Zilun Peng zilunpeng

## deploy_gpt_neo_on_sagemaker.ipynb

      
              1 file
            
          
              1 fork
            
          
              0 comments
            
          
              0 stars
            
          
                zilunpeng
                / deploy_gpt_neo_on_sagemaker.ipynb
            
            
              Last active
              June 23, 2022 18:15
            
              
                example notebook for deploying GPT-Neo in AWS Sagemaker
              
          
      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## gpt_neo_generation.py
"""
Invode the conda environment gpt_neo_generation before running this file.
Specify the prompt in prompt.txt
"""

from transformers import GPTNeoForCausalLM, GPT2Tokenizer
import time

def main():
    start_time = time.time()

## Training_GPT_Neo.ipynb

      
              1 file
            
          
              1 fork
            
          
              0 comments
            
          
              0 stars
            
          
                zilunpeng
                / Training_GPT_Neo.ipynb
            
            
              Last active
              June 23, 2022 18:15
            
              
                example notebook for training GPT-Neo
              
          
      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## GPT_Neo_playground_public.ipynb

      
              1 file
            
          
              1 fork
            
          
              0 comments
            
          
              0 stars
            
          
                zilunpeng
                / GPT_Neo_playground_public.ipynb
            
            
              Last active
              June 23, 2022 18:15
            
              
                example notebook for running GPT-Neo
              
          
      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## run_generation.py
#!/usr/bin/env python
# coding=utf-8
# Copyright 2018 Google AI, Google Brain and Carnegie Mellon University Authors and the HuggingFace Inc. team.
# Copyright (c) 2018, NVIDIA CORPORATION.  All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0

## run_clm.py
#!/usr/bin/env python
# coding=utf-8
# Copyright 2020 The HuggingFace Inc. team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#

## quantized_wav2vec2_perf.csv

          
            Model Name
            WER
            Inference Time
            Model Size
            # of Parameters
            Hardware

            
              wav2vec_big_960h
              2.63%
              4433 s
              1262 mb
              317 M
              CPU (8 cores)

            
              Quantized wav2vec_big_960h
              2.75%
              4079 s
              354 mb
              317 M
              CPU (8 cores)

## student_wav2vec2_perf.csv

          
            Model Name
            # of Parameters
            Model Size
            CPU (8 cores ) Inference Time
            GPU Inference Time
            WER

            
              wav2vec_big_960h
              317 M
              1262 mb
              4433 s
              123 s
              2.63%

            
              Student wav2vec 2.0
              65 M
              262 mb
              1560 s
              51 s
              9.51%

## wav2vec2_perf.csv

          
            Model Name
            WER
            Inference Time
            Model Size
            # of Parameters

            
              wav2vec_big_960h
              2.63%
              123 s
              1262 mb
              317 M

## calc_wer.py
from jiwer import wer
wer_score = wer(ground_truths, predictions)
	"""
	Invode the conda environment gpt_neo_generation before running this file.
	Specify the prompt in prompt.txt
	"""

	from transformers import GPTNeoForCausalLM, GPT2Tokenizer
	import time

	def main():
	start_time = time.time()
	#!/usr/bin/env python
	# coding=utf-8
	# Copyright 2018 Google AI, Google Brain and Carnegie Mellon University Authors and the HuggingFace Inc. team.
	# Copyright (c) 2018, NVIDIA CORPORATION. All rights reserved.
	#
	# Licensed under the Apache License, Version 2.0 (the "License");
	# you may not use this file except in compliance with the License.
	# You may obtain a copy of the License at
	#
	# http://www.apache.org/licenses/LICENSE-2.0
	#!/usr/bin/env python
	# coding=utf-8
	# Copyright 2020 The HuggingFace Inc. team. All rights reserved.
	#
	# Licensed under the Apache License, Version 2.0 (the "License");
	# you may not use this file except in compliance with the License.
	# You may obtain a copy of the License at
	#
	# http://www.apache.org/licenses/LICENSE-2.0
	#
Model Name	WER	Inference Time	Model Size	# of Parameters	Hardware
wav2vec_big_960h	2.63%	4433 s	1262 mb	317 M	CPU (8 cores)
Quantized wav2vec_big_960h	2.75%	4079 s	354 mb	317 M	CPU (8 cores)
	from jiwer import wer
	wer_score = wer(ground_truths, predictions)