tangchongyang

基于pytorch的目标检测（rcnn+resnet50，行人数据集）

import torch
torch.cuda.is_available()

True

!pip install cython
# Install pycocotools, the version by default in Colab
# has a bug fixed in https://github.com/cocodataset/cocoapi/pull/354
!pip install -U 'git+https://github.com/cocodataset/cocoapi.git#subdirectory=PythonAPI'

Requirement already satisfied: cython in /home/qy/.conda/envs/pytorch/lib/python3.6/site-packages (0.29.19)
Collecting git+https://github.com/cocodataset/cocoapi.git#subdirectory=PythonAPI
  Cloning https://github.com/cocodataset/cocoapi.git to /tmp/pip-req-build-al3q5nlj
  Running command git clone -q https://github.com/cocodataset/cocoapi.git /tmp/pip-req-build-al3q5nlj
Requirement already satisfied, skipping upgrade: setuptools>=18.0 in /home/qy/.conda/envs/pytorch/lib/python3.6/site-packages (from pycocotools==2.0) (46.4.0.post20200518)
Requirement already satisfied, skipping upgrade: cython>=0.27.3 in /home/qy/.conda/envs/pytorch/lib/python3.6/site-packages (from pycocotools==2.0) (0.29.19)
Collecting matplotlib>=2.1.0
  Downloading matplotlib-3.2.1-cp36-cp36m-manylinux1_x86_64.whl (12.4 MB)
[K     |████████████████████████████████| 12.4 MB 7.0 MB/s eta 0:00:01
[?25hCollecting kiwisolver>=1.0.1
  Downloading kiwisolver-1.2.0-cp36-cp36m-manylinux1_x86_64.whl (88 kB)
[K     |████████████████████████████████| 88 kB 837 kB/s  eta 0:00:01
[?25hCollecting pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1
  Downloading pyparsing-2.4.7-py2.py3-none-any.whl (67 kB)
[K     |████████████████████████████████| 67 kB 867 kB/s s eta 0:00:01
[?25hRequirement already satisfied, skipping upgrade: python-dateutil>=2.1 in /home/qy/.conda/envs/pytorch/lib/python3.6/site-packages (from matplotlib>=2.1.0->pycocotools==2.0) (2.8.1)
Collecting cycler>=0.10
  Downloading cycler-0.10.0-py2.py3-none-any.whl (6.5 kB)
Requirement already satisfied, skipping upgrade: numpy>=1.11 in /home/qy/.conda/envs/pytorch/lib/python3.6/site-packages (from matplotlib>=2.1.0->pycocotools==2.0) (1.18.1)
Requirement already satisfied, skipping upgrade: six>=1.5 in /home/qy/.conda/envs/pytorch/lib/python3.6/site-packages (from python-dateutil>=2.1->matplotlib>=2.1.0->pycocotools==2.0) (1.14.0)
Building wheels for collected packages: pycocotools
  Building wheel for pycocotools (setup.py) ... [?25ldone
[?25h  Created wheel for pycocotools: filename=pycocotools-2.0-cp36-cp36m-linux_x86_64.whl size=282788 sha256=5e2c6293ef8c17974cee3aeccb1cce1b9c57e5890a4670db02b4e46cbe90bef6
  Stored in directory: /tmp/pip-ephem-wheel-cache-p2ts_ztv/wheels/25/c1/63/8bee2969883497d2785c9bdbe4e89cae5efc59521553d528bf
Successfully built pycocotools
Installing collected packages: kiwisolver, pyparsing, cycler, matplotlib, pycocotools
Successfully installed cycler-0.10.0 kiwisolver-1.2.0 matplotlib-3.2.1 pycocotools-2.0 pyparsing-2.4.7

from PIL import Image
Image.open('PennFudanPed/PNGImages/FudanPed00001.png')

mask = Image.open('PennFudanPed/PedMasks/FudanPed00001_mask.png')
# each mask instance has a different color, from zero to N, where
# N is the number of instances. In order to make visualization easier,
# let's adda color palette to the mask.
mask.putpalette([
    0, 0, 0, # black background
    255, 0, 0, # index 1 is red
    255, 255, 0, # index 2 is yellow
    255, 153, 0, # index 3 is orange
])
mask

import os
import numpy as np
import torch
import torch.utils.data
from PIL import Image


class PennFudanDataset(torch.utils.data.Dataset):
    def __init__(self, root, transforms=None):
        self.root = root
        self.transforms = transforms
        # load all image files, sorting them to
        # ensure that they are aligned
        self.imgs = list(sorted(os.listdir(os.path.join(root, "PNGImages"))))
        self.masks = list(sorted(os.listdir(os.path.join(root, "PedMasks"))))

    def __getitem__(self, idx):
        # load images ad masks
        img_path = os.path.join(self.root, "PNGImages", self.imgs[idx])
        mask_path = os.path.join(self.root, "PedMasks", self.masks[idx])
        img = Image.open(img_path).convert("RGB")
        # note that we haven't converted the mask to RGB,
        # because each color corresponds to a different instance
        # with 0 being background
        mask = Image.open(mask_path)

        mask = np.array(mask)
        # instances are encoded as different colors
        obj_ids = np.unique(mask)
        # first id is the background, so remove it
        obj_ids = obj_ids[1:]

        # split the color-encoded mask into a set
        # of binary masks
        masks = mask == obj_ids[:, None, None]

        # get bounding box coordinates for each mask
        num_objs = len(obj_ids)
        boxes = []
        for i in range(num_objs):
            pos = np.where(masks[i])
            xmin = np.min(pos[1])
            xmax = np.max(pos[1])
            ymin = np.min(pos[0])
            ymax = np.max(pos[0])
            boxes.append([xmin, ymin, xmax, ymax])

        boxes = torch.as_tensor(boxes, dtype=torch.float32)
        # there is only one class
        labels = torch.ones((num_objs,), dtype=torch.int64)
        masks = torch.as_tensor(masks, dtype=torch.uint8)

        image_id = torch.tensor([idx])
        area = (boxes[:, 3] - boxes[:, 1]) * (boxes[:, 2] - boxes[:, 0])
        # suppose all instances are not crowd
        iscrowd = torch.zeros((num_objs,), dtype=torch.int64)

        target = {}
        target["boxes"] = boxes
        target["labels"] = labels
        target["masks"] = masks
        target["image_id"] = image_id
        target["area"] = area
        target["iscrowd"] = iscrowd

        if self.transforms is not None:
            img, target = self.transforms(img, target)

        return img, target

    def __len__(self):
        return len(self.imgs)

dataset = PennFudanDataset('PennFudanPed/')
dataset[0]

(,
 {'boxes': tensor([[159., 181., 301., 430.],
          [419., 170., 534., 485.]]),
  'labels': tensor([1, 1]),
  'masks': tensor([[[0, 0, 0,  ..., 0, 0, 0],
           [0, 0, 0,  ..., 0, 0, 0],
           [0, 0, 0,  ..., 0, 0, 0],
           ...,
           [0, 0, 0,  ..., 0, 0, 0],
           [0, 0, 0,  ..., 0, 0, 0],
           [0, 0, 0,  ..., 0, 0, 0]],
  
          [[0, 0, 0,  ..., 0, 0, 0],
           [0, 0, 0,  ..., 0, 0, 0],
           [0, 0, 0,  ..., 0, 0, 0],
           ...,
           [0, 0, 0,  ..., 0, 0, 0],
           [0, 0, 0,  ..., 0, 0, 0],
           [0, 0, 0,  ..., 0, 0, 0]]], dtype=torch.uint8),
  'image_id': tensor([0]),
  'area': tensor([35358., 36225.]),
  'iscrowd': tensor([0, 0])})

import torchvision
from torchvision.models.detection.faster_rcnn import FastRCNNPredictor
from torchvision.models.detection.mask_rcnn import MaskRCNNPredictor

      
def get_instance_segmentation_model(num_classes):
    # load an instance segmentation model pre-trained on COCO
    model = torchvision.models.detection.maskrcnn_resnet50_fpn(pretrained=True)

    # get the number of input features for the classifier
    in_features = model.roi_heads.box_predictor.cls_score.in_features
    # replace the pre-trained head with a new one
    model.roi_heads.box_predictor = FastRCNNPredictor(in_features, num_classes)

    # now get the number of input features for the mask classifier
    in_features_mask = model.roi_heads.mask_predictor.conv5_mask.in_channels
    hidden_layer = 256
    # and replace the mask predictor with a new one
    model.roi_heads.mask_predictor = MaskRCNNPredictor(in_features_mask,
                                                       hidden_layer,
                                                       num_classes)

    return model

from engine import train_one_epoch, evaluate
import utils
import transforms as T


def get_transform(train):
    transforms = []
    # converts the image, a PIL image, into a PyTorch Tensor
    transforms.append(T.ToTensor())
    if train:
        # during training, randomly flip the training images
        # and ground-truth for data augmentation
        transforms.append(T.RandomHorizontalFlip(0.5))
    return T.Compose(transforms)

# use our dataset and defined transformations
dataset = PennFudanDataset('PennFudanPed', get_transform(train=True))
dataset_test = PennFudanDataset('PennFudanPed', get_transform(train=False))

# split the dataset in train and test set
torch.manual_seed(1)
indices = torch.randperm(len(dataset)).tolist()
dataset = torch.utils.data.Subset(dataset, indices[:-50])
dataset_test = torch.utils.data.Subset(dataset_test, indices[-50:])

# define training and validation data loaders
data_loader = torch.utils.data.DataLoader(
    dataset, batch_size=2, shuffle=True, num_workers=4,
    collate_fn=utils.collate_fn)

data_loader_test = torch.utils.data.DataLoader(
    dataset_test, batch_size=1, shuffle=False, num_workers=4,
    collate_fn=utils.collate_fn)

device = torch.device('cuda') if torch.cuda.is_available() else torch.device('cpu')

# our dataset has two classes only - background and person
num_classes = 2

# get the model using our helper function
model = get_instance_segmentation_model(num_classes)
# move model to the right device
model.to(device)

# construct an optimizer
params = [p for p in model.parameters() if p.requires_grad]
optimizer = torch.optim.SGD(params, lr=0.005,
                            momentum=0.9, weight_decay=0.0005)

# and a learning rate scheduler which decreases the learning rate by
# 10x every 3 epochs
lr_scheduler = torch.optim.lr_scheduler.StepLR(optimizer,
                                               step_size=3,
                                               gamma=0.1)

Downloading: "https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth" to /home/qy/.cache/torch/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
75.3%IOPub message rate exceeded.
The notebook server will temporarily stop sending output
to the client in order to avoid crashing it.
To change this limit, set the config variable
`--NotebookApp.iopub_msg_rate_limit`.

