MultiNet：自主駕駛中的實時聯合語義推理 論文代碼

原創 QiangLi_strong 2020-02-21 12:33

MultiNet論文相關

論文下載地址:原文地址、免翻牆地址

論文Github地址：KittiSeg

論文翻譯參考：csdn

程序

Get Start

git clone https://github.com/MarvinTeichmann/KittiSeg.git
cd KittiSeg
git submodule update --init --recursive
python download_data.py --kitti_url http://kitti.is.tue.mpg.de/kitti/data_road.zip

下載過程

2017-11-24 13:41:12,536 INFO Downloading VGG weights.
2017-11-24 13:41:12,538 INFO Download URL: ftp://mi.eng.cam.ac.uk/pub/mttt2/models/vgg16.npy
2017-11-24 13:41:12,538 INFO Download DIR: DATA
>> Downloading vgg16.npy 100.0%
2017-11-24 13:58:29,855 INFO Downloading Kitti Road Data.
2017-11-24 13:58:29,855 INFO Download URL: http://kitti.is.tue.mpg.de/kitti/data_road.zip
2017-11-24 13:58:29,856 INFO Download DIR: DATA
>> Downloading data_road.zip 100.0%
2017-11-24 14:46:04,978 INFO Extracting kitti_road data.
2017-11-24 14:46:09,240 INFO Preparing kitti_road data.
2017-11-24 14:46:09,244 INFO All data have been downloaded successful.

運行Demo

Windows下運行出錯問題解決

python demo.py --input_image data/demo/demo.png
2017-11-27 14:12:16,096 INFO No environment variable 'TV_PLUGIN_DIR' found. Set to 'C:\Users\10217814/tv-plugins'.
2017-11-27 14:12:16,096 INFO No environment variable 'TV_STEP_SHOW' found. Set to '50'.
2017-11-27 14:12:16,096 INFO No environment variable 'TV_STEP_EVAL' found. Set to '250'.
2017-11-27 14:12:16,096 INFO No environment variable 'TV_STEP_WRITE' found. Set to '1000'.
2017-11-27 14:12:16,097 INFO No environment variable 'TV_MAX_KEEP' found. Set to '10'.
2017-11-27 14:12:16,097 INFO No environment variable 'TV_STEP_STR' found. Set to 'Step {step}/{total_steps}: loss = {loss_value:.2f}; lr = {lr_value:.2e}; {sec_per_batch:.3f} sec (per Batch); {examples_per_sec:.1f} imgs/sec'.
2017-11-27 14:12:16,107 INFO Download URL: ftp://mi.eng.cam.ac.uk/pub/mttt2/models/KittiSeg_pretrained.zip
2017-11-27 14:12:16,108 INFO Download DIR: RUNS
>> Downloading KittiSeg_pretrained.zip 100.0%
2017-11-27 17:13:03,549 INFO Extracting KittiSeg_pretrained.zip
2017-11-27 17:13:24,930 INFO f: <_io.TextIOWrapper name='RUNS\\KittiSeg_pretrained\\model_files\\hypes.json' mode='r' encoding='cp936'>
2017-11-27 17:13:24,931 INFO Hypes loaded successfully.
2017-11-27 17:13:25,072 INFO Modules loaded successfully. Starting to build tf graph.
npy file loaded
Layer name: conv1_1
Layer shape: (3, 3, 3, 64)
2017-11-27 17:13:29,842 INFO Creating Summary for: conv1_1/filter
2017-11-27 17:13:29,890 INFO Creating Summary for: conv1_1/biases
Layer name: conv1_2
Layer shape: (3, 3, 64, 64)
2017-11-27 17:13:29,923 INFO Creating Summary for: conv1_2/filter
2017-11-27 17:13:29,947 INFO Creating Summary for: conv1_2/biases
Layer name: conv2_1
Layer shape: (3, 3, 64, 128)
2017-11-27 17:13:29,979 INFO Creating Summary for: conv2_1/filter
2017-11-27 17:13:30,004 INFO Creating Summary for: conv2_1/biases
Layer name: conv2_2
Layer shape: (3, 3, 128, 128)
2017-11-27 17:13:30,036 INFO Creating Summary for: conv2_2/filter
2017-11-27 17:13:30,061 INFO Creating Summary for: conv2_2/biases
Layer name: conv3_1
Layer shape: (3, 3, 128, 256)
2017-11-27 17:13:30,095 INFO Creating Summary for: conv3_1/filter
2017-11-27 17:13:30,119 INFO Creating Summary for: conv3_1/biases
Layer name: conv3_2
Layer shape: (3, 3, 256, 256)
2017-11-27 17:13:30,152 INFO Creating Summary for: conv3_2/filter
2017-11-27 17:13:30,175 INFO Creating Summary for: conv3_2/biases
Layer name: conv3_3
Layer shape: (3, 3, 256, 256)
2017-11-27 17:13:30,209 INFO Creating Summary for: conv3_3/filter
2017-11-27 17:13:30,232 INFO Creating Summary for: conv3_3/biases
Layer name: conv4_1
Layer shape: (3, 3, 256, 512)
2017-11-27 17:13:30,268 INFO Creating Summary for: conv4_1/filter
2017-11-27 17:13:30,293 INFO Creating Summary for: conv4_1/biases
Layer name: conv4_2
Layer shape: (3, 3, 512, 512)
2017-11-27 17:13:30,332 INFO Creating Summary for: conv4_2/filter
2017-11-27 17:13:30,356 INFO Creating Summary for: conv4_2/biases
Layer name: conv4_3
Layer shape: (3, 3, 512, 512)
2017-11-27 17:13:30,396 INFO Creating Summary for: conv4_3/filter
2017-11-27 17:13:30,422 INFO Creating Summary for: conv4_3/biases
Layer name: conv5_1
Layer shape: (3, 3, 512, 512)
2017-11-27 17:13:30,461 INFO Creating Summary for: conv5_1/filter
2017-11-27 17:13:30,485 INFO Creating Summary for: conv5_1/biases
Layer name: conv5_2
Layer shape: (3, 3, 512, 512)
2017-11-27 17:13:30,565 INFO Creating Summary for: conv5_2/filter
2017-11-27 17:13:30,589 INFO Creating Summary for: conv5_2/biases
Layer name: conv5_3
Layer shape: (3, 3, 512, 512)
2017-11-27 17:13:30,629 INFO Creating Summary for: conv5_3/filter
2017-11-27 17:13:30,655 INFO Creating Summary for: conv5_3/biases
Layer name: fc6
Layer shape: [7, 7, 512, 4096]
2017-11-27 17:13:30,965 INFO Creating Summary for: fc6/weights
2017-11-27 17:13:30,991 INFO Creating Summary for: fc6/biases
Layer name: fc7
Layer shape: [1, 1, 4096, 4096]
2017-11-27 17:13:31,067 INFO Creating Summary for: fc7/weights
2017-11-27 17:13:31,091 INFO Creating Summary for: fc7/biases
2017-11-27 17:13:31,127 INFO Creating Summary for: score_fr/weights
2017-11-27 17:13:31,151 INFO Creating Summary for: score_fr/biases
WARNING:tensorflow:From e:\GitHub\KittiSeg\incl\tensorflow_fcn\fcn8_vgg.py:114: calling argmax (from tensorflow.python.ops.math_ops) with dimension is deprecated and will be removed in a future version.
Instructions for updating:
Use the `axis` argument instead
2017-11-27 17:13:31,465 WARNING From e:\GitHub\KittiSeg\incl\tensorflow_fcn\fcn8_vgg.py:114: calling argmax (from tensorflow.python.ops.math_ops) with dimension is deprecated and will be removed in a future version.
Instructions for updating:
Use the `axis` argument instead
2017-11-27 17:13:31,488 INFO Creating Summary for: upscore2/up_filter
2017-11-27 17:13:31,525 INFO Creating Summary for: score_pool4/weights
2017-11-27 17:13:31,548 INFO Creating Summary for: score_pool4/biases
2017-11-27 17:13:31,594 INFO Creating Summary for: upscore4/up_filter
2017-11-27 17:13:31,634 INFO Creating Summary for: score_pool3/weights
2017-11-27 17:13:31,659 INFO Creating Summary for: score_pool3/biases
2017-11-27 17:13:31,705 INFO Creating Summary for: upscore32/up_filter
2017-11-27 17:13:31,778 INFO Graph build successfully.
2017-11-27 17:13:31.779807: I C:\tf_jenkins\home\workspace\rel-win\M\windows\PY\35\tensorflow\core\platform\cpu_feature_guard.cc:137] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX AVX2
2017-11-27 17:13:31,918 INFO /u/marvin/no_backup/RUNS/KittiSeg/loss_bench/xentropy_kitti_fcn_2016_10_15_01.18/model.ckpt-15999
INFO:tensorflow:Restoring parameters from RUNS\KittiSeg_pretrained\model.ckpt-15999
2017-11-27 17:13:31,919 INFO Restoring parameters from RUNS\KittiSeg_pretrained\model.ckpt-15999
2017-11-27 17:13:35,227 INFO Weights loaded successfully.
2017-11-27 17:13:35,227 INFO Starting inference using data/demo/demo.png as input
2017-11-27 17:13:40,363 INFO
2017-11-27 17:13:40,363 INFO Raw output image has been saved to: e:\GitHub\KittiSeg\data\demo\demo_raw.png
2017-11-27 17:13:40,363 INFO Red-Blue overlay of confs have been saved to: e:\GitHub\KittiSeg\data\demo\demo_rb.png
2017-11-27 17:13:40,364 INFO Green plot of predictions have been saved to: e:\GitHub\KittiSeg\data\demo\demo_green.png
2017-11-27 17:13:40,364 INFO
2017-11-27 17:13:40,364 WARNING Do NOT use this Code to evaluate multiple images.
2017-11-27 17:13:40,364 WARNING Demo.py is **very slow** and designed to be a tutorial to show how the KittiSeg works.
2017-11-27 17:13:40,365 WARNING
2017-11-27 17:13:40,365 WARNING Please see this comment, if you like to apply demo.py tomultiple images see:
2017-11-27 17:13:40,365 WARNING https://github.com/MarvinTeichmann/KittiBox/issues/15#issuecomment-301800058

