Comments (11)
In fact, it works in my environment ... so have you modified the code or made some big changes?
No I have not made any change
from fudanocr.
Please see Readme.md
CUDA_VISIBLE_DEVICES=GPU_NUM python main.py --batch_size=16 --STN --exp_name EXP_NAME --text_focus 【--resume YOUR_MODEL】 --test --test_data_dir ./dataset/mydata/test
or you can fill the absolute address of the pre-trained model in the argpaser of main.py
Hope this advice will help you out.
from fudanocr.
My epoch got interrupted and i want to resume it from the last interrupted so that I don't have to start my epoch from the starting (that will take a lot of my time) . Please help me to resume that . I tried but always cleans the old logs and start from the starting epochs
from fudanocr.
Hello, if you want to resume the checkpoint, you could [modify the name] of the experiment to solve this problem.
from fudanocr.
!python main.py --batch_size=16 --STN --exp_name f2 --text_focus --resume /content/drive/MyDrive/scene-text-telescope/checkpoint/f1/checkpoint.pth
i have used this command (changed the name ) still the training is starting from the first epochs
from fudanocr.
It is quite weird... So have you tried to directly modify the resume path in main.py?
from fudanocr.
Yes, I have modified the main.py, still not working. The epochs always start from the starting I have tried all ways. Please help me because I am not able to train the model for large epochs.
from fudanocr.
In fact, it works in my environment ...
so have you modified the code or made some big changes?
from fudanocr.
my epochs got interrupted here then I tried to resume it from this point but it starts from the starting like this .
Tried everything you told still not working.
can you share your environment so that I can try in that environment
from fudanocr.
i got the same problem.do you solve it? im so disgusting
from fudanocr.
Hello, could you tell me how to solve this problem? I have also met the same situation. Resume doesn't work at all
from fudanocr.
Related Issues (20)
- Question about CLIP-like pre-training in image-ids-CTR HOT 1
- train.py在计算loss有个小错误 HOT 1
- 关于CCR-CLIP HOT 1
- 数据集请求 HOT 2
- 请问text-focused-Transformers model和datasets如何下载 HOT 1
- VCTR dataset read error
- 请问在竖直文本识别那篇论文中,有没有提供预训练模型呢 HOT 2
- How to download VCTR dataset? HOT 2
- How to extract only image after super resolution?
- CCR-CLIP pretraining HOT 2
- About CCR-CLIP pretrained HOT 1
- inference of orientation-independent-CTR ?
- CCR-CLIP使用印刷字体进行增强 HOT 1
- Fix link for Baidu download in text-focused-Transformers/
- train.py的坑:容易误删已训练模型 HOT 2
- lmdb can not find my file. HOT 2
- 关于Chinese-CLIP复现中的几点疑惑,期待答复!
- Question about CCR-CLIP experiment and code HOT 4
- 字典里面的字符是不是不够啊?除了没有逗号,有时候会出现字典里没有这个汉字。 HOT 4
- 是否有HWDB 1.0-1.2 的lmdb数据集 HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from fudanocr.