Closed Molugan closed 3 years ago
The current code is dependent on a specific commit of flashlight which is not on the master branch. I suggest to add the missing classes (forwardSequentialModuleWithPadMaskForCPC and CPCSpecAugment) here directly to facilitate the compatibility.
Added fixes for the build with respect to the recent changes
We have updated the code to make it compatible with the latest flashlight release.
The pretraining and the fine-tuning are working.
But the older checkpoints are no longer compatible, we are relaunching the training;
I suggest that we merge with the new pretrained base model, and add the fine-tuned version on another PR. What do you think ?
We have updated the code to make it compatible with the latest flashlight release.
The pretraining and the fine-tuning are working.
But the older checkpoints are no longer compatible, we are relaunching the training;
* the base pretrained model will be ready tomorrow (no fine-tuning, unsupervised training only) * the auxiliary fine-tuned models will be ready later this week or next week
I suggest that we merge with the new pretrained base model, and add the fine-tuned version on another PR. What do you think ?
Sounds good!
@jacobkahn has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
@Molugan has updated the pull request. You must reimport the pull request before landing.
@tlikhomanenko has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
IMPORTANT: Please do not create a Pull Request without creating an issue first. Changes must be discussed.
Original Issue: https://github.com/facebookresearch/wav2letter/issues/957
closes #[issue 957]
Summary
Patched version of Chaitanya Talnikar's implementation of masked_cpc: we needed to include the pre-training for the VoxPopuli dataset.
Test Plan (required)
Fine-tuning with Common Voices Latvian
After downloading Common Voices:
You should get the following output:
Download and uncompress the checkpoint from https://dl.fbaipublicfiles.com/voxpopuli/wav2letter_100k_small.tar.gz
To fine-tune the model: