stefan-it commited on
Commit
2297386
1 Parent(s): 80f4495

readme: fix figures link

Browse files
Files changed (1) hide show
  1. README.md +2 -2
README.md CHANGED
@@ -54,7 +54,7 @@ resulting in a total corpus size of 21GB.
54
 
55
  The following figure shows an overview of the number of chars per year distribution:
56
 
57
- ![Delpher Corpus Stats](/figures/delpher_corpus_stats.png)
58
 
59
  # Language Model Pretraining
60
 
@@ -81,7 +81,7 @@ python3 run_pretraining.py --input_file gs://delpher-bert/tfrecords/*.tfrecord \
81
  We train the model for 3M steps using a total batch size of 128 on a v3-32 TPU. The pretraining loss curve can be seen
82
  in the next figure:
83
 
84
- ![Delpher Pretraining Loss Curve](/figures/training_loss.png)
85
 
86
  # Evaluation
87
 
 
54
 
55
  The following figure shows an overview of the number of chars per year distribution:
56
 
57
+ ![Delpher Corpus Stats](figures/delpher_corpus_stats.png)
58
 
59
  # Language Model Pretraining
60
 
 
81
  We train the model for 3M steps using a total batch size of 128 on a v3-32 TPU. The pretraining loss curve can be seen
82
  in the next figure:
83
 
84
+ ![Delpher Pretraining Loss Curve](figures/training_loss.png)
85
 
86
  # Evaluation
87