2017TJMtaotao
2017TJMtaotao
修改通道数
144/144 [==============================] - 47s 324ms/step - loss: 73.5387 - val_loss: 154.7555 Epoch 00079: early stopping
2 'targets' to have scalar type Int; but got torch.cuda.FloatTensor instead (while checking arguments for ctc_loss_gpu)
多少 ------------------ 原始邮件 ------------------ 发件人: "notifications"; 发送时间: 2019年6月19日(星期三) 晚上9:52 收件人: "eragonruan/text-detection-ctpn"; 抄送: "Subscribed"; 主题: Re: [eragonruan/text-detection-ctpn] 测试结果不是很准确漏字，有些地方检测出框 (#300) 楼上大佬好，是否愿意帮我们做个文本检测项目哈：）有偿 刚好做了一个这个方面的项目，感兴趣可以加我微信。davidcommander — You are receiving this because you are subscribed to...
作者复现时进行了改进感觉是把特征图尺寸设置为14x14
> I don't understand why grid_num is 14 in code , and why is it divided by 14 in yoloLoss.py line 88? 是由特征图限制的尺寸
[[Node: InceptionV3/Logits/Conv2d_1c_1x1/weights/RMSProp_1 = VariableV2[_class=["loc:@InceptionV3/Logits/Conv2d_1c_1x1/weights"], containe r="", dtype=DT_FLOAT, shape=[1,1,2048,6], shared_name="", _device="/device:GPU:0"]()]]
怎么解决
成功解决 config = tf.ConfigProto(allow_soft_placement=True) slim.learning.train( train_tensor, logdir=FLAGS.train_dir, master=FLAGS.master, is_chief=(FLAGS.task == 0), init_fn=_get_init_fn(), summary_op=summary_op, 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, sync_optimizer=optimizer if FLAGS.sync_replicas else None , session_config= config)