krishvishal

# high-low spread estimator (hlse)
def hlse(ohlc_df, frequency='daily'):
    """
    Computes the high-low spread estimator, an estimate of bid-offer spreads, a measure of liquidity risk.
    See Corwin & Schultz (2011) for details: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1106193

    Parameters
    ----------
    ohlc_df: DataFrame
        DataFrame with DatetimeIndex and Open, High, Low and Close (OHLC) prices from which to compute the high-low spread estimates.

krishvishal

# Use packages to ensure that we trigger Requires.jl.
using Zygote: Zygote
using ReverseDiff: ReverseDiff
using ForwardDiff: ForwardDiff
using Tracker: Tracker
using Memoization: Memoization # used for ReverseDiff.jl cache.

using Turing.Core: ForwardDiffAD, ReverseDiffAD, TrackerAD, ZygoteAD, CHUNKSIZE

const DEFAULT_ADBACKENDS = [

krishvishal

krishvishal

krishvishal

using ReverseDiff

f(x) = maximum(x;dims=1)

if abspath(PROGRAM_FILE) == @__FILE__
	x = rand(3, 3)
	println(x)
	println(ReverseDiff.gradient(sum∘f, x))
end

krishvishal

def getsize(obj_0):
    """Recursively iterate to sum size of object & members."""
    _seen_ids = set()
    def inner(obj):
        obj_id = id(obj)
        if obj_id in _seen_ids:
            return 0
        _seen_ids.add(obj_id)
        size = sys.getsizeof(obj)
        if isinstance(obj, zero_depth_bases):

krishvishal

from model import Net
from trainer import Trainer 
import torch
from torch import nn
from matplotlib import pyplot as plt

model = Net()
ckpt = torch.load('path_to_checkpoint')
model.load_state_dict(ckpt['state_dict'])
filter = model.conv1.weight.data.numpy()