CATIE English FAT5-Flan
Collection
Adapted weights for Google Flan-T5 to use with FAT5 • 4 items • Updated
How to use CATIE-AQ/FAT5-base-flan-en with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("feature-extraction", model="CATIE-AQ/FAT5-base-flan-en", trust_remote_code=True) # Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("CATIE-AQ/FAT5-base-flan-en", trust_remote_code=True, dtype="auto")Adaptation of the flan-t5-base weights to make it compatible with the FAT5 framework (Flash Attention T5).
This adaptation should enable the user to efficiently continue the pre-training of the flan-t5 to adapt it to more recent data, or to specialize it in a specific domain, for example.
from transformers import AutoModel, AutoTokenizer
model = AutoModel.from_pretrained("CATIE-AQ/FAT5-base-flan-en", trust_remote_code=True)
tokenizer = AutoTokenizer.from_pretrained("google/flan-t5-base")