{kantime} is a minimal wrapper for Time Series Kolmogorov Arnold Networks in R. By binding nixtla’s {neuralforecast} engine, KAN, from Python to use in R. Additional steps were made to bridged to {modeltime}, hence the name {kantime}!
You can install the development version of kantime like so:
# devtools::install_github("frankiethull/kantime")
install.packages("kantime", repos = c('https://frankiethull.r-universe.dev', 'https://cloud.r-project.org'))
This is a basic example showing the barebones {reticulate} Python wrappers. These are the core bindings before registering can configuring to {parsnip} and then bridging to {modeltime}.
kantime requires Python and neuralforecast. Python dependencies are
managed automatically via reticulate::py_require() when the package
loads.
# load the R library -------------------
# neuralforecast is installed automatically on first use
library(kantime)
## data comes from sister repo
air_passengers_df <- arrow::read_csv_arrow("https://raw.githubusercontent.com/frankiethull/nixtla-r-tutorial/refs/heads/main/airpassengersDF.csv")
air_passengers_df <- air_passengers_df |>
dplyr::mutate(
ds = as.Date(ds)
)
Sys.sleep(1)
train_df <- air_passengers_df |> dplyr::slice(1:132)
test_df <- air_passengers_df |> dplyr::anti_join(train_df)
#> Joining with `by = join_by(unique_id, ds, y)`
note that nixtla’s design requires a unique_id for each unique time series, ds for the time column, and y for the outcome variable. The internals work as minimal python wrappers to fit and predict on a time series data like so:
# kan specs --
kan_model_specs <- kantime:::kan_spec(h = 12L)
# fit
kan_model_fit <- kantime:::conformal_fit(model_spec = kan_model_specs, df = train_df)
# predict
conf_kan_preds <- kantime:::conformal_predict(
model_spec = kan_model_specs,
model_fit = kan_model_fit,
level = 90L)
These internals are built thanks to {reticulate} and of using Nixtla’s {neuralforecast} via R. While the idea works ok, the results and workflow can be clunky for R, especially when building for many unique IDs (many models for different time series loses pandas row IDs). Additionally, handling parallel processes in R via {reticulate} for cv, tuning and training in Python requires additional robustness.
{kantime} is super experimental in it’s design. Which is also why this is not a full {neuralforecast} binding. This binding in particular is going to leverage the root KAN model, a few helper utils, then bind the base KAN model to {parsnip} and bridge to {modeltime}. This loses pieces of the underlying Nixtla functionality but replacing with {tidymodels} & {modeltime} functionality.
given our underlying internals, we can bind these functions to {modeltime} which is similar to registering a parsnip model but requires an additional bridge. The bridge implementation is shown below, but we have to add predict methods and there’s quite a bit of underlying work that makes it fully functioning.
kan_bridge <- kantime:::kan_bridge_fit_impl(
x = train_df |> dplyr::select(-y),
y = train_df |> dplyr::pull(y),
h = 12L,
input_size = 24L,
max_steps = 10L,
freq = "ME"
)
kan_bridge |> predict()
Remembering that once the model has been bridged to {modeltime}, we lose
some of the underlying nixtla utilities, but get access to tidymodels +
modeltime utilities in R. What’s this mean? We can now use this KAN
binding with tidymodels tools like initial_time_split and we can use
leverage modeltime’s toolkit to further calibrate & validate KAN
models. In fact, we went through these additional bridging steps to have
a full suite of backtesting, ensembling, calibrating, and scoring tools
that will be very familiar if already using {modeltime} and/or
{tidymodels}.
