YOLO2TensorFlow

YOLO2TensorFlow copied to clipboard

could you list the finished function of your code?

4

I help to know whether your code can training successfully and the accuracy, thank you. and how to mail you?

TaihuLight

我看这个网络直接输出分类、框坐标、置信度，并没有混合分类和检测数据集，对分类和检测任务进行联合训练

1

YOLO2提出一种联合训练机制，混合来自检测和分类数据集的图像进行训练。当网络看到标记为检测的图像时，基于完整的yolov2损失函数进行反向传播。当它看到一个分类图像时，只从特定于分类的部分反向传播损失。

jiashihuigithub

object_mask = tf.reduce_sum(gbboxes_batch, 4) object_mask = tf.cast(object_mask > 0, tf.float32) i do not know what is the object_mask what the purpose of the two operations. what's more, the shape of...

Letdreamfly

I had used 'voc_to_tfrecords.py' to generate TFRecords files. When I ran train.py, program would execute to `final_loss = slim.learning.train(train_op, logdir=FLAGS.train_dir, summary_op=summary_op, global_step=global_step, number_of_steps=FLAGS.max_number_of_steps, log_every_n_steps=FLAGS.log_every_n_steps, save_summaries_secs=FLAGS.save_summaries_secs, save_interval_secs=FLAGS.save_interval_secs, session_config=sess_config)` but there was...

scpaliuhc

linydf

Hi! Thank you for your code. And I want to ask you somthing about the Anchors Size.How dou you select the Anchor Size?From you code, the Anchors are [[1, 2],...

ZhongyuanW

The index sequence in conv_width_index and conv_height_index maybe wrong?

ZHANGKEON

preprocess_data with tensorflow is slow

1

def test_tf(): with tf.Session() as sess: array=tf.ones([1024,5],dtype=tf.float32) t0=time.clock() out=0 for i in range(array.shape[0]): out+=array[i] out=sess.run([out]) t1=time.clock() print("test_tf:",out,t1-t0) def test_np(): array=np.ones((1024,5),dtype=np.float32) print array.shape t0=time.clock() out=0 for i in range(array.shape[0]): out+=array[i] t1=time.clock()...

adeagle

Question on the Loss function

2

Hi, thanks for your very useful code! I want to ask you something about the loss function, because I've seen it implemented in many ways but this seems the best...

simo23