Views
No views yet
1abstract = """We describe a system called Overton, whose main design goal is to support engineers in building, monitoring, and improving production
2machine learning systems. Key challenges engineers face are monitoring fine-grained quality, diagnosing errors in sophisticated applications, and
3handling contradictory or incomplete supervision data. Overton automates the life cycle of model construction, deployment, and monitoring by providing a
4set of novel high-level, declarative abstractions. Overton's vision is to shift developers to these higher-level tasks instead of lower-level machine learning tasks.
5In fact, using Overton, engineers can build deep-learning-based applications without writing any code in frameworks like TensorFlow. For over a year,
6Overton has been used in production to support multiple applications in both near-real-time applications and back-of-house processing. In that time,
7Overton-based applications have answered billions of queries in multiple languages and processed trillions of records reducing errors 1.7-2.9 times versus production systems.
8"""1model_name = "snrspeaks/t5-one-line-summary"
2
3from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
4model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
5tokenizer = AutoTokenizer.from_pretrained(model_name)
6input_ids = tokenizer.encode("summarize: " + abstract, return_tensors="pt", add_special_tokens=True)
7generated_ids = model.generate(input_ids=input_ids,num_beams=5,max_length=50,repetition_penalty=2.5,length_penalty=1,early_stopping=True,num_return_sequences=3)
8preds = [tokenizer.decode(g, skip_special_tokens=True, clean_up_tokenization_spaces=True) for g in generated_ids]
9print(preds)
10
11# output
12["Overton: Building, Deploying, and Monitoring Machine Learning Systems for Engineers",
13 "Overton: A System for Building, Monitoring, and Improving Production Machine Learning Systems",
14 "Overton: Building, Monitoring, and Improving Production Machine Learning Systems"]1# pip install --upgrade simplet5
2from simplet5 import SimpleT5
3model = SimpleT5()
4model.load_model("t5","snrspeaks/t5-one-line-summary")
5model.predict(abstract)
6
7# output
8"Overton: Building, Deploying, and Monitoring Machine Learning Systems for Engineers"