issues
search
ybracke
/
transnormer
A lexical normalizer for historical spelling variants using a transformer architecture.
GNU General Public License v3.0
6
stars
1
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Add separate module for exchanging the transliteration component of an hf tokenizer
#43
ybracke
closed
1 year ago
0
Visualization/Analysis: Display tokenization of input string in notebooks for inspection of predictions
#42
ybracke
closed
1 year ago
1
Cache trouble
#41
ybracke
closed
1 year ago
2
Visualization: Add option to change scale type (e.g. linear, log) for loss plot
#40
ybracke
opened
1 year ago
0
Add data loading for DTA JSONL files
#39
ybracke
closed
1 year ago
0
Handling Named Entities
#38
ybracke
opened
1 year ago
2
Update main with changes from dev
#37
ybracke
closed
1 year ago
0
Improve issues
#36
ybracke
closed
1 year ago
0
Data: Add CAB-normalized versions of DTA as training data
#35
ybracke
closed
1 year ago
1
Add functions for unknown tokens
#34
ybracke
closed
1 year ago
0
Add GitHub workflows
#33
ybracke
closed
1 year ago
0
New data loading in training module
#32
ybracke
closed
1 year ago
0
Custom tokenizer as a separate module
#31
ybracke
closed
1 year ago
0
Further development of `merge_datasets`
#30
ybracke
opened
1 year ago
0
Convert data to JSON Lines
#29
ybracke
closed
1 year ago
0
For notebooks: Configure git so as to track only changes in code and output
#28
ybracke
closed
11 months ago
0
A function to concatenate and resample multiple `Dataset`s
#27
ybracke
closed
1 year ago
2
Script for creating an intermediate version of the data (jsonl) for faster loading
#26
ybracke
closed
1 year ago
0
Loading functionality for GerManC-GS
#25
ybracke
opened
1 year ago
1
Dev refactor loading
#24
ybracke
closed
1 year ago
0
Character-wise diff of predicted and correct normalization + visualization
#23
ybracke
opened
1 year ago
1
Make interactive annotation available in notebooks
#22
ybracke
opened
1 year ago
0
Dev refactor training
#21
ybracke
closed
1 year ago
0
Applying the functions in a notebook
#20
ybracke
closed
1 year ago
0
Change datapaths from absolute to relative
#19
ybracke
closed
1 year ago
1
Refactor data loading
#18
ybracke
closed
1 year ago
1
Load training data from XML version DTA EvalCorpus and keep joined tokens
#17
ybracke
closed
1 year ago
1
Add github workflow
#16
ybracke
closed
1 year ago
1
Tokenizer consistency
#15
ybracke
closed
3 months ago
0
Use publication date as input?
#14
ybracke
opened
1 year ago
0
Rework `transnormer/models/model_train.py`
#13
ybracke
closed
1 year ago
1
Experiment with a custom loss-function
#12
ybracke
opened
1 year ago
0
Experiment with a fine-tuned encoder
#11
ybracke
opened
1 year ago
0
Experiment with a fine-tuned decoder
#10
ybracke
opened
1 year ago
0
Experiment with randomly initialized decoder
#9
ybracke
opened
1 year ago
1
Remove `codecarbon` from dependencies
#8
ybracke
closed
1 year ago
1
Move hyperparameters into parameter config file
#7
ybracke
closed
1 year ago
0
Use DVC pipeline
#6
ybracke
closed
1 year ago
1
Create explorative functionality: Plot attention matrix
#3
ybracke
opened
1 year ago
0
Notebook: Manually inspect predictions per time period
#2
ybracke
closed
1 year ago
1
Create explorative functionality: Plot loss
#1
ybracke
closed
1 year ago
0
Previous