Commits · 240236a6882ec18f514be457f0d561789ae854ee · Sciencebeam / Sciencebeam Gym

Jan 16, 2018
- check file list · 240236a6
  Daniel Ecer authored 7 years ago
  
  240236a6
- infer base path from file list · ae473710
  Daniel Ecer authored 7 years ago
  
  ae473710
- skip documents that couldn't be loaded for crf training · 546c8227
  Daniel Ecer authored 7 years ago
  
  546c8227
Jan 15, 2018
- moved get_output_file utility function · 687d4ca3
  Daniel Ecer authored 7 years ago
  
  687d4ca3
- return structured document · 89f66660
  Daniel Ecer authored 7 years ago
  
  89f66660
Jan 12, 2018
- added utility transforms to work with files · 62ef45ce
  Daniel Ecer authored 7 years ago
  
  62ef45ce
Jan 11, 2018
- moved process_sciencebeam_gym_dep_args to sciencebeam-gym · c1b52eb6
  Daniel Ecer authored 7 years ago
  
  c1b52eb6
Jan 10, 2018
- Merge pull request #17 from elifesciences/crf-model-with-cv-input · aebc5368
  Daniel Ecer authored 7 years ago
```
CRF model with CV predictions as an input
```
  aebc5368
- updated readme to reflect change to default ext · f4c6cafe
  Daniel Ecer authored 7 years ago
  
  f4c6cafe
- updated readme with cv predictions vs crf model · 673fd1dd
  Daniel Ecer authored 7 years ago
  
  673fd1dd
- optionally use cv predictions when predicting using the crf model · 10675447
  Daniel Ecer authored 7 years ago
  
  10675447
- don't override tag with empty tag · 072a625b
  Daniel Ecer authored 7 years ago
  
  072a625b
- remove tag if passed in value is None · 5a5150a7
  Daniel Ecer authored 7 years ago
  
  5a5150a7
- log crf model training timings · 32ee96dc
  Daniel Ecer authored 7 years ago
  
  32ee96dc
- added multi threading when loading files · 83506e51
  Daniel Ecer authored 7 years ago
  
  83506e51
- added loading progress · b15a4eec
  Daniel Ecer authored 7 years ago
  
  b15a4eec
- train crf model using cv output · e1c30690
  Daniel Ecer authored 7 years ago
  
  e1c30690
- extract scoped_tags as features · 40d0a405
  Daniel Ecer authored 7 years ago
  
  40d0a405
- merge document tags by using separate scopes (prefixes) · de357014
  Daniel Ecer authored 7 years ago
  
  de357014
Jan 09, 2018
- Merge pull request #16 from elifesciences/fix-output-file-ext-for-pdf-gz · 6f228924
  Daniel Ecer authored 7 years ago
```
fix output file extension for .pdf.gz did not remove .pdf
```
  6f228924
- fix output file extension for .pdf.gz did not remove .pdf · 989dc350
  Daniel Ecer authored 7 years ago
  
  989dc350
- Merge pull request #15 from elifesciences/crf-model · 3f862d94
  Daniel Ecer authored 7 years ago
```
CRF model
```
  3f862d94
- added CRF model documentation · a63a7ed7
  Daniel Ecer authored 7 years ago
  
  a63a7ed7
- added get_output_files to get a list of preprocessing output files · 7989709b
  Daniel Ecer authored 7 years ago
  
  7989709b
- removed unused imports · 635e19be
  Daniel Ecer authored 7 years ago
  
  635e19be
Jan 08, 2018
- optionally limit page range · 5275a4b6
  Daniel Ecer authored 7 years ago
  
  5275a4b6
- added main to annotation of lxml document using crf model; ignore 'none tag' · a7c9dd7c
  Daniel Ecer authored 7 years ago
  
  a7c9dd7c
- added optional limit to crf trainer · f5e69e91
  Daniel Ecer authored 7 years ago
  
  f5e69e91
- improved debugging by raising exception with filename · 3528a3dc
  Daniel Ecer authored 7 years ago
  
  3528a3dc
- fixed should use root node rather than parsed tree · 8c5dfb3a
  Daniel Ecer authored 7 years ago
  
  8c5dfb3a
- annotate document using predictions · 409de7f0
  Daniel Ecer authored 7 years ago
  
  409de7f0
- removed unused imports · dff1ce29
  Daniel Ecer authored 7 years ago
  
  dff1ce29
- added crf model · b5145666
  Daniel Ecer authored 7 years ago
  
  b5145666
Jan 04, 2018
- Merge pull request #14 from elifesciences/embed-color-map-into-cv-model · a6380981
  Daniel Ecer authored 7 years ago
```
embed color map into saved model
```
  a6380981
- embed color map into saved model · 83787957
  Daniel Ecer authored 7 years ago
  
  83787957
- Merge pull request #13 from elifesciences/extract-from-annotated-document · 291f39e1
  Daniel Ecer authored 7 years ago
```
extract from annotated document
```
  291f39e1
Jan 03, 2018
- added .eggs to gitignore · 76a9199d
  Daniel Ecer authored 7 years ago
  
  76a9199d
Jan 02, 2018
- don't combine the title from non-consecutive fragments · a6ca81c2
  Daniel Ecer authored 7 years ago
  
  a6ca81c2
- allow emty job suffix · d64cb869
  Daniel Ecer authored 7 years ago
  
  d64cb869
- removed invalid cloud argument · a72f7397
  Daniel Ecer authored 7 years ago
  
  a72f7397