ssg1 commited on
Commit
24cd098
·
verified ·
1 Parent(s): 709e76a

Add the model card

Browse files
Files changed (1) hide show
  1. README.md +3 -4
README.md CHANGED
@@ -28,15 +28,14 @@ Labels: Dermatology, Orthopedics, ENT, Gastroenterology, Neurology, Urology, Ob-
28
 
29
  Stage 1 only: 6,252 unique comments from *Patient Comments and Specialist Types*
30
  (Mendeley Data, DOI 10.17632/2twgjzpn82.2, CC BY 4.0), with its 68 symptom categories
31
- remapped to the labels above. These are short, one-sentence comments. They appear to be
32
- generated rather than written by real patients (emoji matched to each symptom, inserted
33
- typos), so treat results on this set as a sanity check only.
34
 
35
  ## Results
36
 
37
  On 938 held-out comments from that same public set, the fine-tuned router scores
38
  94.3% top-1 (95% CI 92.7%–95.7%), 99.0% top-3 and
39
- 0.908 macro-F1. That set is short and templated, so this is an easy score:
40
  it says how well the router fits the public data, not how it does on real posts.
41
 
42
  ## Limits
 
28
 
29
  Stage 1 only: 6,252 unique comments from *Patient Comments and Specialist Types*
30
  (Mendeley Data, DOI 10.17632/2twgjzpn82.2, CC BY 4.0), with its 68 symptom categories
31
+ remapped to the labels above. These are short, one-sentence comments, so treat results
32
+ on this set as a sanity check only.
 
33
 
34
  ## Results
35
 
36
  On 938 held-out comments from that same public set, the fine-tuned router scores
37
  94.3% top-1 (95% CI 92.7%–95.7%), 99.0% top-3 and
38
+ 0.908 macro-F1. The comments are short and similar to each other, so this is an easy score:
39
  it says how well the router fits the public data, not how it does on real posts.
40
 
41
  ## Limits