Current values:
NotebookApp.iopub_msg_rate_limit=1000.0 (msgs/sec)
NotebookApp.rate_limit_window=3.0 (secs)

97.2%IOPub message rate exceeded.
The notebook server will temporarily stop sending output
to the client in order to avoid crashing it.
To change this limit, set the config variable
`--NotebookApp.iopub_msg_rate_limit`.

Current values:
NotebookApp.iopub_msg_rate_limit=1000.0 (msgs/sec)
NotebookApp.rate_limit_window=3.0 (secs)

# let's train it for 10 epochs
num_epochs = 10

for epoch in range(num_epochs):
    # train for one epoch, printing every 10 iterations
    train_one_epoch(model, optimizer, data_loader, device, epoch, print_freq=10)
    # update the learning rate
    lr_scheduler.step()
    # evaluate on the test dataset
    evaluate(model, data_loader_test, device=device)

/home/qy/.conda/envs/pytorch/lib/python3.6/site-packages/torch/nn/functional.py:2854: UserWarning: The default behavior for interpolate/upsample with float scale_factor will change in 1.6.0 to align with other frameworks/libraries, and use scale_factor directly, instead of relying on the computed output size. If you wish to keep the old behavior, please set recompute_scale_factor=True. See the documentation of nn.Upsample for details. 
  warnings.warn("The default behavior for interpolate/upsample with float scale_factor will change "
/opt/conda/conda-bld/pytorch_1587428091666/work/torch/csrc/utils/python_arg_parser.cpp:756: UserWarning: This overload of nonzero is deprecated:
	nonzero(Tensor input, *, Tensor out)
Consider using one of the following signatures instead:
	nonzero(Tensor input, *, bool as_tuple)


