Comments (16)
@wljcode
Have you successfully reimplemented the experimental results? I checked the relevant logs and found that you adopt the load_weights to continue training rather than load_checkpoints, the former only load the model weights and the latter will load all training relevant parameters. It is expected to adopt load_checkpoints to continue training.
from vac_cslr.
第一份是使用VAC算法的log文件
log.txt
第二份是使用baseline的log文件
log.txt
from vac_cslr.
@wljcode
Thanks for your attention to our work. It seems like you used batch size=1, which may affect the robustness of the model. Besides, the learning rate does not decay during training? Perhaps because there may exist some bugs in checkpoint loading, I will check this later.
Relevant logs are uploaded for comparson.
baseline.txt
baseline_bn.txt
baseline_VAC.txt
from vac_cslr.
Thank you for your reply. Due to the limitation of GPU memory, we did not continue the experiment reproduction work recently. We will complete your work later when the equipment is ready !
from vac_cslr.
Thank you for your reply. Due to the limitation of GPU memory, we did not continue the experiment reproduction work recently. We will complete your work later when the equipment is ready !
from vac_cslr.
Hi, @ycmin95 , thanks for your great work.
I try to reproduce the work recently. The final result is 0.4% worse than yours.
Here is my training log.
log.txt
After 70th epoch, the performance cannot be improved as yours.
Besides, I find "label_smoothing = 0.1" in your log, but not in the released code.
Could you provide some advice?
from vac_cslr.
Hi, @sunke123, thanks for your attention to our work.
We will explain this performance gap in our next update, perhaps in two weeks,
which can achieve better performance (about 20% WER) with fewer training epochs.
You can conduct further experiments on this codebase, the update won't change the network structure and the training process.
The parameter label_smoothing is adopted in our early experiment about iterative training and I forget to delete this parameter, I will correct it in the next update.
from vac_cslr.
@ycmin95
Cooooool!
Thanks for your reply.
Looking forward to that~
from vac_cslr.
Hi, @sunke123,
the code has been updated~
from vac_cslr.
你好,请问我下载了代码重新训练了一下,但几轮下来DEV wer依旧是100%.我把batchsize调成1,lr调成0.000010.可以给点意见吗,谢谢.
from vac_cslr.
@herochen7372
You can first check whether the evaluation script runs as expected with the provided pretrained model, and than check whether the loss decreases as the iteration progresses.
from vac_cslr.
@ycmin95
Thanks for your reply.
from vac_cslr.
@wljcode Have you successfully reimplemented the experimental results? I checked the relevant logs and found that you adopt the load_weights to continue training rather than load_checkpoints, the former only load the model weights and the latter will load all training relevant parameters. It is expected to adopt load_checkpoints to continue training.
@ycmin95
Hello author, I have encountered the same problem. Can you provide more detailed information on how to solve this problem? Sorry, I didn't understand the method here. Thank you very much. I only have a 3060 GPU, so my batchsize = 1.
I noticed in your log that there are Dev WER and Test WER for each epoch of training, but mine is only Dev WER.
Looking forward to your help.
dev.txt
log.txt
from vac_cslr.
It seems that you only train the baseline without the proposed VAC or SMKD, please follow the Readme.md to set the configuration file. We remove the evaluation on the test set during training for efficiency, you can modify this process in the main.py.
from vac_cslr.
Hello, @ycmin95.
After configuring the settings according to the readme, I ran 40 epoch, but the best result achieved was only 32.3%. Could you please let me know if I might have missed any settings?
log_SMKD_no_ConvCTC.txt
dev_SMKD_no_ConvCTC.txt
I've noticed "# ConvCTC: 1.0" in the baseline.yaml file. I added ConvCTC: 1.0 and retrained 80 epoch, but the results were even worse.
log_SMKD_ConvCTC.txt
dev_SMKD_ConvCTC.txt
from vac_cslr.
Hi, @kido1412y2y
Can you report the evaluation results with a batch size larger than 1? I never run experiments with batch size of 1, and not sure the influence of it.
from vac_cslr.
Related Issues (20)
- Error when i run the training part main.py HOT 1
- unable to run successfully when I run main.py HOT 1
- unable to successfully install CTC HOT 5
- Detailed code running steps HOT 19
- Visualizing the predicted alignments HOT 1
- unable to run these repository on google colab HOT 1
- Data augmention error HOT 4
- about the feature extractor architecture HOT 6
- Successfully inference, but unable to train HOT 3
- Visualize Epoch vs Loss HOT 1
- Hardware and Software Specifications for this research. HOT 3
- Gloss Segment Boundary Assignment Algorithm (GSBA) HOT 1
- Does your PHOENIX-2014-T dataset have /1/ folder? HOT 1
- model or model.parameters()? HOT 2
- Does your model produce different results each time it is run? HOT 4
- Training the baseline (without VAC, SMKD) HOT 2
- How can I use a trained model to test a video? HOT 1
- unable to reconize any word but the loss is decreasing??? HOT 1
- Does preprocess.sh have an effect if using datasets other than phoneix? HOT 1
- IndexError in DataLoader Worker Process with Custom Dataset
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 vac_cslr.