SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
ahotrod
2020-04-19 10:49:24 -07:00
committed by Julien Chaumond
parent a21d4fa410
commit 52c85f847a
1 changed files with 24 additions and 36 deletions
Download Patch File Download Diff File Expand all files Collapse all files

60
model_cards/ahotrod/roberta_large_squad2/README.md
View File

@@ -3,66 +3,54 @@
### with the following results: ### with the following results:
``` ```
"exact": 84.02257222269014, "exact": 84.46896319380106,
"f1": 87.47063479332766, "f1": 87.85388093408943,
"total": 11873, "total": 11873,
"HasAns_exact": 81.19095816464238, "HasAns_exact": 81.37651821862349,
"HasAns_f1": 88.0969714745582, "HasAns_f1": 88.1560607844881,
"HasAns_total": 5928, "HasAns_total": 5928,
"NoAns_exact": 86.84608915054667, "NoAns_exact": 87.55256518082422,
"NoAns_f1": 86.84608915054667, "NoAns_f1": 87.55256518082422,
"NoAns_total": 5945, "NoAns_total": 5945,
"best_exact": 84.02257222269014, "best_exact": 84.46896319380106,
"best_exact_thresh": 0.0, "best_exact_thresh": 0.0,
"best_f1": 87.47063479332759, "best_f1": 87.85388093408929,
"best_f1_thresh": 0.0 "best_f1_thresh": 0.0
``` ```
### from script: ### from script:
``` ```
python -m torch.distributed.launch --nproc_per_node=2 ${RUN_SQUAD_DIR}/run_squad.py \ python ${EXAMPLES}/run_squad.py \
--model_type roberta \ --model_type roberta \
--model_name_or_path roberta-large \ --model_name_or_path roberta-large \
--do_train \ --do_train \
--train_file ${SQUAD_DIR}/train-v2.0.json \ --do_eval \
--predict_file ${SQUAD_DIR}/dev-v2.0.json \ --train_file ${SQUAD}/train-v2.0.json \
--predict_file ${SQUAD}/dev-v2.0.json \
--version_2_with_negative \ --version_2_with_negative \
--num_train_epochs 2 \
--warmup_steps 328 \
--weight_decay 0.01 \
--do_lower_case \ --do_lower_case \
--learning_rate 1.5e-5 \ --num_train_epochs 3 \
--warmup_steps 1642 \
--weight_decay 0.01 \
--learning_rate 3e-5 \
--adam_epsilon 1e-6 \
--max_seq_length 512 \ --max_seq_length 512 \
--doc_stride 128 \ --doc_stride 128 \
--save_steps 1000 \ --per_gpu_train_batch_size 8 \
--per_gpu_train_batch_size 1 \ --gradient_accumulation_steps 6 \
--gradient_accumulation_steps 24 \ --per_gpu_eval_batch_size 48 \
--threads 12 \
--logging_steps 50 \ --logging_steps 50 \
--threads 10 \ --save_steps 2000 \
--overwrite_cache \
--overwrite_output_dir \
--output_dir ${MODEL_PATH}
python ${RUN_SQUAD_DIR}/run_squad.py \
--model_type roberta \
--model_name_or_path ${MODEL_PATH} \
--do_eval \
--train_file ${SQUAD_DIR}/train-v2.0.json \
--predict_file ${SQUAD_DIR}/dev-v2.0.json \
--version_2_with_negative \
--do_lower_case \
--max_seq_length 512 \
--per_gpu_eval_batch_size 24 \
--eval_all_checkpoints \
--overwrite_output_dir \ --overwrite_output_dir \
--output_dir ${MODEL_PATH} --output_dir ${MODEL_PATH}
$@ $@
``` ```
### using the following system & software: ### using the following system & software:
``` ```
OS/Platform: Linux-4.15.0-91-generic-x86_64-with-debian-buster-sid
GPU/CPU: 2 x NVIDIA 1080Ti / Intel i7-8700
Transformers: 2.7.0 Transformers: 2.7.0
PyTorch: 1.4.0 PyTorch: 1.4.0
TensorFlow: 2.1.0 TensorFlow: 2.1.0
Python: 3.7.7 Python: 3.7.7
OS/Platform: Linux-5.3.0-46-generic-x86_64-with-debian-buster-sid
CPU/GPU: Intel i9-9900K / NVIDIA Titan RTX 24GB
``` ```