I ran the SpaceNet_train.py file using the weights file you provided. (The learning step is omitted.)
However, unlike the code, random images are not displayed, and mAP is always calculated as 0.0.
Actually, I'm a beginner who just studied deep learning, so I can't figure out which part is wrong.
I am attaching the file(convert txt to py) I ran, so please tell me where the problem is.
Thank you always.
Configurations:
BACKBONE resnet50
BACKBONE_STRIDES [4, 8, 16, 32, 64]
BATCH_SIZE 1
BBOX_STD_DEV [0.1 0.1 0.2 0.2]
COMPUTE_BACKBONE_SHAPE None
DETECTION_MAX_INSTANCES 350
DETECTION_MIN_CONFIDENCE 0.7
DETECTION_NMS_THRESHOLD 0.3
FPN_CLASSIF_FC_LAYERS_SIZE 1024
GPU_COUNT 1
GRADIENT_CLIP_NORM 5.0
IMAGES_PER_GPU 1
IMAGE_CHANNEL_COUNT 3
IMAGE_MAX_DIM 640
IMAGE_META_SIZE 14
IMAGE_MIN_DIM 640
IMAGE_MIN_SCALE 0
IMAGE_RESIZE_MODE square
IMAGE_SHAPE [640 640 3]
LEARNING_MOMENTUM 0.9
LEARNING_RATE 0.001
LOSS_WEIGHTS {'rpn_class_loss': 1.0, 'rpn_bbox_loss': 1.0, 'mrcnn_class_loss': 1.0, 'mrcnn_bbox_loss': 1.0, 'mrcnn_mask_loss': 1.0}
MASK_POOL_SIZE 14
MASK_SHAPE [28, 28]
MAX_GT_INSTANCES 250
MEAN_PIXEL [123.7 116.8 103.9]
MINI_MASK_SHAPE (56, 56)
NAME SpaceNet
NUM_CLASSES 2
POOL_SIZE 7
POST_NMS_ROIS_INFERENCE 1000
POST_NMS_ROIS_TRAINING 2000
PRE_NMS_LIMIT 6000
ROI_POSITIVE_RATIO 0.33
RPN_ANCHOR_RATIOS [0.25, 1, 4]
RPN_ANCHOR_SCALES (8, 16, 32, 64, 128)
RPN_ANCHOR_STRIDE 1
RPN_BBOX_STD_DEV [0.1 0.1 0.2 0.2]
RPN_NMS_THRESHOLD 0.7
RPN_TRAIN_ANCHORS_PER_IMAGE 256
STEPS_PER_EPOCH 500
TOP_DOWN_PYRAMID_SIZE 256
TRAIN_BN False
TRAIN_ROIS_PER_IMAGE 32
USE_MINI_MASK True
USE_RPN_ROIS True
VALIDATION_STEPS 50
WEIGHT_DECAY 0.0001
1165
C:\Users\Seo\anaconda3\envs\mrcnn\lib\site-packages\mask_rcnn-2.1-py3.6.egg\mrcnn\visualize.py:56: UserWarning:
Matplotlib is currently using agg, which is a non-GUI backend, so cannot show the figure.
1336
788
737
WARNING:tensorflow:From C:\Users\Seo\anaconda3\envs\mrcnn\lib\site-packages\tensorflow\python\framework\op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be
removed in a future version.
Instructions for updating:
Colocations handled automatically by placer.
WARNING:tensorflow:From C:\Users\Seo\anaconda3\envs\mrcnn\lib\site-packages\keras\backend\tensorflow_backend.py:1154: calling reduce_max_v1 (from tensorflow.python.ops.math_ops) with keep_dims is deprecate
d and will be removed in a future version.
Instructions for updating:
keep_dims is deprecated, use keepdims instead
WARNING:tensorflow:From C:\Users\Seo\anaconda3\envs\mrcnn\lib\site-packages\keras\backend\tensorflow_backend.py:1188: calling reduce_sum_v1 (from tensorflow.python.ops.math_ops) with keep_dims is deprecate
d and will be removed in a future version.
Instructions for updating:
keep_dims is deprecated, use keepdims instead
WARNING:tensorflow:From C:\Users\Seo\anaconda3\envs\mrcnn\lib\site-packages\mask_rcnn-2.1-py3.6.egg\mrcnn\model.py:772: to_float (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a
future version.
Instructions for updating:
Use tf.cast instead.
Loading weights from D:\mask_rcnn_spacenet_0151.h5
2021-11-09 01:06:42.125108: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
2021-11-09 01:06:42.267542: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1433] Found device 0 with properties:
name: GeForce RTX 3060 Ti major: 8 minor: 6 memoryClockRate(GHz): 1.695
pciBusID: 0000:01:00.0
totalMemory: 8.00GiB freeMemory: 6.99GiB
2021-11-09 01:06:42.268127: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1512] Adding visible gpu devices: 0
2021-11-09 01:10:02.188573: I tensorflow/core/common_runtime/gpu/gpu_device.cc:984] Device interconnect StreamExecutor with strength 1 edge matrix:
2021-11-09 01:10:02.188862: I tensorflow/core/common_runtime/gpu/gpu_device.cc:990] 0
2021-11-09 01:10:02.189157: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1003] 0: N
2021-11-09 01:10:02.189519: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 6714 MB memory) -> physical GPU (device: 0,
name: GeForce RTX 3060 Ti, pci bus id: 0000:01:00.0, compute capability: 8.6)
original_image shape: (640, 640, 3) min: 6.00000 max: 250.00000 uint8
image_meta shape: (14,) min: 0.00000 max: 650.00000 float64
gt_class_id shape: (0,) min: max: int32
gt_bbox shape: (0, 4) min: max: int32
gt_mask shape: (640, 640, 0) min: max: bool
*** No instances to display ***
C:\Users\Seo\anaconda3\envs\mrcnn\lib\site-packages\mask_rcnn-2.1-py3.6.egg\mrcnn\visualize.py:167: UserWarning:
Matplotlib is currently using agg, which is a non-GUI backend, so cannot show the figure.
Processing 1 images
image shape: (640, 640, 3) min: 6.00000 max: 250.00000 uint8
molded_images shape: (1, 640, 640, 3) min: -117.70000 max: 133.20000 float64
image_metas shape: (1, 14) min: 0.00000 max: 640.00000 int32
anchors shape: (1, 102300, 4) min: -0.20031 max: 1.10016 float32
2021-11-09 01:10:04.638200: I tensorflow/stream_executor/dso_loader.cc:152] successfully opened CUDA library cublas64_100.dll locally
*** No instances to display ***
mAP: 0.
SpaceNet_train_modified.txt
0