運行結果：

demo.png

img

demo_raw.png

img

demo_rb.png

img

demo_green.png

img

代碼筆記：

  1. 輸出日誌中前幾行的No environment variable由incl/tensorvision/utils.py中的325-339輸出，負責讀取環境變量中的一些設置。
  2. KittiSeg_pretrained.zip由demo.py中的125行代碼下載，大小爲2.56GB，默認下載到RUNS/文件夾中，並解壓到同一目錄中。
  3. 接下來讀取hypes和modules.
    1. hypes是指RUNS/KittiSeg_pretrained/model_files/hypes.json文件
    2. modules是一個dict，存放了一些模塊包，包括KittiSeg_pretrained/model_files/ 下的data_input.py, architecture.py, objective.py, solver.py, eval.py.
    3. 以下代碼
      python
      f = os.path.join(model_dir, "eval.py")
      eva = imp.load_source("evaluator_%s" % postfix, f)

      表示將eval.py這個文件進行導入，導入後的模塊名稱爲evluator_%s，eva爲模塊變量。
  4. 開始構建Tensorflow Graph。首先創建2個placeholder
    python
    # Create placeholder for input
    image_pl = tf.placeholder(tf.float32)
    image = tf.expand_dims(image_pl, 0)

    佔位符的shape爲未知

    >>> image_pl
    <tf.Tensor 'Placeholder:0' shape=<unknown> dtype=float32>
    >>> image
    <tf.Tensor 'ExpandDims:0' shape=<unknown> dtype=float32>

  5. 從module中讀取Graph

    
# build Tensorflow graph using the model from logdir

prediction = core.build_inference_graph(hypes, modules, image=image)

----------incl/tensorvision/core.py
def build_inference_graph(hypes, modules, image):
    with tf.name_scope("Validation"):

    logits = modules['arch'].inference(hypes, image, train=False)

    decoded_logits = modules['objective'].decoder(hypes, logits,
                                                  train=False)
return decoded_logits

----------RUNS/KittiSeg_pretrained/model_files/architecture.py
def inference(hypes, images, train=True):
    vgg16_npy_path = os.path.join(hypes['dirs']['data_dir'], "vgg16.npy")
    vgg_fcn = fcn8_vgg.FCN8VGG(vgg16_npy_path=vgg16_npy_path)

    vgg_fcn.wd = hypes['wd']

    vgg_fcn.build(images, train=train, num_classes=2, random_init_fc8=True)

    return vgg_fcn.upscore32

----------RUNS/KittiSeg_pretrained/model_files/objective.py
def decoder(hypes, logits, train):
    decoded_logits = {}
    decoded_logits['logits'] = logits
    decoded_logits['softmax'] = _add_softmax(hypes, logits)
    return decoded_logits