Epoch: [0]  [ 0/60]  eta: 0:02:38  lr: 0.000090  loss: 3.5733 (3.5733)  loss_classifier: 0.7385 (0.7385)  loss_box_reg: 0.1523 (0.1523)  loss_mask: 2.6620 (2.6620)  loss_objectness: 0.0130 (0.0130)  loss_rpn_box_reg: 0.0076 (0.0076)  time: 2.6334  data: 0.0967  max mem: 2303
Epoch: [0]  [10/60]  eta: 0:00:47  lr: 0.000936  loss: 1.5685 (2.1198)  loss_classifier: 0.4478 (0.4967)  loss_box_reg: 0.1842 (0.1909)  loss_mask: 0.9258 (1.4014)  loss_objectness: 0.0196 (0.0203)  loss_rpn_box_reg: 0.0090 (0.0105)  time: 0.9557  data: 0.0131  max mem: 2859
Epoch: [0]  [20/60]  eta: 0:00:34  lr: 0.001783  loss: 0.8704 (1.4306)  loss_classifier: 0.2360 (0.3400)  loss_box_reg: 0.1570 (0.1740)  loss_mask: 0.4029 (0.8831)  loss_objectness: 0.0196 (0.0215)  loss_rpn_box_reg: 0.0099 (0.0120)  time: 0.7708  data: 0.0041  max mem: 2862
Epoch: [0]  [30/60]  eta: 0:00:25  lr: 0.002629  loss: 0.5294 (1.1183)  loss_classifier: 0.0927 (0.2557)  loss_box_reg: 0.1266 (0.1603)  loss_mask: 0.2426 (0.6747)  loss_objectness: 0.0060 (0.0160)  loss_rpn_box_reg: 0.0099 (0.0116)  time: 0.7823  data: 0.0036  max mem: 3595
Epoch: [0]  [40/60]  eta: 0:00:16  lr: 0.003476  loss: 0.4095 (0.9466)  loss_classifier: 0.0617 (0.2089)  loss_box_reg: 0.1078 (0.1514)  loss_mask: 0.2057 (0.5606)  loss_objectness: 0.0037 (0.0133)  loss_rpn_box_reg: 0.0116 (0.0124)  time: 0.8173  data: 0.0039  max mem: 3595
Epoch: [0]  [50/60]  eta: 0:00:08  lr: 0.004323  loss: 0.3405 (0.8259)  loss_classifier: 0.0479 (0.1784)  loss_box_reg: 0.0890 (0.1383)  loss_mask: 0.1798 (0.4847)  loss_objectness: 0.0025 (0.0118)  loss_rpn_box_reg: 0.0116 (0.0128)  time: 0.8091  data: 0.0041  max mem: 3595
Epoch: [0]  [59/60]  eta: 0:00:00  lr: 0.005000  loss: 0.2562 (0.7363)  loss_classifier: 0.0415 (0.1567)  loss_box_reg: 0.0532 (0.1233)  loss_mask: 0.1564 (0.4340)  loss_objectness: 0.0010 (0.0101)  loss_rpn_box_reg: 0.0107 (0.0122)  time: 0.7992  data: 0.0039  max mem: 3595
Epoch: [0] Total time: 0:00:49 (0.8252 s / it)
creating index...
index created!
Test:  [ 0/50]  eta: 0:00:10  model_time: 0.1355 (0.1355)  evaluator_time: 0.0021 (0.0021)  time: 0.2005  data: 0.0621  max mem: 3595
Test:  [49/50]  eta: 0:00:00  model_time: 0.1479 (0.1476)  evaluator_time: 0.0023 (0.0042)  time: 0.1541  data: 0.0019  max mem: 3595
Test: Total time: 0:00:07 (0.1560 s / it)
Averaged stats: model_time: 0.1479 (0.1476)  evaluator_time: 0.0023 (0.0042)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.690
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.983
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.871
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.390
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.700
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.312
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.746
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.746
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.738
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.747
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.694
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.983
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.882
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.373
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.708
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.314
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.735
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.736
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.650
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.742
Epoch: [1]  [ 0/60]  eta: 0:00:47  lr: 0.005000  loss: 0.1709 (0.1709)  loss_classifier: 0.0213 (0.0213)  loss_box_reg: 0.0298 (0.0298)  loss_mask: 0.1055 (0.1055)  loss_objectness: 0.0015 (0.0015)  loss_rpn_box_reg: 0.0128 (0.0128)  time: 0.7915  data: 0.0902  max mem: 3595
Epoch: [1]  [10/60]  eta: 0:00:40  lr: 0.005000  loss: 0.2111 (0.2462)  loss_classifier: 0.0332 (0.0370)  loss_box_reg: 0.0318 (0.0406)  loss_mask: 0.1447 (0.1562)  loss_objectness: 0.0009 (0.0011)  loss_rpn_box_reg: 0.0124 (0.0113)  time: 0.8176  data: 0.0112  max mem: 3595
Epoch: [1]  [20/60]  eta: 0:00:32  lr: 0.005000  loss: 0.2571 (0.2645)  loss_classifier: 0.0402 (0.0460)  loss_box_reg: 0.0350 (0.0439)  loss_mask: 0.1626 (0.1598)  loss_objectness: 0.0009 (0.0016)  loss_rpn_box_reg: 0.0129 (0.0132)  time: 0.8183  data: 0.0037  max mem: 3595
Epoch: [1]  [30/60]  eta: 0:00:24  lr: 0.005000  loss: 0.2145 (0.2412)  loss_classifier: 0.0367 (0.0404)  loss_box_reg: 0.0284 (0.0365)  loss_mask: 0.1468 (0.1511)  loss_objectness: 0.0009 (0.0013)  loss_rpn_box_reg: 0.0086 (0.0118)  time: 0.8109  data: 0.0042  max mem: 3595
Epoch: [1]  [40/60]  eta: 0:00:16  lr: 0.005000  loss: 0.2021 (0.2345)  loss_classifier: 0.0317 (0.0404)  loss_box_reg: 0.0219 (0.0337)  loss_mask: 0.1344 (0.1476)  loss_objectness: 0.0006 (0.0016)  loss_rpn_box_reg: 0.0079 (0.0112)  time: 0.8123  data: 0.0041  max mem: 3595
Epoch: [1]  [50/60]  eta: 0:00:08  lr: 0.005000  loss: 0.2033 (0.2319)  loss_classifier: 0.0358 (0.0412)  loss_box_reg: 0.0232 (0.0330)  loss_mask: 0.1293 (0.1442)  loss_objectness: 0.0006 (0.0018)  loss_rpn_box_reg: 0.0080 (0.0118)  time: 0.8298  data: 0.0039  max mem: 3595
Epoch: [1]  [59/60]  eta: 0:00:00  lr: 0.005000  loss: 0.1682 (0.2241)  loss_classifier: 0.0298 (0.0400)  loss_box_reg: 0.0174 (0.0306)  loss_mask: 0.1251 (0.1404)  loss_objectness: 0.0004 (0.0018)  loss_rpn_box_reg: 0.0080 (0.0112)  time: 0.8060  data: 0.0038  max mem: 3595
Epoch: [1] Total time: 0:00:48 (0.8142 s / it)
creating index...
index created!
Test:  [ 0/50]  eta: 0:00:10  model_time: 0.1453 (0.1453)  evaluator_time: 0.0022 (0.0022)  time: 0.2145  data: 0.0661  max mem: 3595
Test:  [49/50]  eta: 0:00:00  model_time: 0.1585 (0.1521)  evaluator_time: 0.0025 (0.0036)  time: 0.1635  data: 0.0020  max mem: 3595
Test: Total time: 0:00:08 (0.1602 s / it)
Averaged stats: model_time: 0.1585 (0.1521)  evaluator_time: 0.0025 (0.0036)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.771
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.988
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.937
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.524
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.781
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.358
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.820
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.820
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.738
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.826
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.763
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.988
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.916
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.389
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.773
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.346
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.797
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.797
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.713
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.803
Epoch: [2]  [ 0/60]  eta: 0:00:53  lr: 0.005000  loss: 0.1299 (0.1299)  loss_classifier: 0.0173 (0.0173)  loss_box_reg: 0.0072 (0.0072)  loss_mask: 0.0994 (0.0994)  loss_objectness: 0.0007 (0.0007)  loss_rpn_box_reg: 0.0053 (0.0053)  time: 0.8846  data: 0.0853  max mem: 3595
Epoch: [2]  [10/60]  eta: 0:00:40  lr: 0.005000  loss: 0.1884 (0.1756)  loss_classifier: 0.0315 (0.0306)  loss_box_reg: 0.0154 (0.0166)  loss_mask: 0.1135 (0.1197)  loss_objectness: 0.0004 (0.0006)  loss_rpn_box_reg: 0.0066 (0.0081)  time: 0.8066  data: 0.0110  max mem: 3595
Epoch: [2]  [20/60]  eta: 0:00:31  lr: 0.005000  loss: 0.1510 (0.1623)  loss_classifier: 0.0189 (0.0246)  loss_box_reg: 0.0119 (0.0133)  loss_mask: 0.1107 (0.1166)  loss_objectness: 0.0002 (0.0007)  loss_rpn_box_reg: 0.0048 (0.0071)  time: 0.7697  data: 0.0037  max mem: 3595
Epoch: [2]  [30/60]  eta: 0:00:23  lr: 0.005000  loss: 0.1591 (0.1821)  loss_classifier: 0.0207 (0.0295)  loss_box_reg: 0.0104 (0.0172)  loss_mask: 0.1164 (0.1260)  loss_objectness: 0.0003 (0.0007)  loss_rpn_box_reg: 0.0074 (0.0086)  time: 0.7821  data: 0.0038  max mem: 3595
Epoch: [2]  [40/60]  eta: 0:00:15  lr: 0.005000  loss: 0.1827 (0.1824)  loss_classifier: 0.0289 (0.0286)  loss_box_reg: 0.0144 (0.0167)  loss_mask: 0.1318 (0.1276)  loss_objectness: 0.0003 (0.0008)  loss_rpn_box_reg: 0.0077 (0.0087)  time: 0.8254  data: 0.0041  max mem: 3595
Epoch: [2]  [50/60]  eta: 0:00:08  lr: 0.005000  loss: 0.1878 (0.1862)  loss_classifier: 0.0298 (0.0290)  loss_box_reg: 0.0154 (0.0176)  loss_mask: 0.1324 (0.1297)  loss_objectness: 0.0003 (0.0009)  loss_rpn_box_reg: 0.0083 (0.0091)  time: 0.8146  data: 0.0039  max mem: 3595
Epoch: [2]  [59/60]  eta: 0:00:00  lr: 0.005000  loss: 0.1878 (0.1876)  loss_classifier: 0.0323 (0.0300)  loss_box_reg: 0.0160 (0.0176)  loss_mask: 0.1293 (0.1294)  loss_objectness: 0.0005 (0.0010)  loss_rpn_box_reg: 0.0094 (0.0095)  time: 0.8387  data: 0.0037  max mem: 3595
