kubapok/ireland-news-headlines
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Merge branch 'master' of github.com:kubapok/ireland-news

Browse Source
This commit is contained in:
Jakub Pokrywka 2021-10-04 17:30:45 +02:00
commit 7439b3b37b
4 changed files with 62734 additions and 62730 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

76496
dev-0/out.tsv
View File

File diff suppressed because it is too large Load Diff

2
t5_no_year/04_predict.py
View File

@ -4,7 +4,7 @@ from tqdm import tqdm
from config import LABELS_LIST from config import LABELS_LIST
device = 'cuda' device = 'cuda'
model_path= 't5-retrained' model_path= 't5-retrained/checkpoint-110000'
from transformers import AutoModelForSequenceClassification from transformers import AutoModelForSequenceClassification

8
t5_no_year/run.sh
View File

@ -1,15 +1,19 @@
python run_summarization.py \ python run_summarization.py \
--model_name_or_path t5-base \ --model_name_or_path google/t5-v1_1-base \
--do_train \ --do_train \
--train_file ../train/huggingface_format.csv \ --train_file ../train/huggingface_format.csv \
--validation_file ../dev-0/huggingface_format.csv \ --validation_file ../dev-0/huggingface_format_short.csv \
--source_prefix "classify: " \ --source_prefix "classify: " \
--summary_column 'label' \ --summary_column 'label' \
--max_target_length=4 \ --max_target_length=4 \
--max_source_length=64 \ --max_source_length=64 \
--num_train_epochs 20 \
--output_dir ./t5-retrained \ --output_dir ./t5-retrained \
--per_device_train_batch_size=16 \ --per_device_train_batch_size=16 \
--per_device_eval_batch_size=16 \ --per_device_eval_batch_size=16 \
--gradient_accumulation_steps=4 \
--overwrite_output_dir \ --overwrite_output_dir \
--predict_with_generate \ --predict_with_generate \
--save_steps=10000 \ --save_steps=10000 \
--eval_steps=10000 \
--evaluation_strategy steps

48958
test-A/out.tsv
View File

File diff suppressed because it is too large Load Diff