Closed deathsurgeon1 closed 3 years ago
Hi @deathsurgeon1 Please, refer to this example to predict from tflite model. Please, make sure that the first thing in your prediction script is to disable your GPU, because tflite models is not working on PC GPUs.
import os
os.environ["CUDA_VISIBLE_DEVICES"] = "-1"
Try it and let me know.
Hi @deathsurgeon1 Please, refer to this example to predict from tflite model. Please, make sure that the first thing in your prediction script is to disable your GPU, because tflite models is not working on PC GPUs.
import os os.environ["CUDA_VISIBLE_DEVICES"] = "-1"
Try it and let me know.
still not working model loaded log is displayed but after that it keeps on processing without giving any output
@deathsurgeon1 I think this behavior, because still GPU is not disabled. So, please share your OS system details, GPU, CPU, Python version. If you are using colab, share the link. try to enforce using CPU from colab itself.
@deathsurgeon1 I think this behavior, because still GPU is not disabled. So, please share your OS system details, GPU, CPU, Python version. If you are using colab, share the link. try to enforce using CPU from colab itself.
this tflite model loading and prediction was working earlier on colab. Can u verify this once on your side??
Sure, I'll do it may be tonight or tomorrow and let you know, because I don't have access to my machine now :)
@deathsurgeon1 It's working on my machine. Nothing changed! If you want any help, please provide :
So, please share your OS system details, GPU, CPU, Python version. If you are using colab, share the link.
Closed due to: Can't reproduced and no other information provided
nlu_model = TransformerNLU.load(save_path, quantized=True, num_process=1) utterance = "add sabrina salerno to the grime instrumentals playlist" result = nlu_model.predict(utterance) print(result) for item in result['slots']: print(item['value'])
After loading the tflite model, it shows processing only and not giving the prediction for the utterance