Maltehb commited on
Commit
5dd75e3
1 Parent(s): e4259fc

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -54,7 +54,7 @@ model = AutoModelForPreTraining.from_pretrained("Maltehb/-l-ctra-danish-electra-
54
  On [DaNE](https://danlp.alexandra.dk/304bd159d5de/datasets/ddt.zip) (Hvingelby et al., 2020), Ælæctra scores slightly worse than both cased and uncased Multilingual BERT (Devlin et al., 2019) and Danish BERT (Danish BERT, 2019/2020), however, Ælæctra is less than one third the size, and uses significantly fewer computational resources to pretrain and instantiate. For a full description of the evaluation and specification of the model read the thesis: 'Ælæctra - A Step Towards More Efficient Danish Natural Language Processing'.
55
 
56
  ### Pretraining
57
- To pretrain Ælæctra it is recommended to build a Docker Container from the [Dockerfile](https://github.com/MalteHB/Ælæctra/tree/master/notebooks/fine-tuning/). Next, simply follow the [pretraining notebooks](https://github.com/MalteHB/-l-ctra/blob/master/notebooks/pretraining/)
58
 
59
  The pretraining was done by utilizing a single NVIDIA Tesla V100 GPU with 16 GiB, endowed by the Danish data company [KMD](https://www.kmd.dk/). The pretraining took approximately 4 days and 9.5 hours for both the cased and uncased model
60
 
 
54
  On [DaNE](https://danlp.alexandra.dk/304bd159d5de/datasets/ddt.zip) (Hvingelby et al., 2020), Ælæctra scores slightly worse than both cased and uncased Multilingual BERT (Devlin et al., 2019) and Danish BERT (Danish BERT, 2019/2020), however, Ælæctra is less than one third the size, and uses significantly fewer computational resources to pretrain and instantiate. For a full description of the evaluation and specification of the model read the thesis: 'Ælæctra - A Step Towards More Efficient Danish Natural Language Processing'.
55
 
56
  ### Pretraining
57
+ To pretrain Ælæctra it is recommended to build a Docker Container from the [Dockerfile](https://github.com/MalteHB/-l-ctra/blob/master/infrastructure/Dockerfile). Next, simply follow the [pretraining notebooks](https://github.com/MalteHB/-l-ctra/blob/master/notebooks/pretraining/)
58
 
59
  The pretraining was done by utilizing a single NVIDIA Tesla V100 GPU with 16 GiB, endowed by the Danish data company [KMD](https://www.kmd.dk/). The pretraining took approximately 4 days and 9.5 hours for both the cased and uncased model
60