philippelaban/summary_loop24

Try out in the Hosted inference API

In the right panel, you can try to the model (although it only handles a short sequence length). Enter the document you want to summarize in the panel on the right.

Model Loading

The model (based on a GPT2 base architecture) can be loaded in the following way:

from transformers import GPT2LMHeadModel, GPT2TokenizerFast

model = GPT2LMHeadModel.from_pretrained("philippelaban/summary_loop46")
tokenizer = GPT2TokenizerFast.from_pretrained("philippelaban/summary_loop46")

Example Use

document = "Bouncing Boulders Point to Quakes on Mars. A preponderance of boulder tracks on the red planet may be evidence of recent seismic activity. If a rock falls on Mars, and no one is there to see it, does it leave a trace? Yes, and it's a beautiful herringbone-like pattern, new research reveals. Scientists have now spotted thousands of tracks on the red planet created by tumbling boulders. Delicate chevron-shaped piles of Martian dust and sand frame the tracks, the team showed, and most fade over the course of a few years. Rockfalls have been spotted elsewhere in the solar system, including on the moon and even a comet. But a big open question is the timing of these processes on other worlds — are they ongoing or did they predominantly occur in the past?"

tokenized_document = tokenizer([document], max_length=300, truncation=True, return_tensors="pt")["input_ids"].cuda()
input_shape = tokenized_document.shape
outputs = model.generate(tokenized_document, do_sample=False, max_length=500, num_beams=4, num_return_sequences=4, no_repeat_ngram_size=6, return_dict_in_generate=True, output_scores=True)
candidate_sequences = outputs.sequences[:, input_shape[1]:] # Remove the encoded text, keep only the summary
candidate_scores = outputs.sequences_scores.tolist()

for candidate_tokens, score in zip(candidate_sequences, candidate_scores):
    summary = tokenizer.decode(candidate_tokens)
    print("[Score: %.3f] %s" % (score, summary[:summary.index("END")]))

Example output

[Score: -0.113]  These tracks have been spotted elsewhere in the solar system, including on the red planet, and no one is there to see it, does it leave a trace? Yes, and
[Score: -0.119]  Now researchers have spotted thousands of tracks on the red planet created by tumbling boulders in Mars, and no one is there to see it, does it leave a trace?
[Score: -0.214]  Here are answers to those questions posed by scientists investigating the tracks discovered by scientists examining the tracks discovered by scientists exploring the tracks discovered by scientists exploring the tracks discovered by scientists exploring the
[Score: -0.388]  These are the kinds of questions swirling around whether these tracks exist on Mars, and whether they should be noticed sooner rather than later. Here are some answers: -- The tracks detected

Github repo

You can access more information, access to the scoring function, the training script, or an example training log on the Github repo: https://github.com/CannyLab/summary_loop