Comments (1)
The loss starts low because of the warmup stage, which is defaulted to 100. It's advisable to leave it like that as warmup training is very important. Instead, you can start training the second time with the "pretrained_weights" in the config.json file set to "best_weights.h5". It will start from this weights.
It's not a good idea to use the train set as the test set. I think you should leave the test set empty. The code in train.py will automatically split the given train set into the train set and test set using the ratio 0.8 if it cannot find the test set.
from keras-yolo2.
Related Issues (20)
- can't find RESNET50_BACKEND_PATH
- how to claculate MAP?
- Inaccurate Result and stepsize
- issue with model layer lambda, HOT 1
- issue with warmup_batches
- perlex with coord mask & conf_mask HOT 1
- why still need to use conf_mask but not y_true[...,4] HOT 6
- Doesn't work anymore
- You are trying to load a weight file containing 17 layers into a model with 2 layers
- Raccoon Backend / Full Model Links Point to Same Place HOT 1
- script that convert coco format to VOC format
- Issue in running the BloodCell Detection.ipynb
- RuntimeError: The layer has never been called and thus has no defined output shape. HOT 2
- Convert *.h5 (keras model) into *.pb and *.pbtxt (tensorflow model), and load them by opencv dnn
- About your Loss version function, and switch it for other HOT 1
- File "<string>", line unknown ParseError: syntax error: line 1, column 0
- I can't make the darknet
- NotFoundError: 2 root error(s) found when trying to run the code on Tensorflow 2 HOT 1
- tf.Print is deprecated
- How to train my Custom Dataset with this notebook
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 keras-yolo2.