fairseq · Issues· 1352 open

License Clarification for NLLB-200: Usage in Government-Funded Public Service

what is the minimum GPU RAM size to run the inference of this model?

NameError: name 'CriterionType' is not defined [wav2vec2 evaluation using stt.py]

Prebuilt wheels provided via 3rd party repository

xlsr model running issue

Using RoBERTa with more than 512 tokens

<?xml version="1.0" encoding="UTF-8"?> <rss xmlns:g="http://base.google.com/ns/1.0" version="2.0"> <channel> <title>Your Facebook Page Product Catalog</title> <link>https://www.yourwebsite.com</link> <description>Products for Facebook Shop</description> <item> <g:id>product-123</g:id> <g:title>Example T-Shirt</g:title> <g:description>Comfortable cotton t-shirt in various sizes.</g:description> <g:link>https://www.yourwebsite.com/product/example-t-shirt</g:link> <g:image_link>https://www.yourwebsite.com/images/tshirt.jpg</g:image_link> <g:condition>new</g:condition> <g:availability>in stock</g:availability> <g:price>19.99 USD</g:price> <g:brand>Example Brand</g:brand> <g:gtin>1234567890123</g:gtin> <g:google_product_category>Apparel & Accessories > Clothing > Tops</g:google_product_category> </item> <!-- Add more <item> elements for additional products --> </channel> </rss>

Add support for Python 3.12+

Can we make a option to protect something not to be translated

Is there a path for Commercial license for a fine-tuned version of NLLB-200?

apply_mask in HuBERT does not apply expected percentage of masks if the input shapes differ. which is true in the case of fintuning.

Dataclass error while importing Fairseq in Python 3.11

No such file or directory: 'fairseq\\version.txt

CPU OOM when pretraining hubert

Overflow issue with Fairseq Preprocess for large datasets