Epoch: [2] Total time: 0:00:48 (0.8130 s / it)
creating index...
index created!
Test:  [ 0/50]  eta: 0:00:09  model_time: 0.1336 (0.1336)  evaluator_time: 0.0022 (0.0022)  time: 0.1977  data: 0.0612  max mem: 3595
Test:  [49/50]  eta: 0:00:00  model_time: 0.1465 (0.1468)  evaluator_time: 0.0020 (0.0031)  time: 0.1518  data: 0.0019  max mem: 3595
Test: Total time: 0:00:07 (0.1542 s / it)
Averaged stats: model_time: 0.1465 (0.1468)  evaluator_time: 0.0020 (0.0031)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.803
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.989
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.946
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.470
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.813
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.371
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.841
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.841
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.762
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.847
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.745
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.989
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.925
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.401
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.762
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.345
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.783
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.783
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.613
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.795
Epoch: [3]  [ 0/60]  eta: 0:00:57  lr: 0.000500  loss: 0.1744 (0.1744)  loss_classifier: 0.0190 (0.0190)  loss_box_reg: 0.0126 (0.0126)  loss_mask: 0.1370 (0.1370)  loss_objectness: 0.0001 (0.0001)  loss_rpn_box_reg: 0.0057 (0.0057)  time: 0.9517  data: 0.1323  max mem: 3595
Epoch: [3]  [10/60]  eta: 0:00:43  lr: 0.000500  loss: 0.1701 (0.1784)  loss_classifier: 0.0271 (0.0279)  loss_box_reg: 0.0121 (0.0151)  loss_mask: 0.1171 (0.1272)  loss_objectness: 0.0006 (0.0007)  loss_rpn_box_reg: 0.0060 (0.0076)  time: 0.8605  data: 0.0146  max mem: 3595
Epoch: [3]  [20/60]  eta: 0:00:33  lr: 0.000500  loss: 0.1639 (0.1734)  loss_classifier: 0.0271 (0.0275)  loss_box_reg: 0.0099 (0.0138)  loss_mask: 0.1199 (0.1235)  loss_objectness: 0.0003 (0.0007)  loss_rpn_box_reg: 0.0060 (0.0079)  time: 0.8288  data: 0.0034  max mem: 3595
Epoch: [3]  [30/60]  eta: 0:00:24  lr: 0.000500  loss: 0.1571 (0.1701)  loss_classifier: 0.0241 (0.0262)  loss_box_reg: 0.0085 (0.0129)  loss_mask: 0.1203 (0.1229)  loss_objectness: 0.0004 (0.0008)  loss_rpn_box_reg: 0.0055 (0.0073)  time: 0.7733  data: 0.0038  max mem: 3595
Epoch: [3]  [40/60]  eta: 0:00:16  lr: 0.000500  loss: 0.1577 (0.1724)  loss_classifier: 0.0245 (0.0269)  loss_box_reg: 0.0108 (0.0133)  loss_mask: 0.1130 (0.1232)  loss_objectness: 0.0007 (0.0009)  loss_rpn_box_reg: 0.0065 (0.0082)  time: 0.7781  data: 0.0038  max mem: 3595
Epoch: [3]  [50/60]  eta: 0:00:08  lr: 0.000500  loss: 0.1741 (0.1774)  loss_classifier: 0.0278 (0.0275)  loss_box_reg: 0.0116 (0.0143)  loss_mask: 0.1303 (0.1265)  loss_objectness: 0.0004 (0.0009)  loss_rpn_box_reg: 0.0078 (0.0082)  time: 0.8271  data: 0.0040  max mem: 3595
Epoch: [3]  [59/60]  eta: 0:00:00  lr: 0.000500  loss: 0.1501 (0.1721)  loss_classifier: 0.0269 (0.0266)  loss_box_reg: 0.0093 (0.0133)  loss_mask: 0.1049 (0.1231)  loss_objectness: 0.0003 (0.0009)  loss_rpn_box_reg: 0.0078 (0.0081)  time: 0.8225  data: 0.0041  max mem: 3595
Epoch: [3] Total time: 0:00:48 (0.8145 s / it)
creating index...
index created!
Test:  [ 0/50]  eta: 0:00:09  model_time: 0.1334 (0.1334)  evaluator_time: 0.0021 (0.0021)  time: 0.1983  data: 0.0620  max mem: 3595
Test:  [49/50]  eta: 0:00:00  model_time: 0.1473 (0.1432)  evaluator_time: 0.0019 (0.0030)  time: 0.1507  data: 0.0019  max mem: 3595
Test: Total time: 0:00:07 (0.1504 s / it)
Averaged stats: model_time: 0.1473 (0.1432)  evaluator_time: 0.0019 (0.0030)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.817
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.990
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.941
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.476
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.829
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.378
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.861
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.861
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.762
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.868
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.765
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.990
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.919
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.448
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.777
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.350
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.803
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.803
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.700
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.811
Epoch: [4]  [ 0/60]  eta: 0:00:42  lr: 0.000500  loss: 0.1048 (0.1048)  loss_classifier: 0.0069 (0.0069)  loss_box_reg: 0.0029 (0.0029)  loss_mask: 0.0903 (0.0903)  loss_objectness: 0.0001 (0.0001)  loss_rpn_box_reg: 0.0046 (0.0046)  time: 0.7136  data: 0.0856  max mem: 3595
Epoch: [4]  [10/60]  eta: 0:00:40  lr: 0.000500  loss: 0.1483 (0.1592)  loss_classifier: 0.0231 (0.0214)  loss_box_reg: 0.0106 (0.0124)  loss_mask: 0.1064 (0.1175)  loss_objectness: 0.0002 (0.0004)  loss_rpn_box_reg: 0.0076 (0.0075)  time: 0.8011  data: 0.0111  max mem: 3595
Epoch: [4]  [20/60]  eta: 0:00:31  lr: 0.000500  loss: 0.1483 (0.1608)  loss_classifier: 0.0231 (0.0229)  loss_box_reg: 0.0079 (0.0115)  loss_mask: 0.1064 (0.1193)  loss_objectness: 0.0002 (0.0004)  loss_rpn_box_reg: 0.0061 (0.0067)  time: 0.7832  data: 0.0039  max mem: 3595
Epoch: [4]  [30/60]  eta: 0:00:23  lr: 0.000500  loss: 0.1397 (0.1650)  loss_classifier: 0.0261 (0.0260)  loss_box_reg: 0.0078 (0.0122)  loss_mask: 0.1033 (0.1185)  loss_objectness: 0.0003 (0.0008)  loss_rpn_box_reg: 0.0060 (0.0075)  time: 0.7800  data: 0.0043  max mem: 3595
Epoch: [4]  [40/60]  eta: 0:00:16  lr: 0.000500  loss: 0.1451 (0.1627)  loss_classifier: 0.0235 (0.0257)  loss_box_reg: 0.0093 (0.0119)  loss_mask: 0.1079 (0.1170)  loss_objectness: 0.0003 (0.0009)  loss_rpn_box_reg: 0.0062 (0.0072)  time: 0.8241  data: 0.0042  max mem: 3595
Epoch: [4]  [50/60]  eta: 0:00:08  lr: 0.000500  loss: 0.1587 (0.1632)  loss_classifier: 0.0239 (0.0259)  loss_box_reg: 0.0093 (0.0118)  loss_mask: 0.1092 (0.1169)  loss_objectness: 0.0003 (0.0008)  loss_rpn_box_reg: 0.0068 (0.0078)  time: 0.8400  data: 0.0041  max mem: 3595
Epoch: [4]  [59/60]  eta: 0:00:00  lr: 0.000500  loss: 0.1587 (0.1661)  loss_classifier: 0.0276 (0.0262)  loss_box_reg: 0.0079 (0.0126)  loss_mask: 0.1106 (0.1184)  loss_objectness: 0.0003 (0.0010)  loss_rpn_box_reg: 0.0080 (0.0080)  time: 0.8359  data: 0.0044  max mem: 3595
Epoch: [4] Total time: 0:00:48 (0.8121 s / it)
creating index...
index created!
Test:  [ 0/50]  eta: 0:00:10  model_time: 0.1360 (0.1360)  evaluator_time: 0.0019 (0.0019)  time: 0.2020  data: 0.0633  max mem: 3595
Test:  [49/50]  eta: 0:00:00  model_time: 0.1543 (0.1492)  evaluator_time: 0.0020 (0.0031)  time: 0.1577  data: 0.0019  max mem: 3595
Test: Total time: 0:00:07 (0.1568 s / it)
Averaged stats: model_time: 0.1543 (0.1492)  evaluator_time: 0.0020 (0.0031)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.820
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.990
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.939
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.532
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.833
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.374
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.866
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.866
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.750
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.875
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.759
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.990
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.918
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.406
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.771
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.346
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.803
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.803
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.700
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.810
