| from hfapi import Client | |
| client = Client() | |
| BATCH_SIZE = 4 | |
| LONG_LIST_OF_INPUTS = [ | |
| "I like you. </s></s> I love you.", | |
| "At the other end of Pennsylvania Avenue, people began to line up for a White House tour. </s></s> People formed a line at the end of Pennsylvania Avenue.", | |
| ] * 500 | |
| def chunker(seq, size): | |
| return (seq[pos:pos + size] for pos in range(0, len(seq), size)) | |
| all_results = [] | |
| for inputs in chunker(LONG_LIST_OF_INPUTS, BATCH_SIZE): | |
| result = client.text_classification(inputs, model="roberta-large-mnli") | |
| print(result) | |
| all_results += result | |
| print("Done!") | |