    上文運行download_data.py已經下載好了vgg16.npy，這是VGG-16網絡實現訓練好的參數。日誌顯示的Inference結構爲：

    Layer Name Layer Shape
    conv1_1 (3, 3, 3, 64)
    conv1_2 (3, 3, 64, 64)
    conv2_1 (3, 3, 64, 128)
    conv2_2 (3, 3, 128, 128)
    conv3_1 (3, 3, 128, 256)
    conv3_2 (3, 3, 256, 256)
    conv3_3 (3, 3, 256, 256)
    conv4_1 (3, 3, 256, 512)
    conv4_2 (3, 3, 512, 512)
    conv4_3 (3, 3, 512, 512)
    conv5_1 (3, 3, 512, 512)
    conv5_2 (3, 3, 512, 512)
    conv5_3 (3, 3, 512, 512)
    fc6 [7, 7, 512, 4096]
    fc7 [1, 1, 4096, 4096]
    score_fr
    upscore2
    score_pool4
    upscore4
    score_pool3
    upscore32

  6. 加載本地的權重網絡變量

    
# Create a session for running Ops on the Graph.

sess = tf.Session()
saver = tf.train.Saver()


# Load weights from logdir

core.load_weights(logdir, sess, saver)

    在RUNS/KittiSeg_pretrained下包含了ckpt文件，load_weights函數會自動讀取目錄下的checkpoint文件，並得到實際的參數文件，然後save.restore。

  7. 讀取且重定義測試圖像，demo中未執行。
    python
    # Load and resize input image
    image = scp.misc.imread(input_image)
    if hypes['jitter']['reseize_image']:
    # Resize input only, if specified in hypes
    image_height = hypes['jitter']['image_height']
    image_width = hypes['jitter']['image_width']
    image = scp.misc.imresize(image, size=(image_height, image_width),
    interp='cubic')

    scp.misc.imread返回的類型爲<class 'numpy.ndarray'>類型，原始shape爲(375, 1242, 3), 而hypes中圖像大小爲(384,1248), 利用cubic三次樣條插值算法進行縮放。
  8. Tensorflow運行預測任務。
    python
    # Run KittiSeg model on image
    feed = {image_pl: image}
    softmax = prediction['softmax']
    output = sess.run([softmax], feed_dict=feed)

    softmax爲輸出層，輸出類別爲2，如下

    Tensor("Validation/decoder/Softmax:0", shape=(?, 2), dtype=float32)

    output爲一個list，裏面只有1個元素，該元素大小爲圖像大小，元素爲0-1的概率，表示是目標的概率。如下：

    [array([[ 9.99689460e-01, 3.10521980e-04],
    [ 9.99805272e-01, 1.94725304e-04],
    [ 9.99785841e-01, 2.14181622e-04],
    ...,
    [ 9.99480784e-01, 5.19228633e-04],
    [ 9.99274552e-01, 7.25465012e-04],
    [ 9.98537183e-01, 1.46284746e-03]], dtype=float32)]
  9. 將輸出reshape到圖像大小，output_image的shape爲(375, 1242)
    python
    # Reshape output from flat vector to 2D Image
    shape = image.shape
    output_image = output[0][:, 1].reshape(shape[0], shape[1])
  10. 將每個點的概率映射到原始圖像中。rb_image的shape爲(375, 1242, 3)
    python
    # Plot confidences as red-blue overlay
    rb_image = seg.make_overlay(image, output_image)

  11. 利用閾值分割圖像，用綠色標註。green_image的shape爲(375, 1242, 3)

    
# Accept all pixel with conf >= 0.5 as positive prediction


# This creates a `hard` prediction result for class street

threshold = 0.5
street_prediction = output_image > threshold


# Plot the hard prediction as green overlay

green_image = tv_utils.fast_overlay(image, street_prediction)

  12. 保存圖像。

    
# Save output images to disk.

if FLAGS.output_image is None:
    output_base_name = input_image
else:
    output_base_name = FLAGS.output_image

raw_image_name = output_base_name.split('.')[0] + '_raw.png'
rb_image_name = output_base_name.split('.')[0] + '_rb.png'
green_image_name = output_base_name.split('.')[0] + '_green.png'

scp.misc.imsave(raw_image_name, output_image)
scp.misc.imsave(rb_image_name, rb_image)
scp.misc.imsave(green_image_name, green_image)

