Hugging Face's logo

lostelf
/
scibert_scivocab_uncased_scicite_finetuned

Safetensors
English
bert
scicite
citation-intent
scientific-text
sequence-classification
Eval Results (legacy)
Model card Files
xet
 Community
lostelf commited on
Commit
d6fc4db
·
verified ·
1 Parent(s): 8fb5aa5

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +16 -16
README.md CHANGED
@@ -18,10 +18,10 @@ model-index:
18
  metrics:
19
  - name: Accuracy
20
  type: accuracy
21
- value: 0.8549
22
  - name: Macro F1
23
  type: f1
24
- value: 0.8418
25
  ---
26
 
27
  # SciBERT Fine-tuned for SciCite Intent Classification
@@ -45,9 +45,9 @@ Achieved on the SciCite test set:
45
 
46
  | Metric | Score |
47
  |--------|-------|
48
- | Accuracy | 85.49% |
49
- | Macro F1 | 0.8418 |
50
- | Weighted F1 | 0.8556 |
51
 
52
  **Per-class performance:**
53
 
@@ -130,21 +130,21 @@ print([labels[p] for p in predictions])
130
 
131
  ## Training Results
132
 
133
- Best checkpoint: Epoch 8 (macro F1 = 0.8418). Early stopping patience = 4 epochs.
134
 
135
  Training curve (eval epochs):
136
 
137
  ```
138
  Epoch | Train Loss | Val Loss | Macro F1 | Micro F1
139
  ------|------------|----------|----------|----------
140
- 1 | N/A | 0.41499194502830505 | 0.8109638930380094 | 0.8373362445414847
141
- 2 | N/A | 0.42990684509277344 | 0.8176419374791816 | 0.8427947598253275
142
- 3 | N/A | 0.4241107404232025 | 0.8289871411189474 | 0.8504366812227074
143
- 4 | N/A | 0.44976210594177246 | 0.829161648447125 | 0.8504366812227074
144
- 5 | N/A | 0.48823821544647217 | 0.8310941963882211 | 0.8537117903930131
145
- 6 | N/A | 0.5136175155639648 | 0.8314366485799795 | 0.8526200873362445
146
- 7 | N/A | 0.5481893420219421 | 0.8361351217088923 | 0.8558951965065502
147
- 8 | N/A | 0.5620850324630737 | 0.8295385389161322 | 0.851528384279476
148
  ```
149
 
150
  ## Framework Versions
@@ -157,5 +157,5 @@ Epoch | Train Loss | Val Loss | Macro F1 | Micro F1
157
 
158
  ---
159
 
160
- **Generated:** 2026-04-14 22:17:11
161
- **Training GPU:** 0
 
18
  metrics:
19
  - name: Accuracy
20
  type: accuracy
21
+ value: 0.8560
22
  - name: Macro F1
23
  type: f1
24
+ value: 0.8431
25
  ---
26
 
27
  # SciBERT Fine-tuned for SciCite Intent Classification
 
45
 
46
  | Metric | Score |
47
  |--------|-------|
48
+ | Accuracy | 85.60% |
49
+ | Macro F1 | 0.8431 |
50
+ | Weighted F1 | 0.8566 |
51
 
52
  **Per-class performance:**
53
 
 
130
 
131
  ## Training Results
132
 
133
+ Best checkpoint: Epoch 8 (macro F1 = 0.8431). Early stopping patience = 4 epochs.
134
 
135
  Training curve (eval epochs):
136
 
137
  ```
138
  Epoch | Train Loss | Val Loss | Macro F1 | Micro F1
139
  ------|------------|----------|----------|----------
140
+ 1 | 0.7104 | 0.4155 | 0.8143 | 0.8395
141
+ 2 | 0.5514 | 0.4304 | 0.8176 | 0.8428
142
+ 3 | 0.4605 | 0.4256 | 0.8288 | 0.8504
143
+ 4 | 0.3829 | 0.4514 | 0.8310 | 0.8515
144
+ 5 | 0.3176 | 0.4908 | 0.8311 | 0.8537
145
+ 6 | 0.2678 | 0.5162 | 0.8334 | 0.8548
146
+ 7 | 0.2288 | 0.5507 | 0.8351 | 0.8548
147
+ 8 | N/A | 0.5648 | 0.8275 | 0.8493
148
  ```
149
 
150
  ## Framework Versions
 
157
 
158
  ---
159
 
160
+ **Generated:** 2026-04-14 22:24:37
161
+ **Training GPU:** 2