issues
search
nasa-petal
/
PeTaL-labeller
The PeTaL labeler labels journal articles with biomimicry functions.
https://petal-labeller.readthedocs.io/en/latest/
The Unlicense
6
stars
4
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Add files via upload
#89
ARalevski
closed
2 years ago
0
Preprocess_golden.py
#88
pjuangph
closed
1 year ago
0
Update README.md
#87
pjuangph
opened
2 years ago
0
Match vocab file PeTaL.emb
#86
pjuangph
opened
2 years ago
1
taxonomy update
#85
dsmith111
closed
2 years ago
0
Explore tokenizer optimization strategies to improve precision/recall.
#84
bruffridge
opened
3 years ago
0
Change output of labeller from absolute "selection" to "ranking" using confidence scores.
#83
bruffridge
closed
2 years ago
1
Match with petal
#82
elkong
closed
3 years ago
0
Try using a tree of multilabel classifiers
#81
bruffridge
opened
3 years ago
1
Look into Label Studio to help increase the size of our labelled dataset for training
#80
bruffridge
opened
3 years ago
0
Golden Gate
#79
elkong
closed
3 years ago
0
Try Google Cloud AutoML Natural Language's multilabel text classification on golden dataset
#78
bruffridge
opened
3 years ago
1
Add a description of metrics to MATCH's README
#77
bruffridge
closed
3 years ago
1
try to use a language model like GPT-2 for its general-purpose language understanding capabilities (and then integrate it with the MATCH classification task somehow)
#76
bruffridge
opened
3 years ago
0
See if using an ensemble method with MATCH improves performance.
#75
bruffridge
opened
3 years ago
0
Use a large ensemble of single-label classifiers (so, treat all of the labels independently, ignore the hierarchy, and we have a hundred separate yes/no tasks) and see if this works better than MATCH
#74
bruffridge
opened
3 years ago
0
Run Match on just level 1 labels
#73
bruffridge
closed
3 years ago
4
See how replacing random weights with pretrained and fine-tuned weights in MATCH affects performance
#72
bruffridge
opened
3 years ago
4
auto-labeler/MATCH ought to look better now
#71
elkong
closed
3 years ago
0
Create a multi-label classification model for the most used labels.
#70
bruffridge
closed
2 years ago
0
Create a binary classification filtering model for the top 25% of labels
#69
bruffridge
opened
3 years ago
0
rerun ablation study
#68
bruffridge
closed
3 years ago
1
match, migration, mag, mesh
#67
elkong
closed
3 years ago
7
weights, biases, and ablations
#66
elkong
closed
3 years ago
0
Use open-source Snorkel to create labelling functions to expand our training dataset.
#65
bruffridge
opened
3 years ago
2
MATCH with PeTaL: relevance thresholds, MAG and MeSH terms, flat taxonomy
#64
elkong
closed
3 years ago
0
52 analyze data
#63
dsmith111
closed
3 years ago
0
Add 'leaf only' option to lensCleaner
#62
dsmith111
closed
3 years ago
0
Produce metrics to show which labels are being classified correctly and which aren't, and how they're being misclassified.
#61
bruffridge
closed
3 years ago
5
Compare performance of only including leaf labels in the dataset.
#60
bruffridge
closed
3 years ago
1
Replace MAG topics with another topic taxonomy
#59
bruffridge
closed
3 years ago
1
Do k-fold cross validation to generate ablation study results for including MAG and MeSH labels.
#58
bruffridge
closed
3 years ago
1
cleaned up and added work and analysis from Colab and GRC HPC
#57
elkong
closed
3 years ago
0
Compare MATCH results to auto-sklearn
#56
bruffridge
opened
3 years ago
1
Look into using a relevancy threshold vs. top k for labelling
#55
bruffridge
closed
3 years ago
1
More DOIs
#54
dsmith111
closed
3 years ago
0
Does adding MeSH terms and/or MAG fields of study improve accuracy?
#53
bruffridge
closed
3 years ago
3
Analyze our training dataset to discover ways to improve it to improve labelling accuracy
#52
bruffridge
closed
3 years ago
1
Add DOI search
#51
dsmith111
closed
3 years ago
3
Only 409 papers with labels in cleaned_lens_output.json, should be 701
#50
bruffridge
closed
3 years ago
1
Taxonomy format
#49
dsmith111
closed
3 years ago
0
Verify input format expected by MATCH for unlabelled papers
#48
bruffridge
closed
3 years ago
3
Create a label hierarchy for MATCH input
#47
bruffridge
closed
3 years ago
0
Python MATCH data preparation from Lens
#46
dsmith111
closed
3 years ago
2
Build a data pipeline for running unlabelled papers through the labeller
#45
bruffridge
opened
3 years ago
5
Create a POC for integrating a Colab classification model with weights and biases
#44
bruffridge
closed
3 years ago
0
Prepare train/test data for MATCH
#43
bruffridge
closed
3 years ago
0
Look into using MATCH to improve the labeler
#42
bruffridge
closed
3 years ago
6
Look into how we might use SPECTER to improve our labeller
#41
bruffridge
opened
3 years ago
4
Plot a graph that shows how much adding additional training data improves labeller accuracy.
#40
bruffridge
closed
3 years ago
1
Next