Comments (3)
Thank you for the suggestion @khornlund, the cookiecutter project looks quite interesting.
I've tried your example, and got some questions for you.
I noticed initialization process asking a bunch of questions as following.
repo_name [my_repo_name]: trial_run_ckc
author_name [Karl Hornlund]: Seonkyu Park
author_email [[email protected]]: [email protected]
github_username [khornlund]: SunQparK
package_version [0.0.1]: 0.0.1
package_description [Short description]: what these questions are for?
package_name [trial_run_ckc]:
package_url [https://github.com/SunQparK/trial_run_ckc]:
New project directory appeared after that, but the structure looked weird.
trial_run_ckc
├── trial_run_ckc
│ └─ why source code is in this nested directory??
└─ other file, folders: what these files are for?
I think these things are quite uncomfortable. Are these features necessary for using cookiecutter templates? And, what are the advantages of using cookiecuter over copy script??
As I told in issue #43, I'd like to encourage people to save a copy of this template locally and customize when they need. Python script looks just fine for that purpose.
from pytorch-template.
Hi @SunQpark ,
I've made a bunch of other changes to my repo, so that's why the structure is different. This is not necessary though - you could keep the structure the same and still use cookiecutter
.
The advantage of cookiecutter is the ability to set a list of questions (with default answers) to be asked, and then use those answers to create the template. For example when you created the new trial_run_ckc
directory, it would have created a setup.py
file for you with all the relevant information filled in.
The addition of the setup.py
file is just something I did to my repo - you wouldn't have to do the same. I'm just using it as an example for how cookiecutter
can insert initialisation values into the template you create.
I actually found this repo while looking for a pytorch cookiecutter template. The best I could find was here, but it didn't have as good functionality. As you can see on the cookiecutter repo they link to lots of different templates for different kinds of projects, but there aren't any pytorch
ones yet. I think this repo would be a good candidate to become the pytorch cookiecutter template :)
from pytorch-template.
Currently I have two questions on this. cookiecutter seems to require a specific format to be used as template. Maybe we can maintain a branch for this.
Is it possible to filter out unnecessary files like readme, cache and data while having same directory structure as original repository(for easy maintenance)?
And is it possible to initialize little bit different code depending on python version(eg. can I use different code for python 3.6 and 3.5)?
If so, let's work on this after your PR #45 is done.
from pytorch-template.
Related Issues (20)
- Loss function HOT 1
- Any plans to support Wandb Hyperparameter Searching?
- Strange bugs occur when the number of Gpus trained and tested is inconsistent
- usage of shuffle=True when using SubsetRandomSampler HOT 1
- TODO: also configure logging for sub-processes(not master) HOT 4
- DataLoader example
- python train.py --resume path/to/checkpoint HOT 1
- Save best model HOT 1
- Model is moved to GPU after the optimizer is instatiated, resulting in a performance hit. HOT 4
- Setting 'early_stop: 0' does not disable it HOT 2
- Passing an iterable from config.json HOT 2
- Add custom flag & override from CLI HOT 1
- Some features I have implemented.
- ReduceLROnPlateau lr_scheduler HOT 3
- Only `data_loader.data_loaders` is plural HOT 5
- Any support for multiple loss functions? HOT 5
- Latest checkpoint HOT 8
- Some thing wrong with add_histogram function HOT 2
- Adding mesh to Tensorboard not working via TensorboardWriter HOT 2
- Any plans to support DistributedDataParallel? HOT 2
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 pytorch-template.