Eval

python evaluate.py

日誌輸出：

2017-12-05 14:55:04,423 INFO No environment variable 'TV_PLUGIN_DIR' found. Set to 'C:\Users\10217814/tv-plugins'.
2017-12-05 14:55:04,424 INFO No environment variable 'TV_STEP_SHOW' found. Set to '50'.
2017-12-05 14:55:04,425 INFO No environment variable 'TV_STEP_EVAL' found. Set to '250'.
2017-12-05 14:55:04,426 INFO No environment variable 'TV_STEP_WRITE' found. Set to '1000'.
2017-12-05 14:55:04,427 INFO No environment variable 'TV_MAX_KEEP' found. Set to '10'.
2017-12-05 14:55:04,428 INFO No environment variable 'TV_STEP_STR' found. Set to 'Step {step}/{total_steps}: loss = {loss_value:.2f}; lr = {lr_value:.2e}; {sec_per_batch:.3f} sec (per Batch); {examples_per_sec:.1f} imgs/sec'.
2017-12-05 14:55:04,449 INFO f: <_io.TextIOWrapper name='hypes/KittiSeg.json' mode='r' encoding='cp936'>
2017-12-05 14:55:04,451 INFO Evaluating on Validation data.
2017-12-05 14:55:04,451 INFO f: <_io.TextIOWrapper name='RUNS\\KittiSeg_pretrained\\model_files\\hypes.json' mode='r' encoding='cp936'>
npy file loaded
Layer name: conv1_1
Layer shape: (3, 3, 3, 64)
2017-12-05 14:55:05,124 INFO Creating Summary for: conv1_1/filter
2017-12-05 14:55:05,150 INFO Creating Summary for: conv1_1/biases
Layer name: conv1_2
Layer shape: (3, 3, 64, 64)
2017-12-05 14:55:05,183 INFO Creating Summary for: conv1_2/filter
2017-12-05 14:55:05,207 INFO Creating Summary for: conv1_2/biases
Layer name: conv2_1
Layer shape: (3, 3, 64, 128)
2017-12-05 14:55:05,239 INFO Creating Summary for: conv2_1/filter
2017-12-05 14:55:05,264 INFO Creating Summary for: conv2_1/biases
Layer name: conv2_2
Layer shape: (3, 3, 128, 128)
2017-12-05 14:55:05,298 INFO Creating Summary for: conv2_2/filter
2017-12-05 14:55:05,324 INFO Creating Summary for: conv2_2/biases
Layer name: conv3_1
Layer shape: (3, 3, 128, 256)
2017-12-05 14:55:05,357 INFO Creating Summary for: conv3_1/filter
2017-12-05 14:55:05,383 INFO Creating Summary for: conv3_1/biases
Layer name: conv3_2
Layer shape: (3, 3, 256, 256)
2017-12-05 14:55:05,416 INFO Creating Summary for: conv3_2/filter
2017-12-05 14:55:05,441 INFO Creating Summary for: conv3_2/biases
Layer name: conv3_3
Layer shape: (3, 3, 256, 256)
2017-12-05 14:55:05,476 INFO Creating Summary for: conv3_3/filter
2017-12-05 14:55:05,502 INFO Creating Summary for: conv3_3/biases
Layer name: conv4_1
Layer shape: (3, 3, 256, 512)
2017-12-05 14:55:05,539 INFO Creating Summary for: conv4_1/filter
2017-12-05 14:55:05,562 INFO Creating Summary for: conv4_1/biases
Layer name: conv4_2
Layer shape: (3, 3, 512, 512)
2017-12-05 14:55:05,602 INFO Creating Summary for: conv4_2/filter
2017-12-05 14:55:05,627 INFO Creating Summary for: conv4_2/biases
Layer name: conv4_3
Layer shape: (3, 3, 512, 512)
2017-12-05 14:55:05,665 INFO Creating Summary for: conv4_3/filter
2017-12-05 14:55:05,690 INFO Creating Summary for: conv4_3/biases
Layer name: conv5_1
Layer shape: (3, 3, 512, 512)
2017-12-05 14:55:05,730 INFO Creating Summary for: conv5_1/filter
2017-12-05 14:55:05,756 INFO Creating Summary for: conv5_1/biases
Layer name: conv5_2
Layer shape: (3, 3, 512, 512)
2017-12-05 14:55:05,796 INFO Creating Summary for: conv5_2/filter
2017-12-05 14:55:05,821 INFO Creating Summary for: conv5_2/biases
Layer name: conv5_3
Layer shape: (3, 3, 512, 512)
2017-12-05 14:55:05,860 INFO Creating Summary for: conv5_3/filter
2017-12-05 14:55:05,885 INFO Creating Summary for: conv5_3/biases
Layer name: fc6
Layer shape: [7, 7, 512, 4096]
2017-12-05 14:55:06,204 INFO Creating Summary for: fc6/weights
2017-12-05 14:55:06,275 INFO Creating Summary for: fc6/biases
Layer name: fc7
Layer shape: [1, 1, 4096, 4096]
2017-12-05 14:55:06,365 INFO Creating Summary for: fc7/weights
2017-12-05 14:55:06,389 INFO Creating Summary for: fc7/biases
2017-12-05 14:55:06,423 INFO Creating Summary for: score_fr/weights
2017-12-05 14:55:06,447 INFO Creating Summary for: score_fr/biases
WARNING:tensorflow:From incl\tensorflow_fcn\fcn8_vgg.py:114: calling argmax (from tensorflow.python.ops.math_ops) with dimension is deprecated and will be removed in a future version.
Instructions for updating:
Use the `axis` argument instead
2017-12-05 14:55:06,647 WARNING From incl\tensorflow_fcn\fcn8_vgg.py:114: calling argmax (from tensorflow.python.ops.math_ops) with dimension is deprecated and will be removed in a future version.
Instructions for updating:
Use the `axis` argument instead
2017-12-05 14:55:06,671 INFO Creating Summary for: upscore2/up_filter
2017-12-05 14:55:06,710 INFO Creating Summary for: score_pool4/weights
2017-12-05 14:55:06,735 INFO Creating Summary for: score_pool4/biases
2017-12-05 14:55:06,779 INFO Creating Summary for: upscore4/up_filter
2017-12-05 14:55:06,817 INFO Creating Summary for: score_pool3/weights
2017-12-05 14:55:06,841 INFO Creating Summary for: score_pool3/biases
2017-12-05 14:55:06,885 INFO Creating Summary for: upscore32/up_filter
2017-12-05 14:55:06.951540: I C:\tf_jenkins\home\workspace\rel-win\M\windows\PY\35\tensorflow\core\platform\cpu_feature_guard.cc:137] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX AVX2
2017-12-05 14:55:07,076 INFO /u/marvin/no_backup/RUNS/KittiSeg/loss_bench/xentropy_kitti_fcn_2016_10_15_01.18/model.ckpt-15999
INFO:tensorflow:Restoring parameters from RUNS\KittiSeg_pretrained\model.ckpt-15999
2017-12-05 14:55:07,077 INFO Restoring parameters from RUNS\KittiSeg_pretrained\model.ckpt-15999
2017-12-05 14:55:10,334 INFO Graph loaded succesfully. Starting evaluation.
2017-12-05 14:55:10,334 INFO Output Images will be written to: RUNS\KittiSeg_pretrained\analyse\images/
2017-12-05 14:59:26,795 INFO Evaluation Succesfull. Results:
2017-12-05 14:59:26,795 INFO     MaxF1  :  96.0821
2017-12-05 14:59:26,796 INFO     BestThresh  :  14.5098
2017-12-05 14:59:26,796 INFO     Average Precision  :  92.3620
2017-12-05 14:59:26,796 INFO     Speed (msec)  :  4477.6770
2017-12-05 14:59:26,796 INFO     Speed (fps)  :  0.2233
2017-12-05 14:59:47,150 INFO Creating output on test data.
2017-12-05 14:59:47,151 INFO f: <_io.TextIOWrapper name='RUNS\\KittiSeg_pretrained\\model_files\\hypes.json' mode='r' encoding='cp936'>
npy file loaded
Layer name: conv1_1
Layer shape: (3, 3, 3, 64)
2017-12-05 14:59:47,800 INFO Creating Summary for: conv1_1/filter
2017-12-05 14:59:47,824 INFO Creating Summary for: conv1_1/biases
Layer name: conv1_2
Layer shape: (3, 3, 64, 64)
2017-12-05 14:59:47,855 INFO Creating Summary for: conv1_2/filter
2017-12-05 14:59:47,877 INFO Creating Summary for: conv1_2/biases
Layer name: conv2_1
Layer shape: (3, 3, 64, 128)
2017-12-05 14:59:47,910 INFO Creating Summary for: conv2_1/filter
2017-12-05 14:59:47,933 INFO Creating Summary for: conv2_1/biases
Layer name: conv2_2
Layer shape: (3, 3, 128, 128)
2017-12-05 14:59:47,964 INFO Creating Summary for: conv2_2/filter
2017-12-05 14:59:47,987 INFO Creating Summary for: conv2_2/biases
Layer name: conv3_1
Layer shape: (3, 3, 128, 256)
2017-12-05 14:59:48,022 INFO Creating Summary for: conv3_1/filter
2017-12-05 14:59:48,044 INFO Creating Summary for: conv3_1/biases
Layer name: conv3_2
Layer shape: (3, 3, 256, 256)
2017-12-05 14:59:48,078 INFO Creating Summary for: conv3_2/filter
2017-12-05 14:59:48,101 INFO Creating Summary for: conv3_2/biases
Layer name: conv3_3
Layer shape: (3, 3, 256, 256)
2017-12-05 14:59:48,132 INFO Creating Summary for: conv3_3/filter
2017-12-05 14:59:48,155 INFO Creating Summary for: conv3_3/biases
Layer name: conv4_1
Layer shape: (3, 3, 256, 512)
2017-12-05 14:59:48,189 INFO Creating Summary for: conv4_1/filter
2017-12-05 14:59:48,213 INFO Creating Summary for: conv4_1/biases
Layer name: conv4_2
Layer shape: (3, 3, 512, 512)
2017-12-05 14:59:48,251 INFO Creating Summary for: conv4_2/filter
2017-12-05 14:59:48,274 INFO Creating Summary for: conv4_2/biases
Layer name: conv4_3
Layer shape: (3, 3, 512, 512)
2017-12-05 14:59:48,313 INFO Creating Summary for: conv4_3/filter
2017-12-05 14:59:48,336 INFO Creating Summary for: conv4_3/biases
Layer name: conv5_1
Layer shape: (3, 3, 512, 512)
2017-12-05 14:59:48,375 INFO Creating Summary for: conv5_1/filter
2017-12-05 14:59:48,398 INFO Creating Summary for: conv5_1/biases
Layer name: conv5_2
Layer shape: (3, 3, 512, 512)
2017-12-05 14:59:48,542 INFO Creating Summary for: conv5_2/filter
2017-12-05 14:59:48,564 INFO Creating Summary for: conv5_2/biases
Layer name: conv5_3
Layer shape: (3, 3, 512, 512)
2017-12-05 14:59:48,602 INFO Creating Summary for: conv5_3/filter
2017-12-05 14:59:48,624 INFO Creating Summary for: conv5_3/biases
Layer name: fc6
Layer shape: [7, 7, 512, 4096]
2017-12-05 14:59:48,927 INFO Creating Summary for: fc6/weights
2017-12-05 14:59:48,952 INFO Creating Summary for: fc6/biases
Layer name: fc7
Layer shape: [1, 1, 4096, 4096]
2017-12-05 14:59:49,027 INFO Creating Summary for: fc7/weights
2017-12-05 14:59:49,050 INFO Creating Summary for: fc7/biases
2017-12-05 14:59:49,082 INFO Creating Summary for: score_fr/weights
2017-12-05 14:59:49,106 INFO Creating Summary for: score_fr/biases