Epoch: [5]  [ 0/60]  eta: 0:01:09  lr: 0.000500  loss: 0.1493 (0.1493)  loss_classifier: 0.0173 (0.0173)  loss_box_reg: 0.0064 (0.0064)  loss_mask: 0.1195 (0.1195)  loss_objectness: 0.0003 (0.0003)  loss_rpn_box_reg: 0.0058 (0.0058)  time: 1.1576  data: 0.1904  max mem: 3595
Epoch: [5]  [10/60]  eta: 0:00:39  lr: 0.000500  loss: 0.1460 (0.1500)  loss_classifier: 0.0202 (0.0213)  loss_box_reg: 0.0084 (0.0094)  loss_mask: 0.1132 (0.1126)  loss_objectness: 0.0002 (0.0006)  loss_rpn_box_reg: 0.0055 (0.0061)  time: 0.7865  data: 0.0191  max mem: 3595
Epoch: [5]  [20/60]  eta: 0:00:30  lr: 0.000500  loss: 0.1351 (0.1456)  loss_classifier: 0.0201 (0.0205)  loss_box_reg: 0.0064 (0.0088)  loss_mask: 0.1034 (0.1097)  loss_objectness: 0.0002 (0.0005)  loss_rpn_box_reg: 0.0054 (0.0062)  time: 0.7460  data: 0.0027  max mem: 3595
Epoch: [5]  [30/60]  eta: 0:00:23  lr: 0.000500  loss: 0.1419 (0.1600)  loss_classifier: 0.0244 (0.0244)  loss_box_reg: 0.0083 (0.0117)  loss_mask: 0.1038 (0.1158)  loss_objectness: 0.0003 (0.0007)  loss_rpn_box_reg: 0.0071 (0.0075)  time: 0.7701  data: 0.0037  max mem: 3595
Epoch: [5]  [40/60]  eta: 0:00:15  lr: 0.000500  loss: 0.1620 (0.1620)  loss_classifier: 0.0268 (0.0247)  loss_box_reg: 0.0113 (0.0120)  loss_mask: 0.1163 (0.1168)  loss_objectness: 0.0004 (0.0007)  loss_rpn_box_reg: 0.0085 (0.0078)  time: 0.8153  data: 0.0039  max mem: 3595
Epoch: [5]  [50/60]  eta: 0:00:07  lr: 0.000500  loss: 0.1441 (0.1581)  loss_classifier: 0.0229 (0.0242)  loss_box_reg: 0.0081 (0.0110)  loss_mask: 0.1041 (0.1148)  loss_objectness: 0.0003 (0.0007)  loss_rpn_box_reg: 0.0067 (0.0075)  time: 0.8283  data: 0.0042  max mem: 3595
Epoch: [5]  [59/60]  eta: 0:00:00  lr: 0.000500  loss: 0.1449 (0.1596)  loss_classifier: 0.0217 (0.0249)  loss_box_reg: 0.0070 (0.0111)  loss_mask: 0.1044 (0.1154)  loss_objectness: 0.0002 (0.0007)  loss_rpn_box_reg: 0.0070 (0.0076)  time: 0.8396  data: 0.0043  max mem: 3595
Epoch: [5] Total time: 0:00:48 (0.8076 s / it)
creating index...
index created!
Test:  [ 0/50]  eta: 0:00:10  model_time: 0.1335 (0.1335)  evaluator_time: 0.0023 (0.0023)  time: 0.2000  data: 0.0635  max mem: 3595
Test:  [49/50]  eta: 0:00:00  model_time: 0.1466 (0.1485)  evaluator_time: 0.0020 (0.0031)  time: 0.1563  data: 0.0019  max mem: 3595
Test: Total time: 0:00:07 (0.1559 s / it)
Averaged stats: model_time: 0.1466 (0.1485)  evaluator_time: 0.0020 (0.0031)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.815
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.990
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.940
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.542
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.825
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.376
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.858
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.858
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.775
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.864
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.765
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.990
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.932
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.428
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.776
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.350
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.808
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.808
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.738
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.813
Epoch: [6]  [ 0/60]  eta: 0:00:56  lr: 0.000050  loss: 0.1720 (0.1720)  loss_classifier: 0.0306 (0.0306)  loss_box_reg: 0.0119 (0.0119)  loss_mask: 0.1225 (0.1225)  loss_objectness: 0.0001 (0.0001)  loss_rpn_box_reg: 0.0069 (0.0069)  time: 0.9433  data: 0.0828  max mem: 3595
Epoch: [6]  [10/60]  eta: 0:00:41  lr: 0.000050  loss: 0.1458 (0.1465)  loss_classifier: 0.0210 (0.0226)  loss_box_reg: 0.0059 (0.0078)  loss_mask: 0.1034 (0.1104)  loss_objectness: 0.0002 (0.0004)  loss_rpn_box_reg: 0.0059 (0.0053)  time: 0.8372  data: 0.0108  max mem: 3595
Epoch: [6]  [20/60]  eta: 0:00:33  lr: 0.000050  loss: 0.1448 (0.1537)  loss_classifier: 0.0210 (0.0245)  loss_box_reg: 0.0071 (0.0105)  loss_mask: 0.1019 (0.1116)  loss_objectness: 0.0002 (0.0005)  loss_rpn_box_reg: 0.0059 (0.0066)  time: 0.8248  data: 0.0037  max mem: 3595
Epoch: [6]  [30/60]  eta: 0:00:24  lr: 0.000050  loss: 0.1481 (0.1547)  loss_classifier: 0.0230 (0.0261)  loss_box_reg: 0.0088 (0.0102)  loss_mask: 0.1019 (0.1107)  loss_objectness: 0.0003 (0.0009)  loss_rpn_box_reg: 0.0074 (0.0068)  time: 0.8078  data: 0.0037  max mem: 3595
Epoch: [6]  [40/60]  eta: 0:00:16  lr: 0.000050  loss: 0.1484 (0.1617)  loss_classifier: 0.0260 (0.0269)  loss_box_reg: 0.0094 (0.0116)  loss_mask: 0.1055 (0.1148)  loss_objectness: 0.0003 (0.0011)  loss_rpn_box_reg: 0.0071 (0.0072)  time: 0.7847  data: 0.0035  max mem: 3595
Epoch: [6]  [50/60]  eta: 0:00:08  lr: 0.000050  loss: 0.1445 (0.1607)  loss_classifier: 0.0221 (0.0264)  loss_box_reg: 0.0104 (0.0114)  loss_mask: 0.1101 (0.1146)  loss_objectness: 0.0004 (0.0010)  loss_rpn_box_reg: 0.0071 (0.0073)  time: 0.7750  data: 0.0036  max mem: 3595
Epoch: [6]  [59/60]  eta: 0:00:00  lr: 0.000050  loss: 0.1445 (0.1610)  loss_classifier: 0.0230 (0.0261)  loss_box_reg: 0.0077 (0.0114)  loss_mask: 0.1101 (0.1151)  loss_objectness: 0.0004 (0.0011)  loss_rpn_box_reg: 0.0071 (0.0073)  time: 0.7916  data: 0.0036  max mem: 3595
Epoch: [6] Total time: 0:00:48 (0.8018 s / it)
creating index...
index created!
Test:  [ 0/50]  eta: 0:00:09  model_time: 0.1334 (0.1334)  evaluator_time: 0.0020 (0.0020)  time: 0.1978  data: 0.0616  max mem: 3595
Test:  [49/50]  eta: 0:00:00  model_time: 0.1474 (0.1430)  evaluator_time: 0.0020 (0.0029)  time: 0.1501  data: 0.0019  max mem: 3595
Test: Total time: 0:00:07 (0.1502 s / it)
Averaged stats: model_time: 0.1474 (0.1430)  evaluator_time: 0.0020 (0.0029)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.820
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.990
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.940
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.553
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.830
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.377
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.862
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.862
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.787
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.868
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.764
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.990
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.932
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.418
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.775
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.350
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.806
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.806
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.738
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.811
Epoch: [7]  [ 0/60]  eta: 0:00:49  lr: 0.000050  loss: 0.1097 (0.1097)  loss_classifier: 0.0113 (0.0113)  loss_box_reg: 0.0037 (0.0037)  loss_mask: 0.0925 (0.0925)  loss_objectness: 0.0011 (0.0011)  loss_rpn_box_reg: 0.0011 (0.0011)  time: 0.8293  data: 0.1411  max mem: 3595
Epoch: [7]  [10/60]  eta: 0:00:38  lr: 0.000050  loss: 0.1355 (0.1430)  loss_classifier: 0.0191 (0.0210)  loss_box_reg: 0.0049 (0.0072)  loss_mask: 0.1026 (0.1089)  loss_objectness: 0.0003 (0.0004)  loss_rpn_box_reg: 0.0040 (0.0056)  time: 0.7626  data: 0.0151  max mem: 3595
Epoch: [7]  [20/60]  eta: 0:00:32  lr: 0.000050  loss: 0.1482 (0.1526)  loss_classifier: 0.0247 (0.0256)  loss_box_reg: 0.0073 (0.0095)  loss_mask: 0.1060 (0.1100)  loss_objectness: 0.0002 (0.0005)  loss_rpn_box_reg: 0.0062 (0.0070)  time: 0.8048  data: 0.0031  max mem: 3595
Epoch: [7]  [30/60]  eta: 0:00:23  lr: 0.000050  loss: 0.1438 (0.1497)  loss_classifier: 0.0247 (0.0246)  loss_box_reg: 0.0073 (0.0090)  loss_mask: 0.1056 (0.1087)  loss_objectness: 0.0002 (0.0005)  loss_rpn_box_reg: 0.0069 (0.0069)  time: 0.8128  data: 0.0036  max mem: 3595
Epoch: [7]  [40/60]  eta: 0:00:15  lr: 0.000050  loss: 0.1410 (0.1530)  loss_classifier: 0.0207 (0.0243)  loss_box_reg: 0.0074 (0.0099)  loss_mask: 0.1070 (0.1114)  loss_objectness: 0.0003 (0.0006)  loss_rpn_box_reg: 0.0053 (0.0069)  time: 0.7771  data: 0.0036  max mem: 3595
Epoch: [7]  [50/60]  eta: 0:00:07  lr: 0.000050  loss: 0.1581 (0.1577)  loss_classifier: 0.0201 (0.0241)  loss_box_reg: 0.0089 (0.0108)  loss_mask: 0.1107 (0.1150)  loss_objectness: 0.0007 (0.0007)  loss_rpn_box_reg: 0.0072 (0.0072)  time: 0.7750  data: 0.0035  max mem: 3595
Epoch: [7]  [59/60]  eta: 0:00:00  lr: 0.000050  loss: 0.1485 (0.1578)  loss_classifier: 0.0201 (0.0243)  loss_box_reg: 0.0061 (0.0108)  loss_mask: 0.1099 (0.1149)  loss_objectness: 0.0008 (0.0007)  loss_rpn_box_reg: 0.0059 (0.0072)  time: 0.7780  data: 0.0035  max mem: 3595
Epoch: [7] Total time: 0:00:47 (0.7868 s / it)
creating index...
index created!
Test:  [ 0/50]  eta: 0:00:10  model_time: 0.1344 (0.1344)  evaluator_time: 0.0022 (0.0022)  time: 0.2001  data: 0.0626  max mem: 3595
Test:  [49/50]  eta: 0:00:00  model_time: 0.1460 (0.1428)  evaluator_time: 0.0019 (0.0030)  time: 0.1500  data: 0.0019  max mem: 3595
Test: Total time: 0:00:07 (0.1502 s / it)
Averaged stats: model_time: 0.1460 (0.1428)  evaluator_time: 0.0019 (0.0030)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.820
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.990
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.940
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.553
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.830
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.377
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.861
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.861
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.787
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.867
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.765
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.990
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.932
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.430
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.776
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.352
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.808
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.808
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.738
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.813
Epoch: [8]  [ 0/60]  eta: 0:00:54  lr: 0.000050  loss: 0.1661 (0.1661)  loss_classifier: 0.0293 (0.0293)  loss_box_reg: 0.0071 (0.0071)  loss_mask: 0.1191 (0.1191)  loss_objectness: 0.0076 (0.0076)  loss_rpn_box_reg: 0.0029 (0.0029)  time: 0.9051  data: 0.1844  max mem: 3595
Epoch: [8]  [10/60]  eta: 0:00:39  lr: 0.000050  loss: 0.1512 (0.1534)  loss_classifier: 0.0226 (0.0231)  loss_box_reg: 0.0071 (0.0108)  loss_mask: 0.1086 (0.1121)  loss_objectness: 0.0002 (0.0011)  loss_rpn_box_reg: 0.0044 (0.0063)  time: 0.7955  data: 0.0189  max mem: 3595
Epoch: [8]  [20/60]  eta: 0:00:31  lr: 0.000050  loss: 0.1517 (0.1618)  loss_classifier: 0.0225 (0.0246)  loss_box_reg: 0.0097 (0.0118)  loss_mask: 0.1076 (0.1172)  loss_objectness: 0.0002 (0.0009)  loss_rpn_box_reg: 0.0068 (0.0073)  time: 0.7798  data: 0.0029  max mem: 3595
Epoch: [8]  [30/60]  eta: 0:00:23  lr: 0.000050  loss: 0.1379 (0.1574)  loss_classifier: 0.0225 (0.0241)  loss_box_reg: 0.0091 (0.0121)  loss_mask: 0.1016 (0.1134)  loss_objectness: 0.0002 (0.0006)  loss_rpn_box_reg: 0.0074 (0.0072)  time: 0.7700  data: 0.0036  max mem: 3595
Epoch: [8]  [40/60]  eta: 0:00:15  lr: 0.000050  loss: 0.1505 (0.1647)  loss_classifier: 0.0240 (0.0261)  loss_box_reg: 0.0099 (0.0125)  loss_mask: 0.1127 (0.1177)  loss_objectness: 0.0003 (0.0008)  loss_rpn_box_reg: 0.0073 (0.0076)  time: 0.7909  data: 0.0037  max mem: 3595
Epoch: [8]  [50/60]  eta: 0:00:07  lr: 0.000050  loss: 0.1566 (0.1640)  loss_classifier: 0.0252 (0.0266)  loss_box_reg: 0.0097 (0.0122)  loss_mask: 0.1176 (0.1162)  loss_objectness: 0.0004 (0.0009)  loss_rpn_box_reg: 0.0076 (0.0081)  time: 0.8186  data: 0.0037  max mem: 3595
Epoch: [8]  [59/60]  eta: 0:00:00  lr: 0.000050  loss: 0.1460 (0.1603)  loss_classifier: 0.0229 (0.0254)  loss_box_reg: 0.0062 (0.0113)  loss_mask: 0.1032 (0.1153)  loss_objectness: 0.0003 (0.0009)  loss_rpn_box_reg: 0.0053 (0.0074)  time: 0.7792  data: 0.0036  max mem: 3595
Epoch: [8] Total time: 0:00:47 (0.7848 s / it)
creating index...
index created!
Test:  [ 0/50]  eta: 0:00:09  model_time: 0.1338 (0.1338)  evaluator_time: 0.0022 (0.0022)  time: 0.1986  data: 0.0617  max mem: 3595
Test:  [49/50]  eta: 0:00:00  model_time: 0.1465 (0.1430)  evaluator_time: 0.0020 (0.0029)  time: 0.1504  data: 0.0020  max mem: 3595
Test: Total time: 0:00:07 (0.1502 s / it)
Averaged stats: model_time: 0.1465 (0.1430)  evaluator_time: 0.0020 (0.0029)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.820
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.990
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.939
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.553
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.830
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.378
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.863
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.863
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.787
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.868
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.765
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.990
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.932
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.430
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.775
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.350
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.808
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.808
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.738
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.813
Epoch: [9]  [ 0/60]  eta: 0:00:48  lr: 0.000005  loss: 0.1442 (0.1442)  loss_classifier: 0.0140 (0.0140)  loss_box_reg: 0.0034 (0.0034)  loss_mask: 0.1234 (0.1234)  loss_objectness: 0.0002 (0.0002)  loss_rpn_box_reg: 0.0032 (0.0032)  time: 0.8146  data: 0.0761  max mem: 3595
Epoch: [9]  [10/60]  eta: 0:00:39  lr: 0.000005  loss: 0.1442 (0.1671)  loss_classifier: 0.0200 (0.0212)  loss_box_reg: 0.0095 (0.0119)  loss_mask: 0.1228 (0.1254)  loss_objectness: 0.0003 (0.0022)  loss_rpn_box_reg: 0.0055 (0.0064)  time: 0.7829  data: 0.0101  max mem: 3595
Epoch: [9]  [20/60]  eta: 0:00:31  lr: 0.000005  loss: 0.1593 (0.1659)  loss_classifier: 0.0248 (0.0246)  loss_box_reg: 0.0095 (0.0121)  loss_mask: 0.1177 (0.1209)  loss_objectness: 0.0003 (0.0016)  loss_rpn_box_reg: 0.0055 (0.0066)  time: 0.7862  data: 0.0036  max mem: 3595
Epoch: [9]  [30/60]  eta: 0:00:23  lr: 0.000005  loss: 0.1593 (0.1684)  loss_classifier: 0.0265 (0.0264)  loss_box_reg: 0.0088 (0.0128)  loss_mask: 0.1139 (0.1202)  loss_objectness: 0.0004 (0.0016)  loss_rpn_box_reg: 0.0063 (0.0074)  time: 0.7912  data: 0.0036  max mem: 3595
Epoch: [9]  [40/60]  eta: 0:00:15  lr: 0.000005  loss: 0.1467 (0.1608)  loss_classifier: 0.0233 (0.0248)  loss_box_reg: 0.0077 (0.0115)  loss_mask: 0.1032 (0.1163)  loss_objectness: 0.0004 (0.0013)  loss_rpn_box_reg: 0.0064 (0.0070)  time: 0.7813  data: 0.0036  max mem: 3595
Epoch: [9]  [50/60]  eta: 0:00:07  lr: 0.000005  loss: 0.1487 (0.1593)  loss_classifier: 0.0215 (0.0247)  loss_box_reg: 0.0080 (0.0111)  loss_mask: 0.1077 (0.1152)  loss_objectness: 0.0002 (0.0011)  loss_rpn_box_reg: 0.0060 (0.0072)  time: 0.8089  data: 0.0036  max mem: 3595
Epoch: [9]  [59/60]  eta: 0:00:00  lr: 0.000005  loss: 0.1528 (0.1603)  loss_classifier: 0.0215 (0.0251)  loss_box_reg: 0.0077 (0.0108)  loss_mask: 0.1093 (0.1161)  loss_objectness: 0.0002 (0.0010)  loss_rpn_box_reg: 0.0060 (0.0072)  time: 0.8113  data: 0.0036  max mem: 3595
Epoch: [9] Total time: 0:00:47 (0.7922 s / it)
creating index...
index created!
Test:  [ 0/50]  eta: 0:00:09  model_time: 0.1336 (0.1336)  evaluator_time: 0.0021 (0.0021)  time: 0.1983  data: 0.0618  max mem: 3595
Test:  [49/50]  eta: 0:00:00  model_time: 0.1473 (0.1434)  evaluator_time: 0.0020 (0.0029)  time: 0.1519  data: 0.0019  max mem: 3595
Test: Total time: 0:00:07 (0.1505 s / it)
Averaged stats: model_time: 0.1473 (0.1434)  evaluator_time: 0.0020 (0.0029)
Accumulating evaluation results...
DONE (t=0.01s).
Accumulating evaluation results...
DONE (t=0.01s).
IoU metric: bbox
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.821
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.990
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.939
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.553
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.831
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.379
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.864
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.864
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.787
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.869
IoU metric: segm
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.763
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.990
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.932
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.430
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.773
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.350
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.807
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.807
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.738
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.812

