Uses
from transformers import T5Tokenizer, T5ForConditionalGeneration
tokenizer = T5Tokenizer.from_pretrained("alpeshsonar/lot-t5-small-filter")
model = T5ForConditionalGeneration.from_pretrained("alpeshsonar/lot-t5-small-filter")
input_text = """"Extract lots from given text.
* age ≥18 years * patients with de novo or secondary AML, with an unfavorable or intermediate karyotype (according to the 2017 ELN classification), or patients with relapsing AML who may receive second-line treatment * not candidates for intensive induction, for the following reasons* 75 years or ≥ 18 to 74 years and at least one of the following comorbidities: PS ≥ 2 or a history of heart failure requiring treatment or LVEF ≤ 50% or chronic stable angina or FEV1 ≤ 65% or DLCO ≤ 65% or creatinine clearance <45 ml / min; or liver damage with total bilirubin> 1.5 N or other comorbidities that the hematologist considers incompatible with intensive treatment * ineligible for a classic allogeneic hematopoietic stem cell transplant due to the presence of co-morbidities or too high a risk of toxicity >70 years old or at least one of the following comorbidities: PS ≥ 2 or a history of heart failure requiring treatment or LVEF ≤ 50% or chronic stable angina or FEV1 ≤ 65% or DLCO ≤ 65% or creatinine clearance <45 ml / min; or liver damage with total bilirubin> 1.5 N * may receive chemotherapy with hypomethylating agents have a partially compatible (haplo-identical) major family donor (≥18 years old) eligible for lymphocyte donation.
"""
input_ids = tokenizer(input_text, return_tensors="pt").input_ids
outputs = model.generate(input_ids,max_new_tokens=1024)
print(tokenizer.decode(outputs[0],skip_special_tokens=True))
