Commit
•
6a4f769
1
Parent(s):
144e7cb
Update README.md
Browse files
README.md
CHANGED
@@ -15,13 +15,13 @@ tags:
|
|
15 |
---
|
16 |
|
17 |
# Huggingface Model: BART-MNLI-ZeroShot-Text-Classification
|
18 |
-
This is a Huggingface model fine-tuned on the CNN news dataset for zero-shot text classification task using
|
19 |
|
20 |
## Authors
|
21 |
This work was done by [CHERGUELAINE Ayoub](https://www.linkedin.com/in/ayoub-cherguelaine/) & [BOUBEKRI Faycal](https://www.linkedin.com/in/faycal-boubekri-832848199/)
|
22 |
|
23 |
## Model Architecture
|
24 |
-
The model architecture is based on the
|
25 |
|
26 |
## Dataset
|
27 |
The CNN news dataset was used for fine-tuning the model. This dataset contains news articles from the CNN website and is labeled into 6 categories, including politics, health, entertainment, tech, travel, world, and sports.
|
@@ -30,7 +30,7 @@ The CNN news dataset was used for fine-tuning the model. This dataset contains n
|
|
30 |
The model was fine-tuned for 1 epoch on a maximum length of 256 tokens. The training took approximately 6 hours to complete.
|
31 |
|
32 |
## Evaluation Metrics
|
33 |
-
The model achieved an f1 score of
|
34 |
|
35 |
# Usage
|
36 |
The model can be used for zero-shot text classification tasks on news articles. It can be accessed via the Huggingface Transformers library using the following code:
|
|
|
15 |
---
|
16 |
|
17 |
# Huggingface Model: BART-MNLI-ZeroShot-Text-Classification
|
18 |
+
This is a Huggingface model fine-tuned on the CNN news dataset for zero-shot text classification task using DistilBART-MNLI. The model achieved an f1 score of 93% and an accuracy of 93% on the CNN test dataset with a maximum length of 128 tokens.
|
19 |
|
20 |
## Authors
|
21 |
This work was done by [CHERGUELAINE Ayoub](https://www.linkedin.com/in/ayoub-cherguelaine/) & [BOUBEKRI Faycal](https://www.linkedin.com/in/faycal-boubekri-832848199/)
|
22 |
|
23 |
## Model Architecture
|
24 |
+
The model architecture is based on the DistilBART-MNLI transformer model. DistilBART is a smaller and faster version of BART that is pre-trained on a large corpus of text and fine-tuned on downstream natural language processing tasks.
|
25 |
|
26 |
## Dataset
|
27 |
The CNN news dataset was used for fine-tuning the model. This dataset contains news articles from the CNN website and is labeled into 6 categories, including politics, health, entertainment, tech, travel, world, and sports.
|
|
|
30 |
The model was fine-tuned for 1 epoch on a maximum length of 256 tokens. The training took approximately 6 hours to complete.
|
31 |
|
32 |
## Evaluation Metrics
|
33 |
+
The model achieved an f1 score of 93% and an accuracy of 93% on the CNN test dataset with a maximum length of 128 tokens.
|
34 |
|
35 |
# Usage
|
36 |
The model can be used for zero-shot text classification tasks on news articles. It can be accessed via the Huggingface Transformers library using the following code:
|