 2
VIS_PERIOD: 0

Training Script:

#!/bin/bash

cd ../tools

python train_net.py \
    --dataset_name          paper-layout \
    --json_annotation_train ../data/qp/train.json \
    --image_path_train      ../data/qp \
    --json_annotation_val   ../data/qp/test.json \
    --image_path_val        ../data/qp/ \
    --config-file           ../configs/prima/fast_rcnn_R_50_FPN_3x.yaml \
    SOLVER.IMS_PER_BATCH    2 \
    OUTPUT_DIR              ../outputs/prima/fast_rcnn_R_50_FPN_3x/

Training works fine on GPU computer, and model also detects fine on GPU.
Log AFTER using trained model WITHOUT GPU:

(Traceback):

raise KeyError("Non-existent config key: {}".format(full_key))
KeyError: 'Non-existent config key: MODEL.ROI_BOX_HEAD.FED_LOSS_FREQ_WEIGHT_POWER'

[akshayrakate]

Hey @rudyoactiv,
I also went into the same error. But I resolved as follows,

    cfg = get_cfg() 
    cfg.set_new_allowed(True)  # Add this line before merging the file 
    cfg.merge_from_file(cfg_file_path)```


reference: https://github.com/rbgirshick/yacs/issues/25

[beniben0]

Hey @rudyoactiv, I also went into the same error. But I resolved as follows,

    cfg = get_cfg() 
    cfg.set_new_allowed(True)  # Add this line before merging the file 
    cfg.merge_from_file(cfg_file_path)```


reference: https://github.com/rbgirshick/yacs/issues/25

Thank you. This solved the problem for me

[Jaoan1]

Thankyou@rudyoactiv,
from fvcore.common.config import CfgNode

Create an empty CfgNode object

_C = CfgNode()

Allow creating new keys recursively

_C.set_new_allowed(True)

Merge settings from a configuration file into the CfgNode object

filename = "config.yaml"
_C.merge_from_file(filename)