阳阳 hotbaby

## asr_fbank_mfcc.ipynb

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                hotbaby
                / asr_fbank_mfcc.ipynb
            
            
              Created
              March 4, 2024 02:01
                — forked from murphypei/asr_fbank_mfcc.ipynb
            
              
                ASR FBANK MFCC 特征提取
              
          
      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## whisper_distribtued_infer.py
# encoding: utf8

import json
import torch
import argparse
import whisper
from whisper import Whisper, ModelDimensions
from torch.utils.data import Dataset, DataLoader

from lightning import Trainer

## torch_nccl_test.py
# encoding: utf8

import logging
import torch
import torch.distributed
from torch.distributed import ReduceOp


def print_rank_0(msg, *args, **kwargs):
    rank = torch.distributed.get_rank()

## multihead_attention.py
# encoding: utf8

import math
import torch
import torch.nn as nn


class MultiHeadAttention(nn.Module):

    def __init__(self, n_feat, n_head=4):

## pytorch_collective_communication.py
# encoding: utf8

import os
import torch
import torch.distributed as dist
import torch.multiprocessing as mp


def cm_broadcast_object_demo(rank: int, world_size: int):
    dist.init_process_group("gloo", world_size=world_size, rank=rank)

## ddp_demo.py
# encoding: utf8

import os
import time
import random
import contextlib
import torch
import torch.nn as nn
import torch.optim as optim
import torch.distributed as dist

## list2tensor.py
l = [[1, 1675, 4028, 8271, 42, 2], [1, 1675, 4028, 8844, 10135, 10757, 9478, 42, 2]]

# maximum rows and columns
max_rows = len(l)
max_cols = max(len(row) for row in l)

# padded rows
padded = [row + [0] * (max_cols - len(row)) for row in l]

# unpack nested list
	# encoding: utf8

	import json
	import torch
	import argparse
	import whisper
	from whisper import Whisper, ModelDimensions
	from torch.utils.data import Dataset, DataLoader

	from lightning import Trainer
	# encoding: utf8

	import logging
	import torch
	import torch.distributed
	from torch.distributed import ReduceOp


	def print_rank_0(msg, args, *kwargs):
	rank = torch.distributed.get_rank()
	# encoding: utf8

	import math
	import torch
	import torch.nn as nn


	class MultiHeadAttention(nn.Module):

	def __init__(self, n_feat, n_head=4):
	# encoding: utf8

	import os
	import torch
	import torch.distributed as dist
	import torch.multiprocessing as mp


	def cm_broadcast_object_demo(rank: int, world_size: int):
	dist.init_process_group("gloo", world_size=world_size, rank=rank)
	# encoding: utf8

	import os
	import time
	import random
	import contextlib
	import torch
	import torch.nn as nn
	import torch.optim as optim
	import torch.distributed as dist
	l = [[1, 1675, 4028, 8271, 42, 2], [1, 1675, 4028, 8844, 10135, 10757, 9478, 42, 2]]

	# maximum rows and columns
	max_rows = len(l)
	max_cols = max(len(row) for row in l)

	# padded rows
	padded = [row + [0] * (max_cols - len(row)) for row in l]

	# unpack nested list