Comments (4)
I think you can test whether the object can be detected by using the weight of 120000 rounds of training. The loss in my training is also around 2, and if you want to obtain lower loss, you can consider reducing the learning rate.
from ssd_pytorch.
how to test not a single image ,and how to eval this train and test result,such as AP and total recall?can you help me?
from ssd_pytorch.
I'm sorry that this version doesn't provide the method to eval train and test result. I will complete this function in my spare time.
from ssd_pytorch.
thank you for you answer,when use the ssd-pytorch-master on https://github.com/amdegroot/ssd.pytorch, there is a error in training
iter 900 || Loss: 6.6272 || timer: 0.1010 sec.
iter 910 || Loss: 7.0335 || timer: 0.1023 sec.
iter 920 || Loss: 6.6000 || timer: 0.1001 sec.
iter 930 || Loss: 6.7137 || timer: 0.1013 sec.
iter 940 || Loss: 6.9450 || timer: 0.1027 sec.
iter 950 || Loss: 6.5815 || timer: 0.1038 sec.
iter 960 || Loss: 6.8804 || timer: 0.1021 sec.
iter 970 || Loss: 6.6749 || timer: 0.1279 sec.
iter 980 || Loss: 6.4802 || timer: 0.1018 sec.
iter 990 || Loss: 6.1978 || timer: 0.1184 sec.
iter 1000 || Loss: 6.7934 || timer: 0.1019 sec.
iter 1010 || Loss: 6.5664 || timer: 0.1028 sec.
iter 1020 || Loss: 6.6167 || timer: 0.0977 sec.
iter 1030 || Loss: 6.3809 || Traceback (most recent call last):
File "train.py", line 261, in
train()
File "train.py", line 166, in train
images, targets = next(batch_iterator)
File "/home/xlm/anaconda3/envs/M2Det/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 627, in next
raise StopIteration
StopIteration
when i follow the issues in https://github.com/amdegroot/ssd.pytorch
change
images, targets = next(batch_iterator)
with
try:
images,targets = next(batch_iterator)
except StopIteration:
bath_interator = iter(data_loader)
images,targets = next(batch_iterator)
but this error was still occured.
can you help me to address this problem
from ssd_pytorch.
Related Issues (5)
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 ssd_pytorch.