File size: 1,837 Bytes
092176d |
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 |
---
language: en
tags:
- distilbert
- needmining
license: apache-2.0
metric:
- f1
---
# Finetuned-Distilbert-needmining (uncased)
This model is a finetuned version of the [Distilbert base model](https://huggingface.co/distilbert-base-uncased). It was
trained to predict need-containing sentences from amazon product reviews.
## Model description
This mode is part of ongoing research, after the publication of the research more information will be added.
## Intended uses & limitations
You can use this model to identify sentences that contain customer needs in user-generated content. This can act as a filtering process to remove uninformative content for market research.
### How to use
You can use this model directly with a pipeline for text classification:
```python
>>> from transformers import pipeline
>>> classifier = pipeline("text-classification", model="svenstahlmann/finetuned-distilbert-needmining")
>>> classifier("the plasic feels super cheap.")
[{'label': 'contains need', 'score': 0.9397542476654053}]
```
### Limitations and bias
We are not aware of any bias in the training data.
## Training data
The training was done on a dataset of 6400 sentences. The sentences were taken from product reviews off amazon and coded if they express customer needs.
## Training procedure
For the training, we used [Population Based Training (PBT)](https://www.deepmind.com/blog/population-based-training-of-neural-networks) and optimized for f1 score on a validation set of 1600 sentences.
### Preprocessing
The preprocessing follows the [Distilbert base model](https://huggingface.co/distilbert-base-uncased).
### Pretraining
The model was trained on a titan RTX for 1 hour.
## Evaluation results
Results on the validation set:
| F1 |
|:----:|
| 76.0 |
### BibTeX entry and citation info
coming soon |