From 75b8990d9068a2c6ef448c190f2595c17fbcb993 Mon Sep 17 00:00:00 2001 From: Chungman Lee <50270612+fullyz@users.noreply.github.com> Date: Mon, 2 Aug 2021 19:58:43 +0900 Subject: [PATCH] fix typo in example/text-classification README (#12974) * fix typo in example/text-classification README * add space to align the table --- examples/flax/text-classification/README.md | 2 +- examples/pytorch/text-classification/README.md | 8 ++++---- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/examples/flax/text-classification/README.md b/examples/flax/text-classification/README.md index cb2c27d141..8fd64f0e44 100644 --- a/examples/flax/text-classification/README.md +++ b/examples/flax/text-classification/README.md @@ -100,7 +100,7 @@ In the Tensorboard results linked below, the random seed of each model is equal | Task | Metric | Acc (best run) | Acc (avg/5runs) | Stdev | Metrics | |-------|------------------------------|----------------|-----------------|-----------|--------------------------------------------------------------------------| -| CoLA | Matthew's corr | 60.57 | 59.04 | 1.06 | [tfhub.dev](https://tensorboard.dev/experiment/lfr2adVpRtmLDALKrElkzg/) | +| CoLA | Matthews corr | 60.57 | 59.04 | 1.06 | [tfhub.dev](https://tensorboard.dev/experiment/lfr2adVpRtmLDALKrElkzg/) | | SST-2 | Accuracy | 92.66 | 92.23 | 0.57 | [tfhub.dev](https://tensorboard.dev/experiment/jYvfv2trRHKMjoWnXVwrZA/) | | MRPC | F1/Accuracy | 89.90/85.78 | 88.97/84.36 | 0.72/1.09 | [tfhub.dev](https://tensorboard.dev/experiment/bo3W3DEoRw2Q7YXjWrJkfg/) | | STS-B | Pearson/Spearman corr. | 89.04/88.70 | 88.94/88.63 | 0.07/0.07 | [tfhub.dev](https://tensorboard.dev/experiment/fxVwbLD7QpKhbot0r9rn2w/) | diff --git a/examples/pytorch/text-classification/README.md b/examples/pytorch/text-classification/README.md index a2dec9c673..4874787529 100644 --- a/examples/pytorch/text-classification/README.md +++ b/examples/pytorch/text-classification/README.md @@ -51,10 +51,10 @@ single Titan RTX was used): | Task | Metric | Result | Training time | |-------|------------------------------|-------------|---------------| -| CoLA | Matthew's corr | 56.53 | 3:17 | +| CoLA | Matthews corr | 56.53 | 3:17 | | SST-2 | Accuracy | 92.32 | 26:06 | | MRPC | F1/Accuracy | 88.85/84.07 | 2:21 | -| STS-B | Person/Spearman corr. | 88.64/88.48 | 2:13 | +| STS-B | Pearson/Spearman corr. | 88.64/88.48 | 2:13 | | QQP | Accuracy/F1 | 90.71/87.49 | 2:22:26 | | MNLI | Matched acc./Mismatched acc. | 83.91/84.10 | 2:35:23 | | QNLI | Accuracy | 90.66 | 40:57 | @@ -90,10 +90,10 @@ Using mixed precision training usually results in 2x-speedup for training with t | Task | Metric | Result | Training time | Result (FP16) | Training time (FP16) | |-------|------------------------------|-------------|---------------|---------------|----------------------| -| CoLA | Matthew's corr | 56.53 | 3:17 | 56.78 | 1:41 | +| CoLA | Matthews corr | 56.53 | 3:17 | 56.78 | 1:41 | | SST-2 | Accuracy | 92.32 | 26:06 | 91.74 | 13:11 | | MRPC | F1/Accuracy | 88.85/84.07 | 2:21 | 88.12/83.58 | 1:10 | -| STS-B | Person/Spearman corr. | 88.64/88.48 | 2:13 | 88.71/88.55 | 1:08 | +| STS-B | Pearson/Spearman corr. | 88.64/88.48 | 2:13 | 88.71/88.55 | 1:08 | | QQP | Accuracy/F1 | 90.71/87.49 | 2:22:26 | 90.67/87.43 | 1:11:54 | | MNLI | Matched acc./Mismatched acc. | 83.91/84.10 | 2:35:23 | 84.04/84.06 | 1:17:06 | | QNLI | Accuracy | 90.66 | 40:57 | 90.96 | 20:16 |