fairseq
fairseq copied to clipboard
Facebook AI Research Sequence-to-Sequence Toolkit written in Python.
Given groups=1, weight of size [1024, 80, 5], expected input[11, 240, 2630] to have 80 channels, but got 240 channels instead #### Code fairseq-train $DATA_ROOT \ --config-yaml config.yaml --multitask-config-yaml config_multitask.yaml...
## ❓ Questions and Help i have given the input_feature_per_channel 80 but it gets the following help what can you help us ### Before asking: 1. search the issues. 2....
# Before submitting - [x] Was this discussed/approved via a Github issue? (no need for typos, doc improvements) - [x] Did you read the [contributor guideline](https://github.com/pytorch/fairseq/blob/main/CONTRIBUTING.md)? - [x] Did you...
I don't seem to be able to install `fairseq==0.10.2` with python 3.9 on macOS 11.3 ``` $ pip install fairseq==0.10.2 Collecting fairseq==0.10.2 Using cached fairseq-0.10.2.tar.gz (938 kB) Installing build dependencies...
I am trying to train a model with pipeline ``` echo `date` exp_dir=$1 model_arch=${2:-"transformer_18_18"} pretrained_ckpt=$3 fairseq-train $exp_dir/final_bin \ --max-source-positions=256 \ --max-target-positions=256 \ --source-lang=SRC \ --target-lang=TGT \ --max-update=1000000 \ --save-interval-updates=1000 \...
Please add one of South Africa's languages if not all.
Hello guys i have just trained my wav2vec2 model on Nepali speech dataset from [this](http://www.openslr.org/43/) page. I want to now fine tune this model but looks like i need a...
## 🐛 Bug Making recommended modifications to config.py from separate [issue](https://github.com/facebookresearch/fairseq/pull/5062) to allow 3.11 compatibility break fairseq and or runtime compatibility with hydra. ### Windows 10 machine, Python 3.11.3, fairseq==0.12.2,...
## 🚀 Feature Request I saw Newari Language Support there which is a minority language of Nepal, but I coulld not find Nepali Language support. Would be grateful, if you...