system HF staff commited on
Commit
fa895f5
1 Parent(s): 686e8de

Update .ipynb_checkpoints/README-checkpoint.md

Browse files
.ipynb_checkpoints/README-checkpoint.md ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ thumbnail: https://avatars3.githubusercontent.com/u/32437151?s=460&u=4ec59abc8d21d5feea3dab323d23a5860e6996a4&v=4
5
+ tags:
6
+ - text-classification
7
+ - emotion
8
+ - pytorch
9
+ license: apache-2.0
10
+ datasets:
11
+ - emotion
12
+ metrics:
13
+ - accuracy
14
+ ---
15
+
16
+ # bert-base-uncased-emotion
17
+
18
+ ## Model description
19
+
20
+ `bert-base-uncased` finetuned on the emotion dataset using PyTorch Lightning. Sequence length 128, learning rate 2e-5, batch size 32, 2 GPUs, 4 epochs.
21
+
22
+ For more details, please see, [the emotion dataset on nlp viewer](https://huggingface.co/nlp/viewer/?dataset=emotion).
23
+
24
+
25
+ #### Limitations and bias
26
+
27
+ - Not the best model, but it works in a pinch I guess...
28
+ - Code not available as I just hacked this together.
29
+ - [Follow me on github](https://github.com/nateraw) to get notified when code is made available.
30
+
31
+ ## Training data
32
+
33
+ Data came from HuggingFace's `datasets` package. The data can be viewed [on nlp viewer](https://huggingface.co/nlp/viewer/?dataset=emotion).
34
+
35
+
36
+ ## Training procedure
37
+ ...
38
+
39
+ ## Eval results
40
+
41
+ val_acc - 0.931 (useless, as this should be precision/recall/f1)
42
+
43
+ The score was calculated using PyTorch Lightning metrics.