search

justinkay / aldi

Official implementation of "Align and Distill: Unifying and Improving Domain Adaptive Object Detection"
https://aldi-daod.github.io/
38 stars 7 forks source link

How to run SADA on ALDI? #14

Closed Henryluoluoluo closed 4 months ago

Henryluoluoluo commented 4 months ago

I run my own data on aldi-cmt to train SADA, but get this error. However, it's not found on other model. This error is also found on branch extras. Snipaste_2024-05-14_00-20-46

justinkay commented 4 months ago

Can you post your config file and full training log please?

Henryluoluoluo commented 4 months ago

This is the config file and full training log(aldi-cmt) log.txt

AUG:
  INCLUDE_RANDOM_ERASING: true
  LABELED_MIC_AUG: false
  MIC_BLOCK_SIZE: 32
  MIC_RATIO: 0.5
  UNLABELED_MIC_AUG: false
  WEAK_INCLUDES_MULTISCALE: true
CUDNN_BENCHMARK: false
DATALOADER:
  ASPECT_RATIO_GROUPING: true
  FILTER_EMPTY_ANNOTATIONS: true
  NUM_WORKERS: 4
  REPEAT_THRESHOLD: 0.0
  SAMPLER_TRAIN: TrainingSampler
DATASETS:
  BATCH_CONTENTS:
  - labeled_weak
  - unlabeled_weak
  BATCH_RATIOS:
  - 1
  - 1
  PRECOMPUTED_PROPOSAL_TOPK_TEST: 1000
  PRECOMPUTED_PROPOSAL_TOPK_TRAIN: 2000
  PROPOSAL_FILES_TEST: []
  PROPOSAL_FILES_TRAIN: []
  TEST:
  - VisDark_test
  TRAIN:
  - VisLight
  UNLABELED:
  - VisDark_train
DOMAIN_ADAPT:
  LOSSES:
    LOC_LOSS_ENABLED: true
    QUALITY_LOSS_WEIGHT_ENABLED: false
  TEACHER:
    ENABLED: false
    PSEUDO_LABEL_METHOD: thresholding
    THRESHOLD: 0.8
EMA:
  ALPHA: 0.9996
  ENABLED: false
GLOBAL:
  HACK: 1.0
GRCNN:
  EFL: false
  EFL_LAMBDA:
  - 0.5
  - 0.5
  LEARN_ANCHORS_LABELED: false
  LEARN_ANCHORS_UNLABELED: false
  MODEL_TYPE: GAUSSIAN
  TAU:
  - 0.5
  - 0.5
INPUT:
  CROP:
    ENABLED: false
    SIZE:
    - 0.9
    - 0.9
    TYPE: relative_range
  FORMAT: BGR
  MASK_FORMAT: polygon
  MAX_SIZE_TEST: 1333
  MAX_SIZE_TRAIN: 1333
  MIN_SIZE_TEST: 1024
  MIN_SIZE_TRAIN:
  - 800
  - 832
  - 864
  - 896
  - 928
  - 960
  - 992
  - 1024
  MIN_SIZE_TRAIN_SAMPLING: choice
  RANDOM_FLIP: horizontal
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
  CMT:
    CONTRASTIVE_LOSS_WEIGHT: 0.05
    ENABLED: false
  DA:
    DIS_LOSS_WEIGHT: 0.05
    DIS_TYPE: p2
    ENABLED: false
  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: DARCNN
  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_LOSS_TYPE: smooth_l1
    BBOX_REG_WEIGHTS: &id002
    - 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
    NORM: ''
    NUM_CLASSES: 80
    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:
    - &id001
      - 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_LOSS_TYPE: smooth_l1
    BBOX_REG_LOSS_WEIGHT: 1.0
    BBOX_REG_WEIGHTS: *id001
    CLS_AGNOSTIC_BBOX_REG: false
    CONV_DIM: 256
    FC_DIM: 1024
    FED_LOSS_FREQ_WEIGHT_POWER: 0.5
    FED_LOSS_NUM_CLASSES: 50
    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
    USE_FED_LOSS: false
    USE_SIGMOID_CE: 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: 4
    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_LOSS_TYPE: smooth_l1
    BBOX_REG_LOSS_WEIGHT: 1.0
    BBOX_REG_WEIGHTS: *id002
    BOUNDARY_THRESH: -1
    CONV_DIMS:
    - -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
  SADA:
    COS_WEIGHT: 0.1
    DA_IMG_GRL_WEIGHT: 0.01
    DA_INS_GRL_WEIGHT: 0.1
    ENABLED: true
  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: 54
  WEIGHTS: detectron2://ImageNetPretrained/MSRA/R-50.pkl
OUTPUT_DIR: ../output/UDA_sada/
SEED: -1
SOLVER:
  AMP:
    ENABLED: true
  BACKWARD_AT_END: false
  BASE_LR: 0.01
  BASE_LR_END: 0.0
  BIAS_LR_FACTOR: 1.0
  CHECKPOINT_PERIOD: 6250
  CLIP_GRADIENTS:
    CLIP_TYPE: value
    CLIP_VALUE: 1.0
    ENABLED: false
    NORM_TYPE: 2.0
  GAMMA: 0.1
  IMS_PER_BATCH: 8
  IMS_PER_GPU: 8
  LR_SCHEDULER_NAME: WarmupMultiStepLR
  MAX_ITER: 24000
  MOMENTUM: 0.9
  NESTEROV: false
  NUM_DECAYS: 3
  OPTIMIZER: SGD
  REFERENCE_WORLD_SIZE: 0
  RESCALE_INTERVAL: false
  STEPS:
  - 18000
  WARMUP_FACTOR: 0.0016
  WARMUP_ITERS: 1000
  WARMUP_METHOD: linear
  WEIGHT_DECAY: 0.0001
  WEIGHT_DECAY_BIAS: null
  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: 1000
  EXPECTED_RESULTS: []
  KEYPOINT_OKS_SIGMAS: []
  PRECISE_BN:
    ENABLED: false
    NUM_ITER: 200
VERSION: 2
VIS_PERIOD: 625
Henryluoluoluo commented 4 months ago

