Interactive Neural Machine Translation-lite (INMT-lite) is a framework to train and develop lite versions (.tflite) of models for neural machine translation (NMT) that can be run on embedded devices like mobile phones and tablets that have low computation power and space. The tflite models generated can be used to build the offline version of INMT mobile, a mobile version of INMT web.
License: MIT License
The model generated has to be manually pasted into the app asset folder and also some code has to be manually changed to build the app.
A script to automatically build the app using model configuration providing all the parameters about model including vocabulary and model path would smooth the process
wget https://inmtlite.blob.core.windows.net/inmtlite-public-access-models/Hi_Gondi_Deployment_mt5_model.zip
->
ERROR 409: Public access is not permitted on this storage account..
Hi,
Thank you for your great work.
I followed your instructions and tried to train a TFLite model.
But it looks like the models do not converge well.
The loss is as follows:
Epoch 1 Batch 0 Loss 10.1416
Epoch 1 Batch 100 Loss 6.4002
Epoch 1 Batch 200 Loss 5.1489
Epoch 1 Training Loss 6.0853
Time taken for 1 epoch 76.00192999839783 sec
Epoch 1 Validation Loss 11.4574
Time taken for validation 2.3923656940460205 sec
......
Epoch 99 Batch 0 Loss 1.8126
Epoch 99 Batch 100 Loss 2.3800
Epoch 99 Batch 200 Loss 2.4746
Epoch 99 Training Loss 2.4855
Time taken for 1 epoch 52.326969385147095 sec
Epoch 99 Validation Loss 25.6011
Time taken for validation 2.445949077606201 sec
Epoch 100 Batch 0 Loss 2.5111
Epoch 100 Batch 100 Loss 2.7126
Epoch 100 Batch 200 Loss 2.5386
Epoch 100 Training Loss 2.5118
Time taken for 1 epoch 56.2775776386261 sec
Epoch 100 Validation Loss 25.9406
Time taken for validation 6.213912010192871 sec
And I tested the ACC of the model. It is only around 8.3.
What should I do to improve the performance of the model.
Presently, the model can train on 320,000 sentence pairs consisting 14 tokens each on a Tesla P100 GPU. The dataset is loaded all at once in memory.
Constructing data pipelines would allow picking up only batch size of data into memory allowing it to train on larger datasets
Hello experts,
Thank you for your contribution.
I tried to train a model with 90000 sentences from English to Spanish, but the performance of my model is not good.
I tried to change recurrent_hidden to 1024, also, I changed --src_word_vec_size and --tgt_word_vec_size, as you recomended, but the problem is the same.
Below I show the results for 20 epochs. As you can see, the validation lost is bad.
Could you help me, please?
Epoch 10 Batch 0 Loss 5.6960
Epoch 10 Batch 100 Loss 5.3570
Epoch 10 Training Loss 5.4667
Time taken for 1 epoch 529.2065329551697 sec
Epoch 10 Validation Loss 10.6312
Time taken for validation 60.65908360481262 sec
Epoch 11 Batch 0 Loss 4.7367
Epoch 11 Batch 100 Loss 5.1832
Epoch 11 Training Loss 5.1685
Time taken for 1 epoch 529.9154849052429 sec
Epoch 11 Validation Loss 10.8019
Time taken for validation 59.725284814834595 sec
Epoch 12 Batch 0 Loss 4.7126
Epoch 12 Batch 100 Loss 5.1652
Epoch 12 Training Loss 5.1701
Time taken for 1 epoch 532.0332324504852 sec
Epoch 12 Validation Loss 11.5809
Time taken for validation 60.39210081100464 sec
Epoch 13 Batch 0 Loss 4.7674
Epoch 13 Batch 100 Loss 5.1166
Epoch 13 Training Loss 5.0376
Time taken for 1 epoch 527.922877073288 sec
Epoch 13 Validation Loss 10.9454
Time taken for validation 60.049590826034546 sec
Epoch 14 Batch 0 Loss 4.6097
Epoch 14 Batch 100 Loss 5.0026
Epoch 14 Training Loss 4.9551
Time taken for 1 epoch 533.555011510849 sec
Epoch 14 Validation Loss 11.2711
Time taken for validation 62.934301137924194 sec
Epoch 15 Batch 0 Loss 4.5907
Epoch 15 Batch 100 Loss 4.6141
Epoch 15 Training Loss 4.8234
Time taken for 1 epoch 530.599461555481 sec
Epoch 15 Validation Loss 11.9423
Time taken for validation 59.88873314857483 sec
Epoch 16 Batch 0 Loss 4.5649
Epoch 16 Batch 100 Loss 4.8182
Epoch 16 Training Loss 4.7378
Time taken for 1 epoch 533.0982737541199 sec
Epoch 16 Validation Loss 11.9251
Time taken for validation 62.00506854057312 sec
Epoch 17 Batch 0 Loss 4.2148
Epoch 17 Batch 100 Loss 4.1748
Epoch 17 Training Loss 4.5810
Time taken for 1 epoch 530.5870060920715 sec
Epoch 17 Validation Loss 11.7330
Time taken for validation 60.24014854431152 sec
Epoch 18 Batch 0 Loss 4.2759
Epoch 18 Batch 100 Loss 4.6324
Epoch 18 Training Loss 4.5561
Time taken for 1 epoch 534.9815211296082 sec
Epoch 18 Validation Loss 12.2905
Time taken for validation 61.87094283103943 sec
Epoch 19 Batch 0 Loss 4.3838
Epoch 19 Batch 100 Loss 4.7557
Epoch 19 Training Loss 4.4805
Time taken for 1 epoch 533.8275711536407 sec
Epoch 19 Validation Loss 12.4593
Time taken for validation 62.41018629074097 sec
Epoch 20 Batch 0 Loss 4.0156
Epoch 20 Batch 100 Loss 4.3589
Epoch 20 Training Loss 4.3792
Time taken for 1 epoch 533.2257282733917 sec
Epoch 20 Validation Loss 12.6180
Time taken for validation 63.08395957946777 sec
You mentioned that for making the model Android-Compatible: We use an entirely different tokenization procedure.
Could you let us know where are these files?
I couldn't find them in GitHub.
We have to provide a mechanism to plot training and validation graphs by storing the metrics
Present implementation does not allow to continue training after an interruption by restoring the last checkpoint.
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 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.
An Open Source Machine Learning Framework for Everyone
Django
The Web framework for perfectionists with deadlines.
Laravel
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
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.