flax-community
/

roberta-swahili

Inference Endpoints

Model card Files Files and versions Metrics Training metrics Community

fgaim commited on Jul 13, 2021

Commit

0e04b95

•

1 Parent(s): 227f66d

Update data sources in run script

Files changed (1) hide show

run.sh +2 -2

run.sh CHANGED Viewed

@@ -3,8 +3,8 @@
     --model_type="roberta" \
     --config_name="./" \
     --tokenizer_name="./" \
-    --dataset_name="mc4" \
-    --dataset_config_name="sw" \
     --max_seq_length="128" \
     --weight_decay="0.025" \
     --per_device_train_batch_size="128" \

     --model_type="roberta" \
     --config_name="./" \
     --tokenizer_name="./" \
+    --train_file="/home/shared/clean_swahili/train_v1.4.txt" \
+    --validation_split_percentage="2" \
     --max_seq_length="128" \
     --weight_decay="0.025" \
     --per_device_train_batch_size="128" \