macedonizer
commited on
Commit
•
af91809
1
Parent(s):
b0d62ce
Update README.md
Browse files
README.md
CHANGED
@@ -1,21 +1,20 @@
|
|
1 |
---
|
2 |
language:
|
3 |
-
-
|
4 |
-
thumbnail: https://huggingface.co/macedonizer/
|
5 |
license: Apache 2.0
|
6 |
datasets:
|
7 |
-
- wiki-
|
8 |
-
- time-mk-news-2010-2015
|
9 |
---
|
10 |
|
11 |
-
#
|
12 |
Test the whole generation capabilities here: https://transformer.huggingface.co/doc/gpt2-large
|
13 |
Pretrained model on English language using a causal language modeling (CLM) objective. It was introduced in
|
14 |
[this paper](https://d4mucfpksywv.cloudfront.net/better-language-models/language_models_are_unsupervised_multitask_learners.pdf)
|
15 |
and first released at [this page](https://openai.com/blog/better-language-models/).
|
16 |
|
17 |
## Model description
|
18 |
-
|
19 |
means it was pretrained on the raw texts only, with no humans labelling them in any way (which is why it can use lots
|
20 |
of publicly available data) with an automatic process to generate inputs and labels from those texts. More precisely,
|
21 |
it was trained to guess the next word in sentences.
|
@@ -32,10 +31,10 @@ Here is how to use this model to get the features of a given text in PyTorch:
|
|
32 |
import random
|
33 |
from transformers import AutoTokenizer, AutoModelWithLMHead
|
34 |
|
35 |
-
tokenizer = AutoTokenizer.from_pretrained('macedonizer/
|
36 |
-
model = AutoModelWithLMHead.from_pretrained('macedonizer/
|
37 |
|
38 |
-
input_text = '
|
39 |
|
40 |
if len(input_text) == 0: \
|
41 |
encoded_input = tokenizer(input_text, return_tensors="pt") \
|
@@ -59,8 +58,7 @@ else: \
|
|
59 |
num_return_sequences=1, \
|
60 |
)
|
61 |
|
62 |
-
decoded_output = [] \
|
63 |
-
for sample in output: \
|
64 |
decoded_output.append(tokenizer.decode(sample, skip_special_tokens=True))
|
65 |
|
66 |
print(decoded_output)
|
1 |
---
|
2 |
language:
|
3 |
+
- sl
|
4 |
+
thumbnail: https://huggingface.co/macedonizer/mkgpt2/lets-talk-about-nlp.jpg
|
5 |
license: Apache 2.0
|
6 |
datasets:
|
7 |
+
- wiki-sl
|
|
|
8 |
---
|
9 |
|
10 |
+
# sl-gpt2
|
11 |
Test the whole generation capabilities here: https://transformer.huggingface.co/doc/gpt2-large
|
12 |
Pretrained model on English language using a causal language modeling (CLM) objective. It was introduced in
|
13 |
[this paper](https://d4mucfpksywv.cloudfront.net/better-language-models/language_models_are_unsupervised_multitask_learners.pdf)
|
14 |
and first released at [this page](https://openai.com/blog/better-language-models/).
|
15 |
|
16 |
## Model description
|
17 |
+
sl-gpt2 is a transformers model pretrained on a very large corpus of Slovenian data in a self-supervised fashion. This
|
18 |
means it was pretrained on the raw texts only, with no humans labelling them in any way (which is why it can use lots
|
19 |
of publicly available data) with an automatic process to generate inputs and labels from those texts. More precisely,
|
20 |
it was trained to guess the next word in sentences.
|
31 |
import random
|
32 |
from transformers import AutoTokenizer, AutoModelWithLMHead
|
33 |
|
34 |
+
tokenizer = AutoTokenizer.from_pretrained('macedonizer/sl-gpt2') \
|
35 |
+
model = AutoModelWithLMHead.from_pretrained('macedonizer/sl-gpt2')
|
36 |
|
37 |
+
input_text = 'Ljubljana '
|
38 |
|
39 |
if len(input_text) == 0: \
|
40 |
encoded_input = tokenizer(input_text, return_tensors="pt") \
|
58 |
num_return_sequences=1, \
|
59 |
)
|
60 |
|
61 |
+
decoded_output = [] \\nfor sample in output: \
|
|
|
62 |
decoded_output.append(tokenizer.decode(sample, skip_special_tokens=True))
|
63 |
|
64 |
print(decoded_output)
|