Instructions to use Reyansh4/T5_on_Billsum with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Reyansh4/T5_on_Billsum with Transformers:
# 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="Reyansh4/T5_on_Billsum")# Load model directly from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("Reyansh4/T5_on_Billsum") model = AutoModelForSeq2SeqLM.from_pretrained("Reyansh4/T5_on_Billsum", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Model Card for Model ID
This is the T5 Model on that is trained on the BillSum dataset.
Trained on the Google Colab's T4 GPU
Training Arguments
training_args = Seq2SeqTrainingArguments(
"bert-on-the-billsum",
evaluation_strategy="epoch",
learning_rate=2e-5,
per_device_train_batch_size=2,
per_device_eval_batch_size=2,
weight_decay=0.01,
save_total_limit=3,
num_train_epochs=4,
predict_with_generate=True,
fp16=True,
push_to_hub=True,
)
Training Results
Epoch Training_Loss Validation_Loss Rouge1 Rouge2 Rougel Rougelsum Gen_Len
1 No log 1.920353 0.193900 0.105400 0.168900 0.168800 19.000000
2 2.296300 1.860363 0.195300 0.105100 0.171400 0.171400 19.000000
3 1.927000 1.834611 0.195500 0.106800 0.171600 0.171500 19.000000
4 1.849100 1.826394 0.195000 0.105200 0.171400 0.171300 19.000000
- Downloads last month
- 3