library(parsnip)
library(modeltime)
# time split
splits <- rsample::initial_time_split(air_passengers_df, prop = .92)
training <- rsample::training(splits)
testing <- rsample::testing(splits)
# kantime fit
kantime_fit <- kan(h = 12L,
input_size = 24L,
max_steps = 10L,
freq = "ME") |>
set_engine("kan") |>
fit(y ~ ds + unique_id, data = training)
#> Sanity Checking: | | 0/? [00:00<?, ?it/s]Sanity Checking: | | 0/? [00:00<?, ?it/s]Sanity Checking DataLoader 0: 0%| | 0/1 [00:00<?, ?it/s]Sanity Checking DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 1858.35it/s] Training: | | 0/? [00:00<?, ?it/s]Training: | | 0/? [00:00<?, ?it/s]Epoch 0: 0%| | 0/1 [00:00<?, ?it/s]Epoch 0: 100%|██████████| 1/1 [00:00<00:00, 2.79it/s]Epoch 0: 100%|██████████| 1/1 [00:00<00:00, 2.79it/s, v_num=30, train_loss_step=1.860]Epoch 0: 100%|██████████| 1/1 [00:00<00:00, 2.79it/s, v_num=30, train_loss_step=1.860, train_loss_epoch=1.860]Epoch 0: 0%| | 0/1 [00:00<?, ?it/s, v_num=30, train_loss_step=1.860, train_loss_epoch=1.860] Epoch 1: 0%| | 0/1 [00:00<?, ?it/s, v_num=30, train_loss_step=1.860, train_loss_epoch=1.860]Epoch 1: 100%|██████████| 1/1 [00:00<00:00, 2.50it/s, v_num=30, train_loss_step=1.860, train_loss_epoch=1.860]Epoch 1: 100%|██████████| 1/1 [00:00<00:00, 2.50it/s, v_num=30, train_loss_step=1.810, train_loss_epoch=1.860]Epoch 1: 100%|██████████| 1/1 [00:00<00:00, 2.50it/s, v_num=30, train_loss_step=1.810, train_loss_epoch=1.810]Epoch 1: 0%| | 0/1 [00:00<?, ?it/s, v_num=30, train_loss_step=1.810, train_loss_epoch=1.810] Epoch 2: 0%| | 0/1 [00:00<?, ?it/s, v_num=30, train_loss_step=1.810, train_loss_epoch=1.810]Epoch 2: 100%|██████████| 1/1 [00:00<00:00, 2.36it/s, v_num=30, train_loss_step=1.810, train_loss_epoch=1.810]Epoch 2: 100%|██████████| 1/1 [00:00<00:00, 2.36it/s, v_num=30, train_loss_step=1.740, train_loss_epoch=1.810]Epoch 2: 100%|██████████| 1/1 [00:00<00:00, 2.36it/s, v_num=30, train_loss_step=1.740, train_loss_epoch=1.740]Epoch 2: 0%| | 0/1 [00:00<?, ?it/s, v_num=30, train_loss_step=1.740, train_loss_epoch=1.740] Epoch 3: 0%| | 0/1 [00:00<?, ?it/s, v_num=30, train_loss_step=1.740, train_loss_epoch=1.740]Epoch 3: 100%|██████████| 1/1 [00:00<00:00, 2.23it/s, v_num=30, train_loss_step=1.740, train_loss_epoch=1.740]Epoch 3: 100%|██████████| 1/1 [00:00<00:00, 2.23it/s, v_num=30, train_loss_step=1.680, train_loss_epoch=1.740]Epoch 3: 100%|██████████| 1/1 [00:00<00:00, 2.22it/s, v_num=30, train_loss_step=1.680, train_loss_epoch=1.680]Epoch 3: 0%| | 0/1 [00:00<?, ?it/s, v_num=30, train_loss_step=1.680, train_loss_epoch=1.680] Epoch 4: 0%| | 0/1 [00:00<?, ?it/s, v_num=30, train_loss_step=1.680, train_loss_epoch=1.680]Epoch 4: 100%|██████████| 1/1 [00:00<00:00, 