| 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!") |
|
|
