Comments (2)
Looks like your numbers are 0.3-0.4 worse than what I got.
The final train/validation loss shown in my LM's train.log
is:
| epoch 025 | loss 3.956 | ppl 15.52 | wps 19483 | ups 1 | wpb 27910.561 | bsz 1782.919 | num_updates 33875 | lr 9.76563e-07 | gnorm 0.160 | clip 0.021 | oom 0.000 | wall 48813 | train_wall 43183
| epoch 025 | valid on 'valid' subset | loss 4.119 | ppl 17.37 | num_updates 33875 | best_loss 4.11477
and those in my attention model's train.log
is:
| epoch 035 | loss 1.970 | nll_loss 0.649 | ppl 1.57 | wps 560 | ups 0 | wpb 1644.168 | bsz 69.317 | num_updates 97300 | lr 1e-05 | gnorm 0.341 | clip 0.000 | oom 0.000 | wall 32856 | train_wall 116514
| epoch 035 | valid on 'valid' subset | loss 6.214 | nll_loss 5.392 | ppl 41.99 | num_updates 97300 | best_wer 14.2055 | wer 14.2757 | cer 15.1188
LM on eval2000:
| Evaluated 60411 tokens in 3.4s (17886.71 tokens/s) | Loss: 3.5959, Perplexity: 36.45
LM on rt03:
| Evaluated 109828 tokens in 5.5s (19909.00 tokens/s) | Loss: 3.7694, Perplexity: 43.35
Anyways, it's common that we get slightly different results, since the type/number of GPUs, the PyTorch/CUDA version yield different numbers. For my experiments, I was using 2 GeForce 1080 GTX Ti GPUs with the following environment (you can obtain this info by following the instruction from PyTorch github when you are creating new Issue there):
Collecting environment information...
PyTorch version: 1.1.0
Is debug build: No
CUDA used to build PyTorch: 9.0.176
OS: Debian GNU/Linux 9.4 (stretch)
GCC version: (Debian 6.3.0-18+deb9u1) 6.3.0 20170516
CMake version: version 3.7.2
Python version: 3.7
Is CUDA available: Yes
CUDA runtime version: Could not collect
GPU models and configuration:
GPU 0: GeForce GTX 1080 Ti
GPU 1: GeForce GTX 1080 Ti
GPU 2: GeForce GTX 1080 Ti
GPU 3: GeForce GTX 1080 Ti
Nvidia driver version: 387.26
cuDNN version: Could not collect
Versions of relevant libraries:
[pip3] numpy==1.12.1
[conda] blas 1.0 mkl
[conda] mkl 2019.1 144
[conda] mkl-service 1.1.2 py37h90e4bf4_5
[conda] mkl_fft 1.0.10 py37ha843d7b_0
[conda] mkl_random 1.0.2 py37hd81dba3_0
[conda] pytorch 1.1.0 py3.7_cuda9.0.176_cudnn7.5.1_0 pytorch
[conda] torchvision 0.3.0 py37_cu9.0.176_1 pytorch
from espresso.
Thanks!
from espresso.
Related Issues (20)
- Stream ASR HOT 1
- TypeError: get_asr_dataset_from_json() got an unexpected keyword argument 'combined' HOT 1
- Different WERs when decoding with different batch size (--max-sentences) HOT 5
- ONNX exportation of speech_lstm based model HOT 1
- Build PyChain with CPU HOT 6
- WER difference when decode with different batch size HOT 5
- SpecAug slows down training time HOT 4
- WSJ Recipe: "wsj_data_prep.sh: Spot check of command line arguments failed" HOT 1
- SpecAugment is not used due to a typo in the prefetch_called assignment HOT 1
- Getting OOM error at the middle of the training in asr_swbd recipe on lstm encoder decoder model HOT 4
- How can I find Kaldi directory? HOT 37
- Google Colab research creation HOT 1
- token_text as outputs HOT 2
- Which is the difference between stage 5 and stage 7 HOT 1
- Error in training stage of run_chain_e2e_bichar.sh: 'odict_items' object is not an iterator HOT 4
- Transformer LM in ASR HOT 2
- 'frame_subsampling_factor' in chain e2e setup HOT 2
- hydra.errors.ConfigCompositionException: Could not override 'task.data'. HOT 1
- Android Espresso not able to test fragement
- SHA hashes in 'main' branch are different from those in the 'origin/main' 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 espresso.