2017-12-05 14:59:49,150 INFO Creating Summary for: upscore2/up_filter
2017-12-05 14:59:49,184 INFO Creating Summary for: score_pool4/weights
2017-12-05 14:59:49,207 INFO Creating Summary for: score_pool4/biases
2017-12-05 14:59:49,249 INFO Creating Summary for: upscore4/up_filter
2017-12-05 14:59:49,284 INFO Creating Summary for: score_pool3/weights
2017-12-05 14:59:49,307 INFO Creating Summary for: score_pool3/biases
2017-12-05 14:59:49,350 INFO Creating Summary for: upscore32/up_filter
2017-12-05 14:59:49,544 INFO /u/marvin/no_backup/RUNS/KittiSeg/loss_bench/xentropy_kitti_fcn_2016_10_15_01.18/model.ckpt-15999
INFO:tensorflow:Restoring parameters from RUNS\KittiSeg_pretrained\model.ckpt-15999
2017-12-05 14:59:49,544 INFO Restoring parameters from RUNS\KittiSeg_pretrained\model.ckpt-15999
2017-12-05 14:59:52,804 INFO Images will be written to test_images//test_images_{green, rg}
2017-12-05 14:59:57,461 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000000.png
2017-12-05 15:00:02,348 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000001.png
2017-12-05 15:00:07,223 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000002.png
2017-12-05 15:00:12,061 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000003.png
2017-12-05 15:00:16,949 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000004.png
2017-12-05 15:00:21,853 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000005.png
2017-12-05 15:00:26,717 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000006.png
2017-12-05 15:00:31,517 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000007.png
2017-12-05 15:00:36,363 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000008.png
2017-12-05 15:00:41,237 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000009.png
2017-12-05 15:00:46,344 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000010.png
2017-12-05 15:00:51,325 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000011.png
2017-12-05 15:00:56,322 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000012.png
2017-12-05 15:01:01,255 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000013.png
2017-12-05 15:01:06,227 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000014.png
2017-12-05 15:01:11,209 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000015.png
2017-12-05 15:01:16,227 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000016.png
2017-12-05 15:01:21,133 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000017.png
2017-12-05 15:01:26,015 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000018.png
2017-12-05 15:01:30,831 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000019.png
2017-12-05 15:01:35,669 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000020.png
2017-12-05 15:01:40,528 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000021.png
2017-12-05 15:01:45,387 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000022.png
2017-12-05 15:01:50,216 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000023.png
2017-12-05 15:01:55,223 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000024.png
2017-12-05 15:02:00,105 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000025.png
2017-12-05 15:02:05,010 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000026.png
2017-12-05 15:02:09,813 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000027.png
2017-12-05 15:02:14,699 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000028.png
2017-12-05 15:02:19,589 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000029.png
2017-12-05 15:02:24,495 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000030.png
2017-12-05 15:02:29,399 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000031.png
2017-12-05 15:02:34,284 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000032.png
2017-12-05 15:02:39,152 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000033.png
2017-12-05 15:02:43,985 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000034.png
2017-12-05 15:02:48,823 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000035.png
2017-12-05 15:02:53,796 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000036.png
2017-12-05 15:02:58,849 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000037.png
2017-12-05 15:03:03,744 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000038.png
2017-12-05 15:03:08,647 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000039.png
2017-12-05 15:03:13,522 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000040.png
2017-12-05 15:03:18,384 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000041.png
2017-12-05 15:03:23,182 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000042.png
2017-12-05 15:03:28,069 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000043.png
2017-12-05 15:03:32,960 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000044.png
2017-12-05 15:03:37,831 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000045.png
2017-12-05 15:03:42,714 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000046.png
2017-12-05 15:03:47,612 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000047.png
2017-12-05 15:03:52,527 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000048.png
2017-12-05 15:03:57,440 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000049.png
2017-12-05 15:04:02,381 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000050.png
2017-12-05 15:04:07,287 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000051.png
2017-12-05 15:04:12,161 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000052.png
2017-12-05 15:04:17,087 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000053.png
2017-12-05 15:04:21,988 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000054.png
2017-12-05 15:04:26,960 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000055.png
2017-12-05 15:04:31,889 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000056.png
2017-12-05 15:04:36,739 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000057.png
2017-12-05 15:04:41,628 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000058.png
2017-12-05 15:04:46,555 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000059.png
2017-12-05 15:04:51,468 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000060.png
2017-12-05 15:04:56,434 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000061.png
2017-12-05 15:05:01,358 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000062.png
2017-12-05 15:05:06,235 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000063.png
2017-12-05 15:05:11,182 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000064.png
2017-12-05 15:05:16,128 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000065.png
2017-12-05 15:05:21,090 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000066.png
2017-12-05 15:05:26,048 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000067.png
2017-12-05 15:05:31,033 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000068.png
2017-12-05 15:05:35,943 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000069.png
2017-12-05 15:05:40,808 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000070.png
2017-12-05 15:05:45,679 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000071.png
2017-12-05 15:05:50,457 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000072.png
2017-12-05 15:05:55,265 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000073.png
2017-12-05 15:06:00,046 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000074.png
2017-12-05 15:06:04,821 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000075.png
2017-12-05 15:06:09,609 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000076.png
2017-12-05 15:06:14,362 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000077.png
2017-12-05 15:06:19,160 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000078.png
2017-12-05 15:06:23,969 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000079.png
2017-12-05 15:06:28,786 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000080.png
2017-12-05 15:06:33,557 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000081.png
2017-12-05 15:06:38,215 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000082.png
2017-12-05 15:06:42,989 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000083.png
2017-12-05 15:06:47,764 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000084.png
2017-12-05 15:06:52,520 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000085.png
2017-12-05 15:06:57,252 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000086.png
2017-12-05 15:07:02,030 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000087.png
2017-12-05 15:07:06,759 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000088.png
2017-12-05 15:07:11,461 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000089.png
2017-12-05 15:07:16,125 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000090.png