2.82it/s, v_num=30, train_loss_step=1.680, train_loss_epoch=1.680]Epoch 4: 100%|██████████| 1/1 [00:00<00:00, 2.81it/s, v_num=30, train_loss_step=1.620, train_loss_epoch=1.680]Epoch 4: 100%|██████████| 1/1 [00:00<00:00, 2.81it/s, v_num=30, train_loss_step=1.620, train_loss_epoch=1.620]Epoch 4: 0%| | 0/1 [00:00<?, ?it/s, v_num=30, train_loss_step=1.620, train_loss_epoch=1.620] Epoch 5: 0%| | 0/1 [00:00<?, ?it/s, v_num=30, train_loss_step=1.620, train_loss_epoch=1.620]Epoch 5: 100%|██████████| 1/1 [00:00<00:00, 4.05it/s, v_num=30, train_loss_step=1.620, train_loss_epoch=1.620]Epoch 5: 100%|██████████| 1/1 [00:00<00:00, 4.05it/s, v_num=30, train_loss_step=1.570, train_loss_epoch=1.620]Epoch 5: 100%|██████████| 1/1 [00:00<00:00, 4.04it/s, v_num=30, train_loss_step=1.570, train_loss_epoch=1.570]Epoch 5: 0%| | 0/1 [00:00<?, ?it/s, v_num=30, train_loss_step=1.570, train_loss_epoch=1.570] Epoch 6: 0%| | 0/1 [00:00<?, ?it/s, v_num=30, train_loss_step=1.570, train_loss_epoch=1.570]Epoch 6: 100%|██████████| 1/1 [00:00<00:00, 2.98it/s, v_num=30, train_loss_step=1.570, train_loss_epoch=1.570]Epoch 6: 100%|██████████| 1/1 [00:00<00:00, 2.97it/s, v_num=30, train_loss_step=1.500, train_loss_epoch=1.570]Epoch 6: 100%|██████████| 1/1 [00:00<00:00, 2.97it/s, v_num=30, train_loss_step=1.500, train_loss_epoch=1.500]Epoch 6: 0%| | 0/1 [00:00<?, ?it/s, v_num=30, train_loss_step=1.500, train_loss_epoch=1.500] Epoch 7: 0%| | 0/1 [00:00<?, ?it/s, v_num=30, train_loss_step=1.500, train_loss_epoch=1.500]Epoch 7: 100%|██████████| 1/1 [00:00<00:00, 3.06it/s, v_num=30, train_loss_step=1.500, train_loss_epoch=1.500]Epoch 7: 100%|██████████| 1/1 [00:00<00:00, 3.06it/s, v_num=30, train_loss_step=1.430, train_loss_epoch=1.500]Epoch 7: 100%|██████████| 1/1 [00:00<00:00, 3.05it/s, v_num=30, train_loss_step=1.430, train_loss_epoch=1.430]Epoch 7: 0%| | 0/1 [00:00<?, ?it/s, v_num=30, train_loss_step=1.430, train_loss_epoch=1.430] Epoch 8: 0%| | 0/1 [00:00<?, ?it/s, v_num=30, train_loss_step=1.430, train_loss_epoch=1.430]Epoch 8: 100%|██████████| 1/1 [00:00<00:00, 2.21it/s, v_num=30, train_loss_step=1.430, train_loss_epoch=1.430]Epoch 8: 100%|██████████| 1/1 [00:00<00:00, 2.20it/s, v_num=30, train_loss_step=1.390, train_loss_epoch=1.430]Epoch 8: 100%|██████████| 1/1 [00:00<00:00, 2.20it/s, v_num=30, train_loss_step=1.390, train_loss_epoch=1.390]Epoch 8: 0%| | 0/1 [00:00<?, ?it/s, v_num=30, train_loss_step=1.390, train_loss_epoch=1.390] Epoch 9: 0%| | 0/1 [00:00<?, ?it/s, v_num=30, train_loss_step=1.390, train_loss_epoch=1.390]Epoch 9: 100%|██████████| 1/1 [00:00<00:00, 1.95it/s, v_num=30, train_loss_step=1.390, train_loss_epoch=1.390]Epoch 9: 100%|██████████| 1/1 [00:00<00:00, 1.95it/s, v_num=30, train_loss_step=1.320, train_loss_epoch=1.390]
