File size: 1,398 Bytes
060ac52 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 |
# This is the official config file.
PECCAVI_TEXT:
Entailment:
task: "text-classification"
model: "ynie/roberta-large-snli_mnli_fever_anli_R1_R2_R3-nli"
Masking:
task: "fill-mask"
tokenizer: "bert-base-uncased"
model: "bert-base-uncased"
# tokenizer: "bert-large-cased-whole-word-masking"
# model: "bert-large-cased-whole-word-masking"
Vocabulary:
tokenizer: "bert-base-uncased"
model: "bert-base-uncased"
# permissible_ratio: 0.5
# tokenizer: "bert-large-cased-whole-word-masking"
# model: "bert-large-cased-whole-word-masking"
permissible_ratio: 1.0
Sampling:
tokenizer: "bert-base-uncased"
model: "bert-base-uncased"
# tokenizer: "bert-large-cased-whole-word-masking"
# model: "bert-large-cased-whole-word-masking"
Metrics:
EuclideanDistance: "sentence-transformers/all-MiniLM-L6-v2"
Distortion: "gpt2"
Detector:
tokenizer: "bert-base-uncased"
model: "bert-base-uncased"
# tokenizer: "bert-large-cased-whole-word-masking"
# model: "bert-large-cased-whole-word-masking"
Paraphrase:
tokenizer: "humarin/chatgpt_paraphraser_on_T5_base"
model: "humarin/chatgpt_paraphraser_on_T5_base"
num_beams: 10
num_beam_groups: 10
num_return_sequences: 10
repetition_penalty: 10.0
diversity_penalty: 3.0
no_repeat_ngram_size: 2
temperature: 0.7
max_length: 64
|