tflite-android-transformers icon indicating copy to clipboard operation
tflite-android-transformers copied to clipboard

DistilBERT / GPT-2 for on-device inference thanks to TensorFlow Lite with Android demo apps

Results 16 tflite-android-transformers issues
Sort by recently updated
recently updated
newest added

when I set this: ![image](https://user-images.githubusercontent.com/48509221/176408711-9e36f014-ec1b-4894-9bf5-b38bebe20eb3.png) but when I check the model.I got this: ![image](https://user-images.githubusercontent.com/48509221/176408377-24a1ad49-650a-40f2-b7d4-ac4f3aca7173.png) and I also test the true model,it's right ![image](https://user-images.githubusercontent.com/48509221/176408990-045d735f-0731-4474-ad73-0dd7dd2e36c8.png) how to solve this problem?

I tried to convert tensorflow model to tflite using `model_generation/distilbert.py`. I was able to convert and save the model without error, but could not allocate_tensor with python API and invoke...

When I was clone & run your GPT2 example code, there are some runtime error when try to load model. How can I run this code? - tried tflite version...

Hi, I'd like to download the default distilbert model with the [link](https://s3.amazonaws.com/models.huggingface.co/bert/distilbert-base-uncased-distilled-squad-384.tflite) in [download.gradle](https://github.com/huggingface/tflite-android-transformers/blob/master/bert/download.gradle). It seems the model contains a FlexErf node, which causes failure when I try to run...

Hi, I cannot achieve the speed demonstrated in the gif: https://github.com/huggingface/tflite-android-transformers/tree/master/gpt2 It takes about 7 seconds to generate a single word on my build. I am even using gpuDelegate to...

@Pierrci I used the DistilBERT model with the SST-2 dataset for text classification. I then converted the trained model to TensorFlow Lite using float16 quantization. Here's my [notebook](https://github.com/sayakpaul/BERT-for-Mobile/blob/master/DistilBERT_SST-2_TPU.ipynb). Then when...

Hi, I was wondering how to run the tflite models on Linux with arm architectures (Raspberry Pi) instead of andriod/ios? Is there a transformers library for such systems? Thanks, Mohamed.

For tensorflow>=2.2.0, the model._set_inputs failed in https://github.com/huggingface/tflite-android-transformers/blob/dcd6da1bfb28e3cd6bc83b58a112cdcd3d6cc2fe/models_generation/gpt2.py#L7 , and the model.inputs and model.outpus are all None. The following error will happen when running in Android: ```java.lang.IllegalStateException: Internal error: Unexpected failure...