#> Validation: | | 0/? [00:00<?, ?it/s][A
#> Validation: | | 0/? [00:00<?, ?it/s][A
#> Validation DataLoader 0: 0%| | 0/1 [00:00<?, ?it/s][A
#> Validation DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 1460.92it/s][A
#> [AEpoch 9: 100%|██████████| 1/1 [00:00<00:00, 1.94it/s, v_num=30, train_loss_step=1.320, train_loss_epoch=1.390]Epoch 9: 100%|██████████| 1/1 [00:00<00:00, 1.94it/s, v_num=30, train_loss_step=1.320, train_loss_epoch=1.320]Epoch 9: 100%|██████████| 1/1 [00:00<00:00, 1.89it/s, v_num=30, train_loss_step=1.320, train_loss_epoch=1.320]
#> Predicting: | | 0/? [00:00<?, ?it/s]Predicting: | | 0/? [00:00<?, ?it/s]Predicting DataLoader 0: 0%| | 0/1 [00:00<?, ?it/s]Predicting DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 144.60it/s]Predicting DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 137.94it/s]
#> Sanity Checking: | | 0/? [00:00<?, ?it/s]Sanity Checking: | | 0/? [00:00<?, ?it/s]Sanity Checking DataLoader 0: 0%| | 0/1 [00:00<?, ?it/s]Sanity Checking DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 2272.10it/s] Training: | | 0/? [00:00<?, ?it/s]Training: | | 0/? [00:00<?, ?it/s]Epoch 0: 0%| | 0/1 [00:00<?, ?it/s]Epoch 0: 100%|██████████| 1/1 [00:00<00:00, 3.20it/s]Epoch 0: 100%|██████████| 1/1 [00:00<00:00, 3.20it/s, v_num=32, train_loss_step=1.210]Epoch 0: 100%|██████████| 1/1 [00:00<00:00, 3.19it/s, v_num=32, train_loss_step=1.210, train_loss_epoch=1.210]Epoch 0: 0%| | 0/1 [00:00<?, ?it/s, v_num=32, train_loss_step=1.210, train_loss_epoch=1.210] Epoch 1: 0%| | 0/1 [00:00<?, ?it/s, v_num=32, train_loss_step=1.210, train_loss_epoch=1.210]Epoch 1: 100%|██████████| 1/1 [00:00<00:00, 1.92it/s, v_num=32, train_loss_step=1.210, train_loss_epoch=1.210]Epoch 1: 100%|██████████| 1/1 [00:00<00:00, 1.92it/s, v_num=32, train_loss_step=1.160, train_loss_epoch=1.210]Epoch 1: 100%|██████████| 1/1 [00:00<00:00, 1.92it/s, v_num=32, train_loss_step=1.160, train_loss_epoch=1.160]Epoch 1: 0%| | 0/1 [00:00<?, ?it/s, v_num=32, train_loss_step=1.160, train_loss_epoch=1.160] Epoch 2: 0%| | 0/1 [00:00<?, ?it/s, v_num=32, train_loss_step=1.160, train_loss_epoch=1.160]Epoch 2: 100%|██████████| 1/1 [00:00<00:00, 2.47it/s, v_num=32, train_loss_step=1.160, train_loss_epoch=1.160]Epoch 2: 100%|██████████| 1/1 [00:00<00:00, 2.47it/s, v_num=32, train_loss_step=1.090, train_loss_epoch=1.160]Epoch 2: 100%|██████████| 1/1 [00:00<00:00, 2.46it/s, v_num=32, train_loss_step=1.090, train_loss_epoch=1.090]Epoch 2: 0%| | 0/1 [00:00<?, ?it/s, v_num=32, train_loss_step=1.090, train_loss_epoch=1.090] Epoch 3: 0%| | 0/1 [00:00<?, ?it/s, v_num=32, train_loss_step=1.090, train_loss_epoch=1.090]Epoch 