This is the config file and full training log(aldi-extras) log.txt

AUG:
  LABELED_INCLUDE_RANDOM_ERASING: true
  LABELED_MIC_AUG: false
  MIC_BLOCK_SIZE: 32
  MIC_RATIO: 0.5
  UNLABELED_INCLUDE_RANDOM_ERASING: true
  UNLABELED_MIC_AUG: false
  WEAK_INCLUDES_MULTISCALE: true
CUDNN_BENCHMARK: false
DATALOADER:
  ASPECT_RATIO_GROUPING: true
  FILTER_EMPTY_ANNOTATIONS: true
  NUM_WORKERS: 4
  REPEAT_THRESHOLD: 0.0
  SAMPLER_TRAIN: TrainingSampler
DATASETS:
  BATCH_CONTENTS:
  - labeled_weak
  - unlabeled_weak
  BATCH_RATIOS:
  - 1
  - 1
  PRECOMPUTED_PROPOSAL_TOPK_TEST: 1000
  PRECOMPUTED_PROPOSAL_TOPK_TRAIN: 2000
  PROPOSAL_FILES_TEST: []
  PROPOSAL_FILES_TRAIN: []
  TEST:
  - VisDark_test
  TRAIN:
  - VisLight
  UNLABELED:
  - VisDark_train
DOMAIN_ADAPT:
  ALIGN:
    IMG_DA_ENABLED: false
    IMG_DA_IMPL: ours
    IMG_DA_LAYER: p2
    IMG_DA_WEIGHT: 0.01
    INS_DA_ENABLED: false
    INS_DA_WEIGHT: 0.01
    SADA_COS_WEIGHT: 0.1
    SADA_ENABLED: true
    SADA_IMG_GRL_WEIGHT: 0.01
    SADA_INS_GRL_WEIGHT: 0.1
  CLS_LOSS_TYPE: CE
  DISTILL:
    CLS_TMP: 1.0
    HARD_OBJ_ENABLED: false
    HARD_ROIH_CLS_ENABLED: false
    HARD_ROIH_REG_ENABLED: false
    HARD_RPN_REG_ENABLED: false
    HINT_CHANNELS: 512
    HINT_ENABLED: false
    HINT_LAYER: res3
    OBJ_ENABLED: false
    OBJ_TMP: 1.0
    ROIH_CLS_ENABLED: false
    ROIH_REG_ENABLED: false
    RPN_REG_ENABLED: false
  LOSSES:
    QUALITY_LOSS_WEIGHT_ENABLED: false
  TEACHER:
    ENABLED: false
    THRESHOLD: 0.8
EMA:
  ALPHA: 0.9996
  ENABLED: false
  LOAD_FROM_EMA_ON_START: true
GLOBAL:
  HACK: 1.0
GRCNN:
  EFL: false
  EFL_LAMBDA:
  - 0.5
  - 0.5
  LEARN_ANCHORS_LABELED: false
  LEARN_ANCHORS_UNLABELED: false
  MODEL_TYPE: GAUSSIAN
  TAU:
  - 0.5
  - 0.5
INPUT:
  CROP:
    ENABLED: false
    SIZE:
    - 0.9
    - 0.9
    TYPE: relative_range
  FORMAT: BGR
  MASK_FORMAT: polygon
  MAX_SIZE_TEST: 1333
  MAX_SIZE_TRAIN: 1333
  MIN_SIZE_TEST: 1024
  MIN_SIZE_TRAIN:
  - 800
  - 832
  - 864
  - 896
  - 928
  - 960
  - 992
  - 1024
  MIN_SIZE_TRAIN_SAMPLING: choice
  RANDOM_FLIP: horizontal
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: ALDI
  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_LOSS_TYPE: smooth_l1