2017-12-05 15:07:20,847 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000091.png
2017-12-05 15:07:25,597 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000092.png
2017-12-05 15:07:30,367 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000093.png
2017-12-05 15:07:35,041 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000094.png
2017-12-05 15:07:39,878 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/um_road_000095.png
2017-12-05 15:07:44,772 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000000.png
2017-12-05 15:07:49,713 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000001.png
2017-12-05 15:07:54,652 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000002.png
2017-12-05 15:07:59,567 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000003.png
2017-12-05 15:08:04,506 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000004.png
2017-12-05 15:08:09,394 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000005.png
2017-12-05 15:08:14,264 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000006.png
2017-12-05 15:08:19,172 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000007.png
2017-12-05 15:08:24,038 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000008.png
2017-12-05 15:08:28,876 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000009.png
2017-12-05 15:08:33,741 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000010.png
2017-12-05 15:08:38,566 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000011.png
2017-12-05 15:08:43,371 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000012.png
2017-12-05 15:08:48,278 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000013.png
2017-12-05 15:08:53,173 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000014.png
2017-12-05 15:08:58,058 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000015.png
2017-12-05 15:09:02,945 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000016.png
2017-12-05 15:09:07,823 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000017.png
2017-12-05 15:09:12,706 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000018.png
2017-12-05 15:09:17,606 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000019.png
2017-12-05 15:09:22,511 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000020.png
2017-12-05 15:09:27,425 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000021.png
2017-12-05 15:09:32,353 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000022.png
2017-12-05 15:09:37,271 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000023.png
2017-12-05 15:09:42,127 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000024.png
2017-12-05 15:09:47,000 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000025.png
2017-12-05 15:09:51,880 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000026.png
2017-12-05 15:09:56,769 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000027.png
2017-12-05 15:10:01,659 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000028.png
2017-12-05 15:10:06,557 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000029.png
2017-12-05 15:10:11,478 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000030.png
2017-12-05 15:10:16,426 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000031.png
2017-12-05 15:10:21,363 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000032.png
2017-12-05 15:10:26,311 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000033.png
2017-12-05 15:10:31,238 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000034.png
2017-12-05 15:10:36,185 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000035.png
2017-12-05 15:10:41,134 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000036.png
2017-12-05 15:10:46,070 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000037.png
2017-12-05 15:10:51,042 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000038.png
2017-12-05 15:10:55,956 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000039.png
2017-12-05 15:11:00,887 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000040.png
2017-12-05 15:11:05,879 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000041.png
2017-12-05 15:11:10,948 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000042.png
2017-12-05 15:11:16,079 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000043.png
2017-12-05 15:11:21,366 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000044.png
2017-12-05 15:11:26,465 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000045.png
2017-12-05 15:11:31,596 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000046.png
2017-12-05 15:11:36,729 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000047.png
2017-12-05 15:11:41,803 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000048.png
2017-12-05 15:11:46,803 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000049.png
2017-12-05 15:11:51,918 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000050.png
2017-12-05 15:11:57,060 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000051.png
2017-12-05 15:12:02,089 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000052.png
2017-12-05 15:12:07,133 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000053.png
2017-12-05 15:12:12,138 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000054.png
2017-12-05 15:12:17,101 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000055.png
2017-12-05 15:12:22,004 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000056.png
2017-12-05 15:12:27,061 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000057.png
2017-12-05 15:12:32,084 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000058.png
2017-12-05 15:12:37,078 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000059.png
2017-12-05 15:12:42,055 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000060.png
2017-12-05 15:12:47,004 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000061.png
2017-12-05 15:12:51,926 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000062.png
2017-12-05 15:12:56,927 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000063.png
2017-12-05 15:13:01,903 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000064.png
2017-12-05 15:13:06,917 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000065.png
2017-12-05 15:13:11,883 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000066.png
2017-12-05 15:13:16,850 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000067.png
2017-12-05 15:13:21,859 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000068.png
2017-12-05 15:13:26,780 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000069.png
2017-12-05 15:13:31,801 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000070.png
2017-12-05 15:13:36,831 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000071.png
2017-12-05 15:13:41,825 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000072.png
2017-12-05 15:13:46,840 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000073.png
2017-12-05 15:13:51,825 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000074.png
2017-12-05 15:13:56,806 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000075.png
2017-12-05 15:14:01,863 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000076.png
2017-12-05 15:14:06,955 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000077.png
2017-12-05 15:14:12,017 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000078.png
2017-12-05 15:14:17,187 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000079.png
2017-12-05 15:14:22,356 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000080.png
2017-12-05 15:14:27,529 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000081.png
2017-12-05 15:14:32,744 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000082.png
2017-12-05 15:14:37,933 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000083.png
2017-12-05 15:14:43,074 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000084.png
2017-12-05 15:14:48,216 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000085.png
2017-12-05 15:14:53,365 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000086.png
2017-12-05 15:14:58,503 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000087.png
2017-12-05 15:15:03,648 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000088.png
2017-12-05 15:15:08,813 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000089.png
2017-12-05 15:15:13,960 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000090.png
2017-12-05 15:15:18,972 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000091.png
2017-12-05 15:15:24,061 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000092.png
2017-12-05 15:15:29,157 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/umm_road_000093.png
2017-12-05 15:15:34,265 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000000.png
2017-12-05 15:15:39,228 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000001.png
2017-12-05 15:15:44,272 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000002.png