3: 100%|██████████| 1/1 [00:00<00:00, 2.34it/s, v_num=32, train_loss_step=1.090, train_loss_epoch=1.090]Epoch 3: 100%|██████████| 1/1 [00:00<00:00, 2.34it/s, v_num=32, train_loss_step=1.030, train_loss_epoch=1.090]Epoch 3: 100%|██████████| 1/1 [00:00<00:00, 2.33it/s, v_num=32, train_loss_step=1.030, train_loss_epoch=1.030]Epoch 3: 0%| | 0/1 [00:00<?, ?it/s, v_num=32, train_loss_step=1.030, train_loss_epoch=1.030] Epoch 4: 0%| | 0/1 [00:00<?, ?it/s, v_num=32, train_loss_step=1.030, train_loss_epoch=1.030]Epoch 4: 100%|██████████| 1/1 [00:00<00:00, 2.85it/s, v_num=32, train_loss_step=1.030, train_loss_epoch=1.030]Epoch 4: 100%|██████████| 1/1 [00:00<00:00, 2.85it/s, v_num=32, train_loss_step=0.975, train_loss_epoch=1.030]Epoch 4: 100%|██████████| 1/1 [00:00<00:00, 2.84it/s, v_num=32, train_loss_step=0.975, train_loss_epoch=0.975]Epoch 4: 0%| | 0/1 [00:00<?, ?it/s, v_num=32, train_loss_step=0.975, train_loss_epoch=0.975] Epoch 5: 0%| | 0/1 [00:00<?, ?it/s, v_num=32, train_loss_step=0.975, train_loss_epoch=0.975]Epoch 5: 100%|██████████| 1/1 [00:00<00:00, 2.07it/s, v_num=32, train_loss_step=0.975, train_loss_epoch=0.975]Epoch 5: 100%|██████████| 1/1 [00:00<00:00, 2.07it/s, v_num=32, train_loss_step=0.909, train_loss_epoch=0.975]Epoch 5: 100%|██████████| 1/1 [00:00<00:00, 2.07it/s, v_num=32, train_loss_step=0.909, train_loss_epoch=0.909]Epoch 5: 0%| | 0/1 [00:00<?, ?it/s, v_num=32, train_loss_step=0.909, train_loss_epoch=0.909] Epoch 6: 0%| | 0/1 [00:00<?, ?it/s, v_num=32, train_loss_step=0.909, train_loss_epoch=0.909]Epoch 6: 100%|██████████| 1/1 [00:00<00:00, 1.77it/s, v_num=32, train_loss_step=0.909, train_loss_epoch=0.909]Epoch 6: 100%|██████████| 1/1 [00:00<00:00, 1.77it/s, v_num=32, train_loss_step=0.857, train_loss_epoch=0.909]Epoch 6: 100%|██████████| 1/1 [00:00<00:00, 1.77it/s, v_num=32, train_loss_step=0.857, train_loss_epoch=0.857]Epoch 6: 0%| | 0/1 [00:00<?, ?it/s, v_num=32, train_loss_step=0.857, train_loss_epoch=0.857] Epoch 7: 0%| | 0/1 [00:00<?, ?it/s, v_num=32, train_loss_step=0.857, train_loss_epoch=0.857]Epoch 7: 100%|██████████| 1/1 [00:00<00:00, 3.09it/s, v_num=32, train_loss_step=0.857, train_loss_epoch=0.857]Epoch 7: 100%|██████████| 1/1 [00:00<00:00, 3.09it/s, v_num=32, train_loss_step=0.809, train_loss_epoch=0.857]Epoch 7: 100%|██████████| 1/1 [00:00<00:00, 3.09it/s, v_num=32, train_loss_step=0.809, train_loss_epoch=0.809]Epoch 7: 0%| | 0/1 [00:00<?, ?it/s, v_num=32, train_loss_step=0.809, train_loss_epoch=0.809] Epoch 8: 0%| | 0/1 [00:00<?, ?it/s, v_num=32, train_loss_step=0.809, train_loss_epoch=0.809]Epoch 8: 100%|██████████| 1/1 [00:00<00:00, 2.65it/s, v_num=32, train_loss_step=0.809, train_loss_epoch=0.809]Epoch 8: 