    BBOX_REG_WEIGHTS: &id002
    - 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
    NORM: ''
    NUM_CLASSES: 80
    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:
    - &id001
      - 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_LOSS_TYPE: smooth_l1
    BBOX_REG_LOSS_WEIGHT: 1.0
    BBOX_REG_WEIGHTS: *id001
    CLS_AGNOSTIC_BBOX_REG: false
    CONV_DIM: 256
    FC_DIM: 1024
    FED_LOSS_FREQ_WEIGHT_POWER: 0.5
    FED_LOSS_NUM_CLASSES: 50
    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
    USE_FED_LOSS: false
    USE_SIGMOID_CE: 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: 4
    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_LOSS_TYPE: smooth_l1
    BBOX_REG_LOSS_WEIGHT: 1.0
    BBOX_REG_WEIGHTS: *id002
    BOUNDARY_THRESH: -1
    CONV_DIMS:
    - -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: 54
  UMT:
    ENABLED: false
  WEIGHTS: detectron2://ImageNetPretrained/MSRA/R-50.pkl
OUTPUT_DIR: ../output/UDA/UDA_SADA/
SEED: -1
SOLVER:
  AMP:
    ENABLED: true
  BACKWARD_AT_END: false
  BASE_LR: 0.01
  BASE_LR_END: 0.0
  BIAS_LR_FACTOR: 1.0
  CHECKPOINT_PERIOD: 5000
  CLIP_GRADIENTS:
    CLIP_TYPE: value
    CLIP_VALUE: 1.0
    ENABLED: false
    NORM_TYPE: 2.0
  GAMMA: 0.1
  IMS_PER_BATCH: 8
  IMS_PER_GPU: 8
  LR_SCHEDULER_NAME: WarmupMultiStepLR
  MAX_ITER: 30000
  MOMENTUM: 0.9
  NESTEROV: false
  NUM_DECAYS: 3
  OPTIMIZER: SGD
  REFERENCE_WORLD_SIZE: 0
  RESCALE_INTERVAL: false
  STEPS:
  - 29999
  WARMUP_FACTOR: 0.01
  WARMUP_ITERS: 0
  WARMUP_METHOD: linear
  WEIGHT_DECAY: 0.0001
  WEIGHT_DECAY_BIAS: null
  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: 1000
  EXPECTED_RESULTS: []
  KEYPOINT_OKS_SIGMAS: []
  PRECISE_BN:
    ENABLED: false
    NUM_ITER: 200
VERSION: 2
VIS_PERIOD: 1000
VIT:
  USE_ACT_CHECKPOINT: true
Henryluoluoluo commented 4 months ago

https://github.com/justinkay/aldi/blob/ef876ff967c8501eec39dbbf7158e6d2b5d4935f/trainer.py#L119 May be there are some bugs when train SADA.

justinkay commented 4 months ago

Thanks @Henryluoluoluo this is fixed by https://github.com/justinkay/aldi/commit/03ebb8b3455f71773c7638a6a8606d7415191938 on the extras branch.