# pick one image from the test set
img, _ = dataset_test[0]
# put the model in evaluation mode
model.eval()
with torch.no_grad():
    prediction = model([img.to(device)])

prediction

[{'boxes': tensor([[ 59.3467,  44.0603, 195.7051, 326.9784],
          [276.5097,  22.4848, 290.8961,  73.5858],
          [ 80.4872,  38.2383, 190.3802, 218.9982]], device='cuda:0'),
  'labels': tensor([1, 1, 1], device='cuda:0'),
  'scores': tensor([0.9994, 0.8514, 0.1090], device='cuda:0'),
  'masks': tensor([[[[0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            ...,
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.]]],
  
  
          [[[0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            ...,
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.]]],
  
  
          [[[0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            ...,
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.],
            [0., 0., 0.,  ..., 0., 0., 0.]]]], device='cuda:0')}]

Image.fromarray(img.mul(255).permute(1, 2, 0).byte().numpy())

Image.fromarray(prediction[0]['masks'][0, 0].mul(255).byte().cpu().numpy())

你可能感兴趣的:(pytorch)

Spring AI与机器学习：智能应用开发新范式 tmjpz04412 人工智能 spring 机器学习
SpringAI与机器学习的整合SpringAI是一个基于Spring生态的AI开发框架，旨在简化智能应用的开发流程。通过SpringAI，开发者可以快速集成机器学习模型，构建高效的智能应用。SpringAI支持多种机器学习库和框架，如TensorFlow、PyTorch和Scikit-learn，提供统一的API接口。SpringAI的核心优势在于其模块化设计和自动化配置。开发者无需关心复杂的依
从零开始构建深度学习环境：基于Pytorch、CUDA与cuDNN的虚拟环境搭建与实践（适合初学者）荣华富贵8 程序员的知识储备2 程序员的知识储备3 深度学习 pytorch 人工智能
摘要：深度学习正在引领人工智能技术的革新，而对于初学者来说，正确搭建深度学习环境是迈向AI研究与应用的第一步。本文将为读者提供一套详尽的教程，指导如何在本地环境中搭建Pytorch、CUDA与cuDNN，以及如何利用Anaconda和PyCharm进行高效开发。内容涵盖从环境配置、常见错误修正，到基础的深度学习模型构建及训练。我们旨在为深度学习零基础的入门者提供一个全面且易于理解的“保姆级”教程，
使用 PyTorch 和 Pandas 进行 Kaggle 房价预测 Clang's Blog AI pytorch pandas 人工智能
文章目录1、环境设置2、数据下载3、数据预处理4、模型构建5、训练和验证6、训练模型并生成预测结果7、完整代码在本篇博文中，我们将探索如何使用PyTorch和Pandas库，构建一个用于Kaggle房价预测的模型。我们将详细讨论数据加载、预处理、模型构建、训练、验证及最终预测的全过程。1、环境设置我们首先需要导入所需的库，包括用于数据处理的pandas和numpy，以及用于深度学习的torch。i
PyTorch 使用指南
PyTorch是一个功能强大且灵活的Python开源机器学习库，以其动态计算图和直观的Pythonic接口而闻名。本指南将带您了解PyTorch的基础操作，包括张量创建、自动求导，以及如何构建、训练和优化神经网络模型。我们还将深入探讨其在图像分类（以CIFAR-10为例）和自然语言处理（以灾难推文分类为例）等特定领域的应用，并概述其在图像分割和强化学习等其他领域的应用。PyTorch使用指南1.P
PyTorch武侠演义第一卷：初入江湖第7章：矿洞中的计算禁制空中湖 pytorch武侠演绎 pytorch 人工智能 python
第一卷：初入江湖第7章：矿洞中的计算禁制矿洞深处罗盘残件在接近矿洞时突然发热，指针疯狂旋转。"就是这里，"欧阳长老抚摸着洞壁上的计算图刻痕，“TensorFlow帮用静态图封印了矿脉。”林小码看到：幽蓝矿脉构成巨大的计算图结构水晶矿簇随呼吸节奏明灭（CUDA核心）矿道中流淌着数据光流（内存带宽）"小心！"大师突然拉回林小码。他刚才踩中的矿砖下陷，触发岩壁上的机关——数十道计算图锁链从四面八方射来！
数字人克隆中SyncTalk算法介绍与部署过程优秘智能UMI 人工智能 ubuntu
SyncTalk算法介绍SyncTalk合成同步的头部说话视频，采用三平面哈希表示来保持主体身份。它可以生成同步的嘴唇动作、面部表情和稳定的头部姿势，并恢复头发细节以创建高分辨率视频。部署在Linux中部署该项目，在Ubuntu18.04、Pytorch1.12.1和CUDA11.3上测试。gitclonehttps://github.com/ZiqiaoPeng/SyncTalk.gitcdSy
风格迁移（Style Transfer）
1.什么是风格迁移（StyleTransfer）：简单介绍风格迁移的概念，指的是将一张图像的内容与另一张图像的艺术风格结合起来，从而生成一个新的图像。例如，将一张风景图像的内容与一幅著名艺术作品（如梵高的《星夜》）的风格结合。应用场景：风格迁移常用于图像生成、艺术创作和增强现实等领域。目标：本文将讲解如何使用PyTorch和VGG19模型实现风格迁移，并展示其核心代码。2.风格迁移的原理在这一部分
标签助手：基于LabelImg和YOLOv5的图像半自动标注工具伏容一Julia
标签助手：基于LabelImg和YOLOv5的图像半自动标注工具项目基础介绍标签助手（labelGo-Yolov5AutoLabelImg）是一个图形化的半自动图像注解工具，它结合了广受欢迎的图像标注工具LabelImg的力量与先进的目标检测框架YOLOv5。这个开源项目旨在简化数据集的标注过程，利用现有YOLOv5PyTorch模型实现快速的半自动化标注，极大地提高了标注效率。项目主要采用Pyt
【MMCV】MMCV安装与踩坑 Elendill Pyhton pytorch python mmcv
确认MMCV版本首先确认项目所需MMCV的版本是多少mmcv2.0版本的代码相比较于=2.0.0安装方法新创建一个conda环境安装pytorch：condainstallpytorchtorchvisiontorchaudiopytorch-cuda=11.8-cpytorch-cnvidia安装mim，这是openmm官方推出的用于安装他们旗下mm系列产品的安装器：pipinstall-Uop
基于Jetson Nano与PyTorch的无人机实时目标跟踪系统搭建指南
引言：边缘计算赋能智能监控在AIoT时代，将深度学习模型部署到嵌入式设备已成为行业刚需。本文将手把手指导读者在NVIDIAJetsonNano（4GB版本）开发板上，构建基于YOLOv5+SORT算法的实时目标跟踪系统，集成无人机控制与地面站监控界面，最终打造低功耗智能监控设备。通过本项目，读者将掌握：嵌入式端模型优化与部署技巧；多目标跟踪算法工程化实现；无人机-地面站协同控制架构；边缘计算场景下
【语义分割专栏】4：deeplab系列实战篇(附上完整可运行的代码pytorch) fouen 语义分割 pytorch 人工智能 python 计算机视觉深度学习
文章目录前言Deeplab系列全流程代码模型搭建(model)backbone的搭建Deeplabv1Deeplabv2Deeplabv3Deeplabv3+数据处理(dataloader)评价指标(metric)训练流程(train)模型测试(test)效果图结语前言Deeplab系列原理篇讲解：【语义分割专栏】4：deeplab系列原理篇_deeplab系列详解-CSDN博客代码地址，下载可复
pytorch学习笔记-自定义卷积墨染枫深度学习 pytorch 学习笔记
未完结的草稿———！大概是准备整合一下常见的层，整合完感觉就可以进行搭建了（还没进行到这一步所以不太确定版）（ps我将在完结这一篇的时候删除上面的小字and二编一下整篇文章的结构，如果看到了这部分文字也是很有缘分了/doge这一部分感觉也没啥好说的==也就是reshape部分值得注意一下？剩下的感觉就是了解一下用法就可以importtorchimporttorch.nnasnnimporttorc
PyTorch武侠演义第一卷：初入江湖第5章：玉如意的秘密
第一卷：初入江湖第5章：玉如意的秘密百年秘辛藏经阁最深处，大师掀开尘封的《门派大事记》，指向一幅泛黄的画卷：“看，这就是百年前的优化器长老——欧阳调参。”画中人手持玉如意，面前悬浮着九个水晶球。林小码凑近细看，发现如意上刻着「lr=0.001」。“当年TensorFlow帮为何要盗损失玉佩？”大师叹息：“因为这块玉佩，正是控制玉如意能量的钥匙…”突然，书架后传来机关转动的咔嗒声。一道暗门缓缓打开，
使用PyTorch实现目标检测与跟踪认真写代码i pytorch 目标检测人工智能 Python
目标检测与跟踪是计算机视觉领域中的重要任务，它可以帮助我们在图像或视频中准确地定位和跟踪特定物体。PyTorch是一个流行的深度学习框架，提供了强大的工具和库，可以用于目标检测与跟踪的实现。本文将详细介绍如何使用PyTorch实现目标检测与跟踪，并提供相应的源代码。安装PyTorch和相关依赖首先，我们需要安装PyTorch和其他必要的依赖项。你可以通过以下命令使用pip安装PyTorch：pip
Pytorch 自定义损失函数 DeniuHe Pytorch
自定义HingeLossclassMyHingeLoss(torch.nn.Module):#不要忘记继承Moduledef__init__(self):super(MyHingeLoss,self).__init__()defforward(self,output,target):"""output和target都是1-D张量,换句话说,每个样例的返回是一个标量."""hinge_loss=1-
Pytorch实现目标检测
importosimportrandomimportpandasaspdimportnumpyasnpimportcv2fromsklearn.model_selectionimporttrain_test_splitimporttorchfromtorch.utils.dataimportDataset,DataLoaderimporttorch.nnasnnimporttorch.nn.fun
解决TensorBoard报错“log_dir is not a directory“的完整指南 SEVEN是7 tensorflow python 深度学习
在使用PyTorch的TensorBoard进行训练可视化时，许多开发者会遇到FailedPreconditionError:./文件名isnotadirectory的错误。本文将深入分析这个问题的根源，并提供完整的解决方案，特别是针对中文路径这一常见但容易被忽视的问题。问题一（目录确实存在的情况下）：路径中包含中文解决：更改文件名为正确的命名格式（注意：连模型的文件名也不要用中文，确保绝对路径全
Pytorch混合精度训练最佳实践贝塔西塔工程经验 pytorch 人工智能深度学习混合精度模型加速
混合精度训练（MixedPrecisionTraining）是一种通过结合单精度（FP32）和半精度（FP16/FP8）计算来加速训练、减少显存占用的技术。它在保持模型精度的同时，通常能带来2-3倍的训练速度提升，并减少约50%的显存使用，是平衡训练效率与数值稳定性的核心技术，尤其在大模型训练中不可或缺。以下从GradScaler底层逻辑、避坑技巧（含NaN解决方案）、PyTorchLightni
Pytorch自定义优化器最佳实践
在PyTorch中，自定义优化器需要遵循特定的规范以兼容PyTorch的训练流程。下面从核心方法、closure的作用，到Ranger优化器的实现，逐步展开说明。一、PyTorch自定义优化器的必要方法自定义优化器必须继承torch.optim.Optimizer，并实现以下核心方法：init(self,params,defaults)作用：初始化优化器，定义超参数（如学习率、动量等），并为参数组
PytorchLightning最佳实践基础篇贝塔西塔工程经验 pytorch Lightning 深度学习编程框架
PyTorchLightning（简称PL）是一个建立在PyTorch之上的高层框架，核心目标是剥离工程代码与研究逻辑，让研究者专注于模型设计和实验思路，而非训练循环、分布式配置、日志管理等重复性工程工作。本文从基础到进阶，全面介绍其功能、核心组件、封装逻辑及最佳实践。一、PyTorchLightning核心价值原生PyTorch训练代码中，大量精力被消耗在：手动编写训练/验证循环（epoch、b
Linux指令&&ros学习&&python深度学习&&bug学习笔记起个别名 Linux ROS Python
##这个文件是关于ros、linux指令，pytorch、python、onnx和相关problem的一些笔记###ROS&&linux**find:在当前路径或指定的路径下递归地搜索文件或目录，并可以根据不同的条件进行过滤和匹配。**```find-name*.pyfind/home/dai/bev_lane_det-main-namemodelsfind/home/dai/bev_lane_d
零基础完整版入门经典深度学习时间序列预测项目实战+最新前沿时间序列预测模型代码讲解学习整理（附完整可运行代码） OverOnEarth 时间序列预测项目实战深度学习学习人工智能
专栏内容本专栏主要整理了作者在时间序列预测领域内的一些学习思路与代码整理，帮助大家在初进入此领域时，可以快速掌握代码进行实战操作，对代码的操作再结合论文阅读肯定是上升更快嘛，作者也愿意和大家一起讨论进步，下面的内容会逐步更新，作者主页的资源列也会放出一些可下载的资源供大家参考学习噢。一、LSTM时间序列预测完整代码示例学习分析（pytorch框架）精选试读文章二、LSTM多变量输入实现多步预测完整
第十四章:AI的数据“集装箱”：彻底搞懂Tensor的Batch与维度爱分享的飘哥 AI新纪元：120日觉醒计划 Tensor PyTorch BatchSize 数据处理 AI基础深度学习教程
AI数据集中箱前言：为什么AI从不“零售”，总是“批发”？1：Batch（批次）——GPU的“灵魂伴侣”1.1单个处理vs.批量处理：CPU与GPU的思维差异1.2DataLoader：PyTorch的“自动化装箱员”2：维度的语言——破译[B,L,D]的含义2.1[L,D]：一个句子的“二维画像”2.2[B,L,D]：一批句子的“三维魔方”2.3用代码直观感受维度的增加3：追踪Tensor的“变
PyTorch中实现早停机制（EarlyStopping）附代码自信的小螺丝钉 AI知识 pytorch python 人工智能 AI 深度学习
1.核心目的当模型在验证集上的性能不再提升时，提前终止训练防止过拟合，节省计算资源2.实现方法监控验证集指标（如损失、准确率），设置耐心值（Patience）3.代码：classEarlyStopping:def__init__(self,patience=10,delta=0):"""EarlystoppingArgs:patience:int,numberofepochstowaitbefor
【已解决】YOLO11模型转wts时报错:PytorchStreamReader failed reading zip archive lxmyzzs bug 人工智能 python 计算机视觉目标检测神经网络深度学习
问题：在把训练好的新YOLO11s模型转wts文件时报错，具体信息如下图（PytorchStreamReaderfailedreadingziparchive:failedfindingcentraldirectory）解决：新老版本pytorch之间的兼容问题，改动一下生成wts文件即可。代码帖在下面。importsys#noqa:F401importargparseimportosimport
Pytorch实现细节解析：Transformer模型的Encoder与Decoder逐行代码讲解 lazycatlove pytorch transformer 人工智能
