lostella

Keybase proof

I hereby claim:

• I am lostella on github.
• I am lostella (https://keybase.io/lostella) on keybase.
• I have a public key whose fingerprint is FBAB AFA4 81E5 F7BA E3D7 9DD3 ACE0 41D6 6164 F46C

To claim this, I am signing this object:

lostella

Keybase proof

I hereby claim:

• I am lostella on github.
• I am lostella (https://keybase.io/lostella) on keybase.
• I have a public key ASAiQnm-kLkopRgiLCGDKC5w5ZcA-ehLSCEZAp-HYjaxLAo

To claim this, I am signing this object:

lostella

import mxnet as mx

class MyBlock(mx.gluon.HybridBlock):
    def __init__(self):
        super().__init__()
        with self.name_scope():
            self.lstm = mx.gluon.rnn.HybridSequentialRNNCell()
            for layer in range(3):
                self.lstm.add(mx.gluon.rnn.LSTMCell(hidden_size=20))

lostella

import mxnet as mx

class MyBlock(mx.gluon.HybridBlock):
    def __init__(self):
        super().__init__()
        with self.name_scope():
            self.lstmcell = mx.gluon.rnn.LSTMCell(hidden_size=20)

    def hybrid_forward(self, F, seq):
        outputs, state = self.lstmcell.unroll(inputs=seq, length=10, layout="NTC", merge_outputs=True)

lostella

from pathlib import Path
from typing import List


class IndexedFile:
    def __init__(self, path: Path) -> None:
        self.path = path
        self.offset: List[int] = []
        self._build_index()
    

lostella

import time
from functools import partial

import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

import mxnet as mx
import torch

lostella

import time

import mxnet as mx

from gluonts.dataset.repository.datasets import get_dataset
from gluonts.dataset.loader import TrainDataLoader
from gluonts.model.deepar import DeepAREstimator
from gluonts.mx.batchify import batchify as mx_batchify

dataset = get_dataset("electricity")

lostella

from gluonts.dataset.repository.datasets import get_dataset, dataset_names

def check_train_test_split(dataset):
    prediction_length = dataset.metadata.prediction_length

    train_end = {}
    for entry in dataset.train:
        assert entry["item_id"] not in train_end, f"item {k} is duplicate"
        train_end[entry["item_id"]] = entry["start"] + len(entry["target"]) * entry["start"].freq

lostella

syntax on
filetype plugin indent on
set tabstop=4 softtabstop=4 shiftwidth=4 expandtab smarttab autoindent
set incsearch ignorecase smartcase hlsearch
set wildmode=longest,list,full wildmenu
set ruler laststatus=2 showcmd showmode
set list listchars=trail:»,tab:»-
set fillchars+=vert:\
set wrap breakindent
set encoding=utf-8

lostella

# Works on gluonts dev branch as of May 30th, 2023
# Assumes "m5-forecasting-accuracy" folder with data next to the script
# Data is obtained from https://www.kaggle.com/c/m5-forecasting-accuracy

import pandas as pd
from pathlib import Path
from gluonts.dataset.pandas import PandasDataset

# Load data from csv files