-
pytorch训练过程中，会在utils.py里的encode()函数卡住，不停迭代
def encode(self, text, depth=0):
"""Support batch or single str."""
if isinstance(text, str):
text = [self.dict[char.lowe…
-
I want to train with my own dataset. First I used the create_dataset.py and it created two files- data.mdb and lock.mdb.
Now i gave the same path of data.mdb to train and validate. If this is not rig…
-
-
I am training a CRNN model in pytorch
max_seq_length=99
number_of_alphabets=96
batch_size=16
output=CRNN(image)
what should be the expected shape of output?
Secondly, should we apply softmax in …
-
- OS Platform and Distribution: windows 10 64bit
- ONNX Runtime installed from (source or binary): binary
- Python version: 0.4.0
- Only CPU
I export my model as onnx with pytorch, and try to lo…
-
Extract the Chinese_dataset.rar to data folder and move all pictures to images, then modify data_test.txt to data_dev.txt.
run main.py, it shows that:
>> Creating directory if it does not exist:…
-
你好，我今天刚刚clone了这个repo，在进行测试的时候发现test.py中的crnn_Rec_done_1.pth并没有包含在repo内，trained_model文件夹内依然是10个月前的mixed_second_finetune_acc97p7.pth。
请问这个模型是最初的训练结果么？因为我在使用该模型进行测试并不能重现出博文中的结果，如图所示：
:
File "crnn_main.py", line 200, in
training()
File "crnn_main.py", line 117, in training
val(crnn, test_dataset, criterion)
File "crnn_…
-
32核服务器
num_workers为2的时候，cpu占用率1600%，100batch平均21s
num_workers为0的时候，cpu占用率1300%，100batch平均29s
你们的也是如此吗？
-
作者你好，请问一下我在训练的时候训练一开始loss为负值，训练一段时间后loss反而变大了，但是绝对值一直是小于1的，调整了很多次学习率都是类似的情况，有人遇到过吗
下面是训练开始的部分
[0/10][5/204976] Loss: -0.127944
[0/10][10/204976] Loss: 0.175103
[0/10][15/204976] Loss: 0.192796
[0…