文章目录摘要一、Transformer1.1为什么要使用attention1.2Transformer的优点二、Transformer模型Encoder和Decoder原理讲解与其Pytorch逐行实现2.1wordembedding2.2单词索引构成源句子和目标句子2.3构建positionembedding2.4构造encoder的self-attentionmask2.5构造intra-at
Transformer Masked loss原理精讲及其PyTorch逐行实现
MaskedLoss的核心原理是：在计算损失函数时，只考虑真实有意义的词元（token），而忽略掉为了数据对齐而填充的无意义的填充词元（paddingtoken）。这是重要的技术，可以确保模型专注于学习有意义的任务，并得到一个正确的性能评估。1.原理精讲为什么需要MaskedLoss？在训练神经网络时，我们通常会用一个批次（batch）的数据进行训练，而不是一次只用一个样本。对于自然语言处理任务，
Transformer模型Decoder原理精讲及其PyTorch逐行实现老鱼说AI transformer pytorch 深度学习人工智能学习 python
原理：Decoder的核心是一个自回归(Auto-regressive)的生成器。它的任务是在给定源序列的编码表示(encoder_outputs)和已生成的目标序列部分(y_1,...,y_{t-1})的条件下，预测出下一个词y_t的概率分布。一个标准的DecoderLayer包含三个核心子层：1.带掩码的多头自注意力(MaskedMulti-HeadSelf-Attention):用于处理已生
2025暑期—07YOLO-YOLOV11 宇称不守恒4.0 人工智能图像处理 YOLO 深度学习人工智能
安装的环境包括YoloV11，torch2.32.4Clip1.0D2LOpenCV4.12等安装1Conda环境安装YOLOcondacreate--prefix=D:/YOLO11/yolo11_envpython=3.10condaactivateD:\YOLO11\yolo11_envPytorch网站确定condainstallpytorch==2.3.0torchvision==0.1
PyTorch中的词嵌入层(nn.Embedding)详解与实践指南慕婉0307 自然语言处理 pytorch embedding 人工智能
一、词嵌入(WordEmbedding)简介词嵌入是自然语言处理(NLP)中的一项核心技术，它将离散的词语映射到连续的向量空间中。通过词嵌入，语义相似的词语在向量空间中的位置也会相近。为什么需要词嵌入？解决维度灾难：传统one-hot编码维度等于词汇表大小，而词嵌入维度可自定义捕捉语义关系：通过向量空间中的距离反映词语间的语义关系迁移学习：预训练的词嵌入可以在不同任务间共享二、PyTorch中的n
eclipse maven IXHONG eclipse
eclipse中使用maven插件的时候，运行run as maven build的时候报错 -Dmaven.multiModuleProjectDirectory system propery is not set. Check $M2_HOME environment variable and mvn script match. 可以设一个环境变量M2_HOME指
timer cancel方法的一个小实例 alleni123 多线程 timer
package com.lj.timer; import java.util.Date; import java.util.Timer; import java.util.TimerTask; public class MyTimer extends TimerTask { private int a; private Timer timer; pub
MySQL数据库在Linux下的安装 ducklsl mysql
1.建好一个专门放置MySQL的目录 /mysql/db数据库目录 /mysql/data数据库数据文件目录 2.配置用户，添加专门的MySQL管理用户 >groupadd mysql ----添加用户组 >useradd -g mysql mysql ----在mysql用户组中添加一个mysql用户 3.配置，生成并安装MySQL >cmake -D
spring------>>cvc-elt.1: Cannot find the declaration of element Array_06 spring bean
将-------- <?xml version="1.0" encoding="UTF-8"?> <beans xmlns="http://www.springframework.org/schema/beans" xmlns:xsi="http://www.w3
maven发布第三方jar的一些问题 cugfy maven
maven中发布第三方jar到nexus仓库使用的是 deploy:deploy-file命令有许多参数，具体可查看 http://maven.apache.org/plugins/maven-deploy-plugin/deploy-file-mojo.html 以下是一个例子： mvn deploy:deploy-file -DgroupId=xpp3
MYSQL下载及安装 357029540 mysql
好久没有去安装过MYSQL，今天自己在安装完MYSQL过后用navicat for mysql去厕测试链接的时候出现了10061的问题，因为的的MYSQL是最新版本为5.6.24，所以下载的文件夹里没有my.ini文件，所以在网上找了很多方法还是没有找到怎么解决问题，最后看到了一篇百度经验里有这个的介绍，按照其步骤也完成了安装，在这里给大家分享下这个链接的地址
ios TableView cell的布局张亚雄 tableview
cell.imageView.image = [UIImage imageNamed:[imageArray objectAtIndex:[indexPath row]]]; CGSize itemSize = CGSizeMake(60, 50); &nbs
Java编码转义 adminjun java 编码转义
import java.io.UnsupportedEncodingException; /** * 转换字符串的编码 */ public class ChangeCharset { /** 7位ASCII字符，也叫作ISO646-US、Unicode字符集的基本拉丁块 */ public static final Strin
Tomcat 配置和spring aijuans spring
简介 Tomcat启动时，先找系统变量CATALINA_BASE，如果没有，则找CATALINA_HOME。然后找这个变量所指的目录下的conf文件夹，从中读取配置文件。最重要的配置文件：server.xml 。要配置tomcat，基本上了解server.xml，context.xml和web.xml。 Server.xml -- tomcat主
Java打印当前目录下的所有子目录和文件 ayaoxinchao 递归 File
其实这个没啥技术含量，大湿们不要操笑哦，只是做一个简单的记录，简单用了一下递归算法。 import java.io.File; /** * @author Perlin * @date 2014-6-30 */ public class PrintDirectory { public static void printDirectory(File f
linux安装mysql出现libs报冲突解决 BigBird2012 linux
linux安装mysql出现libs报冲突解决安装mysql出现 file /usr/share/mysql/ukrainian/errmsg.sys from install of MySQL-server-5.5.33-1.linux2.6.i386 conflicts with file from package mysql-libs-5.1.61-4.el6.i686
jedis连接池使用实例 bijian1013 redis jedis连接池 jedis
实例代码： package com.bijian.study; import java.util.ArrayList; import java.util.List; import redis.clients.jedis.Jedis; import redis.clients.jedis.JedisPool; import redis.clients.jedis.JedisPoo
关于朋友 bingyingao 朋友兴趣爱好维持
成为朋友的必要条件：志相同，道不合，可以成为朋友。譬如马云、周星驰一个是商人，一个是影星，可谓道不同，但都很有梦想，都要在各自领域里做到最好，当他们遇到一起，互相欣赏，可以畅谈两个小时。志不同，道相合，也可以成为朋友。譬如有时候看到两个一个成绩很好每次考试争做第一，一个成绩很差的同学是好朋友。他们志向不相同，但他
【Spark七十九】Spark RDD API一 bit1129 spark
aggregate package spark.examples.rddapi import org.apache.spark.{SparkConf, SparkContext} //测试RDD的aggregate方法 object AggregateTest { def main(args: Array[String]) { val conf = new Spar
ktap 0.1 released bookjovi kernel tracing
Dear, I'm pleased to announce that ktap release v0.1, this is the first official release of ktap project, it is expected that this release is not fully functional or very stable and we welcome bu
能保存Properties文件注释的Properties工具类 BrokenDreams properties
今天遇到一个小需求：由于java.util.Properties读取属性文件时会忽略注释，当写回去的时候，注释都没了。恰好一个项目中的配置文件会在部署后被某个Java程序修改一下，但修改了之后注释全没了，可能会给以后的参数调整带来困难。所以要解决这个问题。 &nb
读《研磨设计模式》-代码笔记-外观模式-Facade bylijinnan java 设计模式
声明：本文只为方便我个人查阅和理解，详细的分析以及源代码请移步原作者的博客http://chjavach.iteye.com/ /* * 百度百科的定义： * Facade（外观）模式为子系统中的各类（或结构与方法）提供一个简明一致的界面， * 隐藏子系统的复杂性，使子系统更加容易使用。他是为子系统中的一组接口所提供的一个一致的界面 * * 可简单地
After Effects教程收集 cherishLC After Effects
1、中文入门 http://study.163.com/course/courseMain.htm?courseId=730009 2、videocopilot英文入门教程（中文字幕） http://www.youku.com/playlist_show/id_17893193.html 英文原址： http://www.videocopilot.net/basic/ 素
Linux Apache 安装过程 crabdave apache
Linux Apache 安装过程下载新版本： apr-1.4.2.tar.gz（下载网站：http://apr.apache.org/download.cgi） apr-util-1.3.9.tar.gz（下载网站：http://apr.apache.org/download.cgi） httpd-2.2.15.tar.gz（下载网站：http://httpd.apac
Shell学习之变量赋值和引用 daizj shell 变量引用赋值
本文转自：http://www.cnblogs.com/papam/articles/1548679.html Shell编程中，使用变量无需事先声明，同时变量名的命名须遵循如下规则：首个字符必须为字母（a-z，A-Z）中间不能有空格，可以使用下划线（_）不能使用标点符号不能使用bash里的关键字（可用help命令查看保留关键字）需要给变量赋值时，可以这么写：
Java SE 第一讲（Java SE入门、JDK的下载与安装、第一个Java程序、Java程序的编译与执行） dcj3sjt126com java jdk
Java SE 第一讲： Java SE：Java Standard Edition Java ME: Java Mobile Edition Java EE：Java Enterprise Edition Java是由Sun公司推出的（今年初被Oracle公司收购）。收购价格：74亿美金 J2SE、J2ME、J2EE JDK：Java Development
YII给用户登录加上验证码 dcj3sjt126com yii
1、在SiteController中添加如下代码： /** * Declares class-based actions. */ public function actions() { return array( // captcha action renders the CAPTCHA image displ
Lucene使用说明 dyy_gusi Lucene search 分词器
Lucene使用说明 1、lucene简介 1.1、什么是lucene Lucene是一个全文搜索框架，而不是应用产品。因此它并不像baidu或者googleDesktop那种拿来就能用，它只是提供了一种工具让你能实现这些产品和功能。 1.2、lucene能做什么要回答这个问题，先要了解lucene的本质。实际
学习编程并不难,做到以下几点即可! gcq511120594 数据结构编程算法
不论你是想自己设计游戏，还是开发iPhone或安卓手机上的应用，还是仅仅为了娱乐，学习编程语言都是一条必经之路。编程语言种类繁多，用途各异，然而一旦掌握其中之一，其他的也就迎刃而解。作为初学者，你可能要先从Java或HTML开始学，一旦掌握了一门编程语言，你就发挥无穷的想象，开发各种神奇的软件啦。 1、确定目标学习编程语言既充满乐趣，又充满挑战。有些花费多年时间学习一门编程语言的大学生到
Java面试十问之三：Java与C++内存回收机制的差别 HNUlanwei java C++finalize()堆栈内存回收
大家知道， Java 除了那 8 种基本类型以外，其他都是对象类型（又称为引用类型）的数据。 JVM 会把程序创建的对象存放在堆空间中，那什么又是堆空间呢？其实，堆（ Heap）是一个运行时的数据存储区，从它可以分配大小各异的空间。一般，运行时的数据存储区有堆（ Heap）和堆栈（ Stack），所以要先看它们里面可以分配哪些类型的对象实体，然后才知道如何均衡使用这两种存储区。一般来说，栈中存放的
第二章 Nginx+Lua开发入门 jinnianshilongnian nginx lua
Nginx入门本文目的是学习Nginx+Lua开发，对于Nginx基本知识可以参考如下文章： nginx启动、关闭、重启 http://www.cnblogs.com/derekchen/archive/2011/02/17/1957209.html agentzh 的 Nginx 教程 http://openresty.org/download/agentzh-nginx-tutor
MongoDB windows安装基本命令 liyonghui160com
windows安装安装目录： D:\MongoDB\ 新建目录 D:\MongoDB\data\db 4.启动进城： cd D:\MongoDB\bin mongod -dbpath D:\MongoDB\data\db &n
Linux下通过源码编译安装程序 pda158 linux
一、程序的组成部分　　Linux下程序大都是由以下几部分组成：　　二进制文件：也就是可以运行的程序文件　　库文件：就是通常我们见到的lib目录下的文件　　配置文件：这个不必多说，都知道　　帮助文档：通常是我们在linux下用man命令查看的命令的文档　　二、linux下程序的存放目录　　linux程序的存放目录大致有三个地方：　　/etc, /b
WEB开发编程的职业生涯４个阶段 shw3588 编程 Web 工作生活
觉得自己什么都会 2007年从学校毕业，凭借自己原创的ASP毕业设计，以为自己很厉害似的，信心满满去东莞找工作，找面试成功率确实很高，只是工资不高，但依旧无法磨灭那过分的自信，那时候什么考勤系统、什么OA系统、什么ERP，什么都觉得有信心，这样的生涯大概持续了约一年。根本不是自己想的那样 2008年开始接触很多工作相关的东西，发现太多东西自己根本不会，都需要去学，不管是asp还是js，
遭遇jsonp同域下变作post请求的坑 vb2005xu jsonp 同域post
今天迁移一个站点时遇到一个坑爹问题,同一个jsonp接口在跨域时都能调用成功,但是在同域下调用虽然成功,但是数据却有问题. 此处贴出我的后端代码片段 $mi_id = htmlspecialchars(trim($_GET['mi_id '])); $mi_cv = htmlspecialchars(trim($_GET['mi_cv '])); 贴出我前端代码片段: $.aj