2017-12-05 15:15:49,329 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000003.png
2017-12-05 15:15:54,296 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000004.png
2017-12-05 15:15:59,296 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000005.png
2017-12-05 15:16:04,332 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000006.png
2017-12-05 15:16:09,360 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000007.png
2017-12-05 15:16:14,318 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000008.png
2017-12-05 15:16:19,305 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000009.png
2017-12-05 15:16:24,339 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000010.png
2017-12-05 15:16:29,424 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000011.png
2017-12-05 15:16:34,434 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000012.png
2017-12-05 15:16:39,510 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000013.png
2017-12-05 15:16:44,500 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000014.png
2017-12-05 15:16:49,497 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000015.png
2017-12-05 15:16:54,575 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000016.png
2017-12-05 15:16:59,563 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000017.png
2017-12-05 15:17:04,646 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000018.png
2017-12-05 15:17:09,872 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000019.png
2017-12-05 15:17:14,897 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000020.png
2017-12-05 15:17:19,902 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000021.png
2017-12-05 15:17:24,923 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000022.png
2017-12-05 15:17:29,986 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000023.png
2017-12-05 15:17:35,160 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000024.png
2017-12-05 15:17:40,244 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000025.png
2017-12-05 15:17:45,399 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000026.png
2017-12-05 15:17:50,633 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000027.png
2017-12-05 15:17:55,675 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000028.png
2017-12-05 15:18:00,674 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000029.png
2017-12-05 15:18:05,722 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000030.png
2017-12-05 15:18:10,824 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000031.png
2017-12-05 15:18:15,823 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000032.png
2017-12-05 15:18:20,810 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000033.png
2017-12-05 15:18:25,872 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000034.png
2017-12-05 15:18:30,990 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000035.png
2017-12-05 15:18:36,192 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000036.png
2017-12-05 15:18:41,195 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000037.png
2017-12-05 15:18:46,205 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000038.png
2017-12-05 15:18:51,157 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000039.png
2017-12-05 15:18:56,218 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000040.png
2017-12-05 15:19:01,204 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000041.png
2017-12-05 15:19:06,138 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000042.png
2017-12-05 15:19:11,079 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000043.png
2017-12-05 15:19:16,025 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000044.png
2017-12-05 15:19:21,059 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000045.png
2017-12-05 15:19:26,124 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000046.png
2017-12-05 15:19:31,152 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000047.png
2017-12-05 15:19:36,268 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000048.png
2017-12-05 15:19:41,319 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000049.png
2017-12-05 15:19:46,395 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000050.png
2017-12-05 15:19:51,521 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000051.png
2017-12-05 15:19:56,704 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000052.png
2017-12-05 15:20:01,742 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000053.png
2017-12-05 15:20:06,779 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000054.png
2017-12-05 15:20:11,909 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000055.png
2017-12-05 15:20:16,943 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000056.png
2017-12-05 15:20:21,960 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000057.png
2017-12-05 15:20:26,942 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000058.png
2017-12-05 15:20:31,969 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000059.png
2017-12-05 15:20:37,024 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000060.png
2017-12-05 15:20:42,045 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000061.png
2017-12-05 15:20:47,163 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000062.png
2017-12-05 15:20:52,581 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000063.png
2017-12-05 15:20:57,870 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000064.png
2017-12-05 15:21:03,084 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000065.png
2017-12-05 15:21:08,138 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000066.png
2017-12-05 15:21:13,257 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000067.png
2017-12-05 15:21:18,286 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000068.png
2017-12-05 15:21:23,297 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000069.png
2017-12-05 15:21:28,458 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000070.png
2017-12-05 15:21:33,523 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000071.png
2017-12-05 15:21:38,423 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000072.png
2017-12-05 15:21:43,471 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000073.png
2017-12-05 15:21:48,509 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000074.png
2017-12-05 15:21:53,635 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000075.png
2017-12-05 15:21:58,703 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000076.png
2017-12-05 15:22:03,674 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000077.png
2017-12-05 15:22:08,672 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000078.png
2017-12-05 15:22:13,776 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000079.png
2017-12-05 15:22:18,827 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000080.png
2017-12-05 15:22:23,891 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000081.png
2017-12-05 15:22:28,967 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000082.png
2017-12-05 15:22:34,022 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000083.png
2017-12-05 15:22:39,076 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000084.png
2017-12-05 15:22:44,178 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000085.png
2017-12-05 15:22:49,210 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000086.png
2017-12-05 15:22:54,305 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000087.png
2017-12-05 15:22:59,279 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000088.png
2017-12-05 15:23:04,186 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000089.png
2017-12-05 15:23:09,103 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000090.png
2017-12-05 15:23:14,044 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000091.png
2017-12-05 15:23:19,067 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000092.png
2017-12-05 15:23:24,212 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000093.png
2017-12-05 15:23:29,356 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000094.png
2017-12-05 15:23:34,412 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000095.png
2017-12-05 15:23:39,554 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000096.png
2017-12-05 15:23:44,733 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000097.png
2017-12-05 15:23:49,775 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000098.png
2017-12-05 15:23:54,883 INFO Writing file: E:\GitHub\KittiSeg\RUNS\KittiSeg_pretrained\test_images/uu_road_000099.png
2017-12-05 15:23:55,479 INFO Analysis for pretrained model complete.
2017-12-05 15:23:55,479 INFO For evaluating your own models I recommend using:`tv-analyze --logdir /path/to/run`.
2017-12-05 15:23:55,480 INFO tv-analysis has a much cleaner interface.
QiangLi_strong
發佈了45 篇原創文章 · 獲贊 61 · 訪問量 7萬+
私信 關注
發表評論
所有評論
還沒有人評論，想成為第一個評論的人麼? 請在上方評論欄輸入並且點擊發布.
相關文章

