yeephycho / densenet-tensorflow Goto Github PK
View Code? Open in Web Editor NEWA clean densenet in tensorflow
License: Apache License 2.0
A clean densenet in tensorflow
License: Apache License 2.0
Hello, I have achieved very good performance in my dataset using your code. Right now, I want to use ImageNet pretrain model to improve performance more. Could you tell me how can I use it based on your code? I have convert imagenet pretrain model based on caffe model at https://github.com/shicai/DenseNet-Caffe
You can download my convert pretrain model at https://drive.google.com/open?id=1nby_ZC7ELkqaVaipPc8eKrzoHb05BGx4
The downloaded dataset is not right
hello,is your pretrained model the imagenet pretrained model?If not, what's your pretrained dataset?I want to use denset to do place recognition , can I use your pretrained model?Thank you thank you ~~
I used python build_image_data.py for own data
"../build_image_data.py"
It creates shard files :
train-00000-of-00002
train-00001-of-00002
Now when I run ./train.py it cant find file train.tfrecord
ValueError: Failed to find file: ./own_data_dir/train.tfrecord
This seems to be hardcoded in data_provider.py (line 77) as:
filenames = [os.path.join(DATA_DIR, "train.tfrecord")]
How can I get the code to produce a file named "train.tfrecord" insteald of several separare files, 1 for each shard?
Or is there some other way to combine shard-files to single file named .tfrecord?
First of all, I would like to thank your code. It is very useful for me. I have used it and obtain 81% in the flower dataset.
Now, I will use it for my dataset which includes 10.000 png images in the range 0-255. I have two questions for making/training my dataset
build_image_data.py
has png_to_jpeg(image_data), so if my dataset has png file, can it still work for converting tfrecord?data provider.py
asimage_raw = tf.image.decode_jpeg(image_encoded, channels=3)
image_raw = tf.cast(image_raw , tf.float32) * (1. / 255)
Am I right?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.