File size: 2,704 Bytes
68499bf af0f99b 789f267 f6aa3fc 68499bf 2b5f963 68499bf 2b5f963 6d66f92 2b5f963 ada5cc0 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 |
---
language: en
license: apache-2.0
datasets:
- sst2
- glue
model-index:
- name: distilbert-base-uncased-finetuned-sst-2-english
results:
- task:
type: text-classification
name: Text Classification
dataset:
name: glue
type: glue
config: sst2
split: validation
metrics:
- name: Accuracy
type: accuracy
value: 0.9105504587155964
verified: true
- name: Precision
type: precision
value: 0.8978260869565218
verified: true
- name: Recall
type: recall
value: 0.9301801801801802
verified: true
- name: AUC
type: auc
value: 0.9716626673402374
verified: true
- name: F1
type: f1
value: 0.9137168141592922
verified: true
- name: loss
type: loss
value: 0.39013850688934326
verified: true
---
# DistilBERT base uncased finetuned SST-2
This model is a fine-tune checkpoint of [DistilBERT-base-uncased](https://huggingface.co/distilbert-base-uncased), fine-tuned on SST-2.
This model reaches an accuracy of 91.3 on the dev set (for comparison, Bert bert-base-uncased version reaches an accuracy of 92.7).
For more details about DistilBERT, we encourage users to check out [this model card](https://huggingface.co/distilbert-base-uncased).
# Fine-tuning hyper-parameters
- learning_rate = 1e-5
- batch_size = 32
- warmup = 600
- max_seq_length = 128
- num_train_epochs = 3.0
# Bias
Based on a few experimentations, we observed that this model could produce biased predictions that target underrepresented populations.
For instance, for sentences like `This film was filmed in COUNTRY`, this binary classification model will give radically different probabilities for the positive label depending on the country (0.89 if the country is France, but 0.08 if the country is Afghanistan) when nothing in the input indicates such a strong semantic shift. In this [colab](https://colab.research.google.com/gist/ageron/fb2f64fb145b4bc7c49efc97e5f114d3/biasmap.ipynb), [Aurélien Géron](https://twitter.com/aureliengeron) made an interesting map plotting these probabilities for each country.
<img src="https://huggingface.co/distilbert-base-uncased-finetuned-sst-2-english/resolve/main/map.jpeg" alt="Map of positive probabilities per country." width="500"/>
We strongly advise users to thoroughly probe these aspects on their use-cases in order to evaluate the risks of this model. We recommend looking at the following bias evaluation datasets as a place to start: [WinoBias](https://huggingface.co/datasets/wino_bias), [WinoGender](https://huggingface.co/datasets/super_glue), [Stereoset](https://huggingface.co/datasets/stereoset).
|