[百度飛槳PaddlePaddle]Python“小白”逆襲大神--訓練營結營心得

學習背景 我是一個可以說混了4年的 phper，高不成低不就的。每天寫業務邏輯，改 bug，日復一日。 爲啥一個 phper 幹了那麼久現在才學 python 呢？是否不務正業什麼的？其實我在17 ~ 18年就開始接觸 pytho

ming明如月 2020-07-06 16:27:04

從RCNN到Faster RCNN

R-CNN (Region-based CNN features) 性能：RCNN在VOC2007上的mAP是58%左右。 簡述：R-CNN要完成目標定位，其流程主要分爲四步： 1.輸入圖像 2.利用選擇性搜索(Selectiv

Elag 2020-07-05 10:51:31

exponentially weighted averages

定義 一個變量的當前值和以前所有值的加權平均 公式 t=0t=0t=0時，Vt=0V_{t} = 0Vt​=0 t>0t>0t>0時，Vt=βVt−1+(1−β)θtV_{t} = βV_{t-1} + (1-β)θ_{t}Vt​

Dunker23 2020-07-04 05:20:50

train/dev/test set 數據分佈不均勻

構建數據集方法 使dev/test set 和真實環境的數據同分布 構建train-dev set：跟train set同分布，但是不用來訓練 若所有數據混合打亂後隨即抽取train/dev/test set，有可能造成dev

Dunker23 2020-07-04 05:20:50

Step by Step

工具安裝（以下皆在Windows下） Miniconda 或 Anaconda 根據操作系統安裝好以後，測試是否正常安裝：conda --version 若出現 conda不是內部或者外部命令 ，那麼就是環境變量還沒有配置，需

DaxinPai 2020-07-04 01:43:00

colab使用方式

Google Colab——用谷歌免費GPU跑你的深度學習代碼

喵咖 2020-07-03 08:30:50

注意Pytorch中GPU與CPU的使用

其實在Pytorch中並不是在任何情況下把數據放到GPU下就能訓練效率，減少訓練時間。 下面舉個例子： import torch import time print(torch.cuda.is_available()) print(

Robert--cao 2020-06-30 20:14:52

win10+pytorch1.0+cuda9

原本安裝pytorch只需要一條命令： conda install pytorch torchvision cudatoolkit=9.0 -c pytorch 結果不是網不行就是各種出錯。。。   於是用pip安裝試了一下，發現還是p

遂言 2020-06-27 09:21:45

vs2015+OpenVINO報錯解決:MKLDNNPlugind.dll,MKLDNNPlugind.dll

使用OpenVINO時，報出以下錯誤： Failed to create plugin xxx\intel64\Debug\MKLDNNPlugind.dll for device CPU Please, check your envir

岁月神偷小拳拳 2020-06-27 05:40:00

win10+vs2015+OpenVINO安裝使用

OpenVINO是英特爾基於自身現有的硬件平臺開發的一種可以加快高性能計算機視覺和深度學習視覺應用開發速度工具套件，支持各種英特爾平臺的硬件加速器上進行深度學習，並且允許直接異構執行。 支持在Windows與Linux系統，Python/

岁月神偷小拳拳 2020-06-27 05:39:48

pytorch安裝詳細步驟

文章目錄環境配置步驟官網選擇型號清華源快速安裝安裝完檢驗pycharm設置項目環境 環境配置步驟 如果顯卡支持GPU，在命令行處，輸入nvidia-smi ，查看自己驅動的版本。 參考文章 官網選擇型號 官網 清華源快速安裝

spatial_coder 2020-06-24 21:57:59

GoogLeNet Inception V1：Going deeper with convolutions論文筆記

論文地址：Going deeper with convolutions 前言 在ILSVRC2014上最耀眼的除了VGG，那就非GoogLeNet莫屬了，它與VGG類似的地方就是，關注於模型的深度，使模型傾向於deeper，與VG

小时候贼聪明 2020-06-24 10:05:00

Deep learning：十五(Self-Taught Learning練習)

本文轉自http://www.cnblogs.com/tornadomeet/archive/2013/03/24/2979408.html 前言： 　　本次實驗主要是練習soft- taught learning的實現。參考的資料

siliang13 2020-06-24 07:32:20

google bert模型詳解 源碼解析

Table of Contents 模型簡介 1. A High-Level Look 2. Encoder(http://jalammar.github.io/illustrated-transformer/) 3. Self-Atte

xxl98330 2020-06-23 14:17:13

ubuntu16.04+GTX1660深度學習相關配置

1.前情提要 本文使用cuda10.1，cuda8不行，cuda9沒試過 cudnn要使用顯卡以後發佈的版本，GTX1660發佈於20190222 考慮到官網總是掛，可以在此下載cuda10.1，cudnn cudnn的deve

random_repick 2020-06-21 10:27:25