Total runs:
1.6M
Run Growth:
2.7K
Growth Rate:
0.18%
dbmdz / convbert-base-turkish-mc4-uncased
Model's Last Updated: September 11 2023
fill-mask
Introduction of convbert-base-turkish-mc4-uncased
Model Details of convbert-base-turkish-mc4-uncased
🇹🇷 Turkish ConvBERT model
We present community-driven BERT, DistilBERT, ELECTRA and ConvBERT models for Turkish 🎉
Some datasets used for pretraining and evaluation are contributed from the awesome Turkish NLP community, as well as the decision for the BERT model name: BERTurk.
Logo is provided by Merve Noyan .
Stats
We've trained an (uncased) ConvBERT model on the recently released Turkish part of the multiligual C4 (mC4) corpus from the AI2 team.
After filtering documents with a broken encoding, the training corpus has a size of 242GB resulting in 31,240,963,926 tokens.
We used the original 32k vocab (instead of creating a new one).
mC4 ConvBERT
In addition to the ELEC TR A base model, we also trained an ConvBERT model on the Turkish part of the mC4 corpus. We use a sequence length of 512 over the full training time and train the model for 1M steps on a v3-32 TPU.
Model usage
All trained models can be used from the DBMDZ Hugging Face model hub page using their model name.
Example usage with 🤗/Transformers:
tokenizer = AutoTokenizer.from_pretrained("dbmdz/convbert-base-turkish-mc4-uncased")
model = AutoModel.from_pretrained("dbmdz/convbert-base-turkish-mc4-uncased")
Citation
You can use the following BibTeX entry for citation:
@software{stefan_schweter_2020_3770924,
author = {Stefan Schweter},
title = {BERTurk - BERT models for Turkish},
month = apr,
year = 2020,
publisher = {Zenodo},
version = {1.0.0},
doi = {10.5281/zenodo.3770924},
url = {https://doi.org/10.5281/zenodo.3770924}
}
Acknowledgments
Thanks to Kemal Oflazer for providing us additional large corpora for Turkish. Many thanks to Reyyan Yeniterzi for providing us the Turkish NER dataset for evaluation.
We would like to thank Merve Noyan for the awesome logo!
Research supported with Cloud TPUs from Google's TensorFlow Research Cloud (TFRC). Thanks for providing access to the TFRC ❤️
Runs of dbmdz convbert-base-turkish-mc4-uncased on huggingface.co
451
Total runs
0
24-hour runs
-19
3-day runs
-339
7-day runs
-1.3K
30-day runs
More Information About convbert-base-turkish-mc4-uncased huggingface.co Model
Url of convbert-base-turkish-mc4-uncased
convbert-base-turkish-mc4-uncased huggingface.co Url
Provider of convbert-base-turkish-mc4-uncased huggingface.co
Other API from dbmdz
Updated: September 07 2023
Total runs:
99.7K
Run Growth:
21.8K
Growth Rate:
21.90%
Total runs:
80.9K
Run Growth:
12.1K
Growth Rate:
14.89%
Total runs:
76.0K
Run Growth:
48.3K
Growth Rate:
63.51%
Total runs:
46.0K
Run Growth:
40.6K
Growth Rate:
88.22%
Total runs:
45.5K
Run Growth:
9.4K
Growth Rate:
20.76%
Total runs:
39.3K
Run Growth:
27.1K
Growth Rate:
69.17%
Total runs:
30.9K
Run Growth:
13.8K
Growth Rate:
44.72%
Total runs:
29.0K
Run Growth:
8.5K
Growth Rate:
29.17%
Total runs:
9.7K
Run Growth:
2.9K
Growth Rate:
29.27%
Total runs:
9.4K
Run Growth:
4.4K
Growth Rate:
46.79%
Total runs:
4.5K
Run Growth:
-212
Growth Rate:
-4.73%
Total runs:
4.1K
Run Growth:
1.6K
Growth Rate:
38.34%
Total runs:
1.8K
Run Growth:
-1.5K
Growth Rate:
-84.15%
Total runs:
1.7K
Run Growth:
225
Growth Rate:
13.12%
Total runs:
1.3K
Run Growth:
552
Growth Rate:
43.91%
Total runs:
817
Run Growth:
281
Growth Rate:
34.39%
Total runs:
599
Run Growth:
489
Growth Rate:
81.64%
Total runs:
559
Run Growth:
-64
Growth Rate:
-11.45%
Total runs:
338
Run Growth:
241
Growth Rate:
71.30%
Total runs:
309
Run Growth:
-20
Growth Rate:
-6.47%
Total runs:
283
Run Growth:
266
Growth Rate:
93.99%
Total runs:
282
Run Growth:
44
Growth Rate:
15.60%
Total runs:
279
Run Growth:
239
Growth Rate:
85.66%
Total runs:
269
Run Growth:
216
Growth Rate:
80.30%
Total runs:
267
Run Growth:
-418
Growth Rate:
-156.55%
Total runs:
213
Run Growth:
57
Growth Rate:
26.76%
Total runs:
206
Run Growth:
-87
Growth Rate:
-42.23%
Total runs:
204
Run Growth:
181
Growth Rate:
88.73%
Total runs:
185
Run Growth:
167
Growth Rate:
90.27%
Total runs:
182
Run Growth:
116
Growth Rate:
63.74%
Total runs:
160
Run Growth:
138
Growth Rate:
86.25%
Total runs:
142
Run Growth:
116
Growth Rate:
81.69%
Total runs:
141
Run Growth:
129
Growth Rate:
91.49%
Total runs:
130
Run Growth:
121
Growth Rate:
93.08%
Total runs:
125
Run Growth:
79
Growth Rate:
63.20%
Total runs:
124
Run Growth:
107
Growth Rate:
86.29%
Total runs:
123
Run Growth:
112
Growth Rate:
91.06%
Total runs:
116
Run Growth:
112
Growth Rate:
96.55%
Total runs:
116
Run Growth:
106
Growth Rate:
91.38%
Total runs:
112
Run Growth:
110
Growth Rate:
98.21%
Total runs:
110
Run Growth:
95
Growth Rate:
86.36%
Total runs:
110
Run Growth:
70
Growth Rate:
63.64%
Total runs:
110
Run Growth:
108
Growth Rate:
98.18%
Total runs:
109
Run Growth:
101
Growth Rate:
92.66%
Total runs:
108
Run Growth:
56
Growth Rate:
51.85%
Total runs:
108
Run Growth:
103
Growth Rate:
95.37%
Total runs:
37
Run Growth:
25
Growth Rate:
67.57%
Total runs:
28
Run Growth:
21
Growth Rate:
75.00%
Total runs:
18
Run Growth:
9
Growth Rate:
50.00%
Total runs:
17
Run Growth:
10
Growth Rate:
58.82%
Total runs:
16
Run Growth:
4
Growth Rate:
25.00%
Total runs:
9
Run Growth:
8
Growth Rate:
88.89%
Total runs:
7
Run Growth:
5
Growth Rate:
71.43%
Total runs:
3
Run Growth:
2
Growth Rate:
66.67%
Total runs:
2
Run Growth:
-1
Growth Rate:
-50.00%
Total runs:
2
Run Growth:
1
Growth Rate:
50.00%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%
Total runs:
0
Run Growth:
0
Growth Rate:
0.00%