|
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!") |
|
|