Comments (3)
I totally understood.
Thank you very much and great work : )
from a-pytorch-tutorial-to-image-captioning.
This is also a good strategy to train I think. However, I'm unsure, as the authors also mentioned, whether this improves anything at all either in terms of speed or accuracy, since the "pack and pad" approach used by PyTorch does not do any extra computation. That's why the implementation here uses " pack and pad" (see lines 177-180 )
from a-pytorch-tutorial-to-image-captioning.
As our implementation requires time proportional to the length of the longest sentence per update, we found training on a random group of captions to be computationally wasteful. To mitigate this problem, in preprocessing we build a dictionary mapping the length of a sentence to the corresponding subset of captions. Then, during training we randomly sample a length and retrieve a mini-batch of size 64 of that length. We found that this greatly improved convergence speed with no noticeable diminishment in performance. On our largest dataset (MS COCO), our soft attention model took less than 3 days to train on an NVIDIA Titan Black GPU.
I assume they did this only from the standpoint of the time taken for each batch. As @kmario23 points out, there is no extra computation in our code since we're handling everything dynamically. But there is a slight time inefficiency in that the loop still runs to the maximum sentence length just for the sake of one or two long sentences in that batch.
I guess the downside with the same-length batch is that the order of training is not truly random in this case, so one might expect a slight decrease in performance, and hence they specifically report that this decrease wasn't noticeable.
I didn't feel the need to do this because training is pretty fast anyway. I felt it is nicer (and more generally applicable) to randomly sample sentences in each batch, and also to demonstrate that PyTorch can be used flexibly/dynamically and about the utility of packed sequences.
from a-pytorch-tutorial-to-image-captioning.
Related Issues (20)
- why use LSTMCell not use LSTM directly HOT 3
- Asking about List of Packages Version
- Failed to create .hdf5 files
- bleu-4 does not increase HOT 2
- Code stopped @ train.py with this error,
- Anyone , Please help to solve this error HOT 3
- If i want to use RL for this , how will i do ?
- eval.val is not working.
- Dimension error HOT 2
- Dataset not available HOT 3
- ValueError: max() arg is an empty sequence HOT 3
- High loss and low bleu-4 for training HOT 7
- Example Notebook?? HOT 1
- In eval.py, it seems that for each image, the model has to recalculate 5 times, is it too inefficient?
- Please help me with the error:RuntimeError: Expected target size [32, 9490], got [32, 51] HOT 1
- I think this a bug. caption.py 140 HOT 5
- RuntimeError: CUDA error: device-side assert triggered
- Only the author mentions half the bleu-4 score HOT 2
- can this model detect and recognize text in images containing text HOT 1
- I get this issues:Dimension out of range (expected to be in range of [-2, 1], but got 2) HOT 1
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 a-pytorch-tutorial-to-image-captioning.