vuiseng9
/

bert-base-squadv1-block-pruning-hybrid-filled-lt-nncf-50.0sparse-qat-lt

Inference Endpoints

Model card Files Files and versions Community

Chua, Vui Seng commited on Jan 9, 2022

Commit

73501a1

•

1 Parent(s): 2524d65

Update eval batch size

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -105,9 +105,9 @@ nohup python run_qa.py  \
     --nncf_config $MODELROOT/nncf_bert_squad_sparsity.json  \
     --to_onnx $OUTDIR/bert-base-squadv1-block-pruning-hybrid-filled-lt-nncf-50.0sparse-qat-lt.onnx  \
     --do_eval  \
-    --per_device_eval_batch_size 16  \
     --max_seq_length 384  \
     --doc_stride 128  \
     --overwrite_output_dir \
     --output_dir $OUTDIR 2>&1 | tee $OUTDIR/run.log &
-```

     --nncf_config $MODELROOT/nncf_bert_squad_sparsity.json  \
     --to_onnx $OUTDIR/bert-base-squadv1-block-pruning-hybrid-filled-lt-nncf-50.0sparse-qat-lt.onnx  \
     --do_eval  \
+    --per_device_eval_batch_size 128  \
     --max_seq_length 384  \
     --doc_stride 128  \
     --overwrite_output_dir \
     --output_dir $OUTDIR 2>&1 | tee $OUTDIR/run.log &
+```