Diverse Beam Search: Decoding Diverse Solutions from Neural Sequence Models
Paper • 1610.02424 • Published • 1
# Load model directly
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained("alexrink/Pegasus_VTSSum_3epochs")
model = AutoModelForSeq2SeqLM.from_pretrained("alexrink/Pegasus_VTSSum_3epochs", device_map="auto")Quick Links
google/pegasus-large trained on sample of VT-SSum (https://github.com/Dod-o/VT-SSum; https://arxiv.org/pdf/1610.02424.pdf)
Sample consists of 600 train, 200 validation, 200 test using categories Computer Science, Data Science, Mathematics
- Train Loss: ~5.3
- Validation Loss: ~4.1
- Downloads last month
- 6
# Use a pipeline as a high-level helper # Warning: Pipeline type "summarization" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # 'pip install "transformers<5.0.0' from transformers import pipeline pipe = pipeline("summarization", model="alexrink/Pegasus_VTSSum_3epochs")