krishvishal

As described in detail here, julia can take really excessive amounts of time to execute the first @everywhere statement on many processes — around 1 hour for thousands of processes — even if the actual code being executed everywhere is trivial. Basically, the Distributed functions need to be precompiled to make this happen quickly.

This gist provides a simple way to do so — at least on Slurm clusters (though the same principles should apply elsewhere). Just submit precompile.jl as a batch job (adjusting the SBATCH directives as needed), and it should create a sysimage that you can use to run future batch jobs. Check end of the log of the Slurm job to see exactly how to use the sysimage.

Note that both the original julia process and all processes created with addprocs should use the --sysimage=/path/to/sys_everywhere.so argument. Doing so reduces the time taken to execute the first @everywhere sta

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()

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):