Comments (4)
Hi Ben,
Thanks for the questions. I'll try.
-
The point of
layer_base
, which is just a1x1
convolution, is to standardize the number of output channels toout_filters
before performing the main operation in a convolutional cell or a normal cell. In_enas_layer
, we do this infinal_conv
. The effect is almost the same, but we found it easier to implement this way. -
I don't understand this point of yours. Both
_fixed_layer
and_enas_layer
use both convolutions and pooling. Forfixed_layer
, I hope the code is quite straightforward. For_enas_layer
, since we need to implement a somewhat dynamic graph, we separate the process into the function_enas_cell
. -
The purpose of
_factorized_reduction
is to reduce both spatial dimensions (width and height) by a factor of 2, and potentially to change the number of output filters. Where you mention it, this function is used to make sure that the outputs of all operations in a convolutional cell or a reduction cell will have the same spatial dimensions, so that they can be concatenated along the depth dimension.
The reason why we cannot just fix normal_arc
and reduce_arc
and use the same code for both the search process and fixed-architecture process is efficiency. Dynamic graphs in TF, at least the way we implement them, are slow and very memory inefficient.
Let us know if you still have more questions 😃
from enas.
For number 2, the point was that you're using pooling w/ stride > 1 in the fixed architecture, but a combination of _factorized_reduction
and pooling w/ stride = 1 in the ENAS cells.
Makes sense about the dynamic graphs being slow.
Thanks for the quick response. (And thanks for releasing the code! I've been working on a similar project for a little while, so am very excited to compare what I've done to your code.)
~ Ben
from enas.
For number 2, the point was that you're using pooling w/ stride > 1 in the fixed architecture, but a combination of
_factorized_reduction
and pooling w/ stride = 1 in the ENAS cells.
I think it's just because we couldn't figure out how to syntactically make _factorized_reduction
run with the output of a dynamic operation, such as tf.case
.
from enas.
@hyhieu I am wondering if the reduction cell in _fixed_layer
and _enas_layer
have the same previous layers
result of _factorized_reduction
is appended to the layers
If I understand it correctly, to make the previous layers consistent, this line should be
layers = [layers[0], x]
from enas.
Related Issues (20)
- Adapting ENAS for chess
- potential change of NHCW to NHWC in micro_child.py
- Do you have plans to update this project to Tensorflow2.0?
- Understanding of the output of the NAS HOT 5
- Reproducibility of the results from the paper (RNN) on new repository
- what is the batch size for the sampling? HOT 2
- enas_PTB child model
- Which version of tensorflow should I use? HOT 2
- How to restore final trained model??
- CNN architechure search run error on tensorflow1.13
- How much memory to run the cifar10 example HOT 4
- Not able to run on GPU HOT 1
- The output of micro search HOT 1
- How to cumpute the gradients of controller variables? HOT 1
- Enas
- About the architecture
- Library requirements
- Unknown data_format
- output ENAS
- Process finished with exit code -1073741571 (0xC00000FD)
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 enas.