From bcd4aa8fe03b4391c55a19e0c46fdaa08e179b64 Mon Sep 17 00:00:00 2001 From: thomwolf Date: Tue, 15 Jan 2019 23:32:34 +0100 Subject: [PATCH] update evaluation example --- examples/eval_transfo_xl.py | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/examples/eval_transfo_xl.py b/examples/eval_transfo_xl.py index 886e826b2c8..15c2665782a 100644 --- a/examples/eval_transfo_xl.py +++ b/examples/eval_transfo_xl.py @@ -44,18 +44,18 @@ parser = argparse.ArgumentParser(description='PyTorch Transformer Language Model parser.add_argument('--model_name', type=str, default='transfo-xl-wt103', choices=['transfo-xl-wt103'], #, 'lm1b', 'enwik8', 'text8'], help='pretrained model name') -parser.add_argument('--split', type=str, default='all', +parser.add_argument('--split', type=str, default='test', choices=['all', 'valid', 'test'], help='which split to evaluate') parser.add_argument('--batch_size', type=int, default=10, help='batch size') -parser.add_argument('--tgt_len', type=int, default=5, +parser.add_argument('--tgt_len', type=int, default=128, help='number of tokens to predict') parser.add_argument('--ext_len', type=int, default=0, help='length of the extended context') -parser.add_argument('--mem_len', type=int, default=0, +parser.add_argument('--mem_len', type=int, default=1600, help='length of the retained previous heads') -parser.add_argument('--clamp_len', type=int, default=-1, +parser.add_argument('--clamp_len', type=int, default=1000, help='max positional embedding index') parser.add_argument('--cuda', action='store_true', help='use CUDA')