| --- |
| language: en |
| license: mit |
| datasets: |
| - glue |
| - mrpc |
| metrics: |
| - f1 |
| tags: |
| - text-classfication |
| - nlp |
| - neural-compressor |
| - PostTrainingDynamic |
| - int8 |
| - Intel® Neural Compressor |
| --- |
| |
| # Dynamically quantized DistilBERT base uncased finetuned MPRC |
|
|
| ## Table of Contents |
| - [Model Details](#model-details) |
| - [How to Get Started With the Model](#how-to-get-started-with-the-model) |
|
|
| ## Model Details |
| **Model Description:** This model is a [DistilBERT](https://huggingface.co/textattack/distilbert-base-uncased-MRPC) fine-tuned on MPRC dynamically quantized with [optimum-intel](https://github.com/huggingface/optimum-intel) through the usage of [huggingface/optimum-intel](https://github.com/huggingface/optimum-intel) through the usage of [Intel® Neural Compressor](https://github.com/intel/neural-compressor). |
| - **Model Type:** Text Classification |
| - **Language(s):** English |
| - **License:** Apache-2.0 |
| - **Parent Model:** For more details on the original model, we encourage users to check out [this](https://huggingface.co/textattack/distilbert-base-uncased-MRPC) model card. |
|
|
| ## How to Get Started With the Model |
|
|
| ### PyTorch |
|
|
| To load the quantized model, you can do as follows: |
|
|
| ```python |
| from optimum.intel.neural_compressor.quantization import IncQuantizedModelForSequenceClassification |
| |
| model = IncQuantizedModelForSequenceClassification.from_pretrained("Intel/distilbert-base-uncased-MRPC-int8-dynamic") |
| ``` |
|
|
| #### Test result |
|
|
| | |INT8|FP32| |
| |---|:---:|:---:| |
| | **Accuracy (eval-f1)** |0.8983|0.9027| |
| | **Model size (MB)** |75|268| |
|
|
