Michael J Clark wassname

## numpy_dataset.py
import torch.utils.data


class NumpyDataset(torch.utils.data.Dataset):
    """Dataset wrapping arrays.

    Each sample will be retrieved by indexing array along the first dimension.

    Arguments:
        *arrays (numpy.array): arrays that have the same size of the first dimension.

## tqdm_dask_progressbar.py

from dask.callbacks import Callback
from tqdm.auto import tqdm

class TQDMDaskProgressBar(Callback, object):
    """
    A tqdm progress bar for dask.
    Usage:
        ```
        with TQDMDaskProgressBar():

## kdtree_scaling_scipy.ipynb

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                wassname
                / kdtree_scaling_scipy.ipynb
            
            
              Last active
              April 30, 2018 03:33
            
              
                kdtree_scaling_scipy
              
          
      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## jupyter_logging.py
"""
In jupyter notebook simple logging to console
"""
import logging
import sys

logging.basicConfig(stream=sys.stdout, level=logging.INFO)

# Test
logger = logging.getLogger('LOGGER_NAME')

## AdamStepLR.py
class AdamStepLR(torch.optim.Adam):
    """Combine Adam and lr_scheduler.StepLR so we can use it as a normal optimiser"""
    def __init__(self, params, lr=0.001, betas=(0.9, 0.999), eps=1e-08, weight_decay=0, step_size=50000, gamma=0.5):
        super().__init__(params, lr, betas, eps, weight_decay)
        self.scheduler = torch.optim.lr_scheduler.StepLR(self, step_size, gamma)

    def step(self):
        self.scheduler.step()
        return super().step()

## pytorch_window_stack.py
def window_stack(x, window=4, pad=True):
    """
    Stack along a moving window of a pytorch timeseries

    Inputs:
        tensor of dims (batches/time, channels)
        pad: if true the left side will be padded to let the output match
    Outputs:
        if pad=True: a tensor of size (batches, channels, window)
        else: tensor of size (batches-window, channels, window)

## layer_norm_conv2d.py
import torch
import torch.nn as nn

class LayerNormConv2d(nn.Module):
    """
    Layer norm the just works on the channel axis for a Conv2d

    Ref:
    - code modified from https://github.com/Scitator/Run-Skeleton-Run/blob/master/common/modules/LayerNorm.py
    - paper: https://arxiv.org/abs/1607.06450

## batchrenorm2d.py
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.autograd as autograd
from torch.nn.parameter import Parameter
from torch.autograd import Variable


def r_d_max_func(itr):
    "Default max r and d provider as recommended in paper."

## hydrosaver.ipynb

      
              2 files
            
          
              0 forks
            
          
              0 comments
            
          
              2 stars
            
          
                wassname
                / hydrosaver.ipynb
            
            
              Last active
              February 13, 2018 03:28
            
              
                starter colab jupyter notebook for the hydrosaver competition
              
          
      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## pandas_profiling_hydrosaver.html
<!doctype html>

<html lang="en">
<head>
  <meta charset="utf-8">

  <title>Profile report</title>
  <meta name="description" content="Profile report generated by pandas-profiling. See GitHub.">
  <meta name="author" content="pandas-profiling">
    <script src="https://ajax.googleapis.com/ajax/libs/jquery/1.11.3/jquery.min.js"></script>
	import torch.utils.data


	class NumpyDataset(torch.utils.data.Dataset):
	"""Dataset wrapping arrays.

	Each sample will be retrieved by indexing array along the first dimension.

	Arguments:
	*arrays (numpy.array): arrays that have the same size of the first dimension.

	from dask.callbacks import Callback
	from tqdm.auto import tqdm

	class TQDMDaskProgressBar(Callback, object):
	"""
	A tqdm progress bar for dask.
	Usage:
	```
	with TQDMDaskProgressBar():
	"""
	In jupyter notebook simple logging to console
	"""
	import logging
	import sys

	logging.basicConfig(stream=sys.stdout, level=logging.INFO)

	# Test
	logger = logging.getLogger('LOGGER_NAME')
	class AdamStepLR(torch.optim.Adam):
	"""Combine Adam and lr_scheduler.StepLR so we can use it as a normal optimiser"""
	def __init__(self, params, lr=0.001, betas=(0.9, 0.999), eps=1e-08, weight_decay=0, step_size=50000, gamma=0.5):
	super().__init__(params, lr, betas, eps, weight_decay)
	self.scheduler = torch.optim.lr_scheduler.StepLR(self, step_size, gamma)

	def step(self):
	self.scheduler.step()
	return super().step()
	def window_stack(x, window=4, pad=True):
	"""
	Stack along a moving window of a pytorch timeseries

	Inputs:
	tensor of dims (batches/time, channels)
	pad: if true the left side will be padded to let the output match
	Outputs:
	if pad=True: a tensor of size (batches, channels, window)
	else: tensor of size (batches-window, channels, window)
	import torch
	import torch.nn as nn

	class LayerNormConv2d(nn.Module):
	"""
	Layer norm the just works on the channel axis for a Conv2d

	Ref:
	- code modified from https://github.com/Scitator/Run-Skeleton-Run/blob/master/common/modules/LayerNorm.py
	- paper: https://arxiv.org/abs/1607.06450
	import torch
	import torch.nn as nn
	import torch.nn.functional as F
	import torch.autograd as autograd
	from torch.nn.parameter import Parameter
	from torch.autograd import Variable


	def r_d_max_func(itr):
	"Default max r and d provider as recommended in paper."
	<!doctype html>

	<html lang="en">
	<head>
	<meta charset="utf-8">

	<title>Profile report</title>
	<meta name="description" content="Profile report generated by pandas-profiling. See GitHub.">
	<meta name="author" content="pandas-profiling">
	<script src="https://ajax.googleapis.com/ajax/libs/jquery/1.11.3/jquery.min.js"></script>