100%|██████████| 1/1 [00:00<00:00, 2.65it/s, v_num=32, train_loss_step=0.763, train_loss_epoch=0.809]Epoch 8: 100%|██████████| 1/1 [00:00<00:00, 2.64it/s, v_num=32, train_loss_step=0.763, train_loss_epoch=0.763]Epoch 8: 0%| | 0/1 [00:00<?, ?it/s, v_num=32, train_loss_step=0.763, train_loss_epoch=0.763] Epoch 9: 0%| | 0/1 [00:00<?, ?it/s, v_num=32, train_loss_step=0.763, train_loss_epoch=0.763]Epoch 9: 100%|██████████| 1/1 [00:00<00:00, 3.59it/s, v_num=32, train_loss_step=0.763, train_loss_epoch=0.763]Epoch 9: 100%|██████████| 1/1 [00:00<00:00, 3.59it/s, v_num=32, train_loss_step=0.722, train_loss_epoch=0.763]
#> Validation: | | 0/? [00:00<?, ?it/s][A
#> Validation: | | 0/? [00:00<?, ?it/s][A
#> Validation DataLoader 0: 0%| | 0/1 [00:00<?, ?it/s][A
#> Validation DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 2283.24it/s][A
#> [AEpoch 9: 100%|██████████| 1/1 [00:00<00:00, 3.56it/s, v_num=32, train_loss_step=0.722, train_loss_epoch=0.763]Epoch 9: 100%|██████████| 1/1 [00:00<00:00, 3.55it/s, v_num=32, train_loss_step=0.722, train_loss_epoch=0.722]Epoch 9: 100%|██████████| 1/1 [00:00<00:00, 3.41it/s, v_num=32, train_loss_step=0.722, train_loss_epoch=0.722]
#> Predicting: | | 0/? [00:00<?, ?it/s]Predicting: | | 0/? [00:00<?, ?it/s]Predicting DataLoader 0: 0%| | 0/1 [00:00<?, ?it/s]Predicting DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 5.68it/s]Predicting DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 5.66it/s]
# kantime predict
kan_point_predictions <-
kantime_fit |>
modeltime_table() |>
modeltime_forecast(actual_data = training, new_data = testing)
#> Predicting: | | 0/? [00:00<?, ?it/s]Predicting: | | 0/? [00:00<?, ?it/s]Predicting DataLoader 0: 0%| | 0/1 [00:00<?, ?it/s]Predicting DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 400.45it/s]Predicting DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 354.52it/s]
kan_point_predictions |>
plot_modeltime_forecast(.conf_interval_show = FALSE, .interactive = FALSE)
# RMoK (Robust Mixture of Kernels) via neuralforecast
rmok_fit <- rmok(h = 12L,
input_size = 24L,
max_steps = 10L,
freq = "ME") |>
set_engine("rmok") |>
fit(y ~ ds + unique_id, data = training)
#> Sanity Checking: | | 0/? [00:00<?, ?it/s]Sanity Checking: | | 0/? [00:00<?, ?it/s]Sanity Checking DataLoader 0: 0%| | 0/1 [00:00<?, ?it/s]Sanity Checking DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 3151.24it/s] Training: | | 0/? [00:00<?, ?it/s]Training: | | 0/? [00:00<?, ?it/s]Epoch 0: 0%| | 0/1 [00:00<?, ?it/s]Epoch 0: 100%|██████████| 1/1 [00:00<00:00, 81.52it/s]Epoch 0: 100%|██████████| 1/1 [00:00<00:00, 79.62it/s, v_num=33, train_loss_step=1.890]Epoch 0: 100%|██████████| 1/1 [00:00<00:00, 77.92it/s, v_num=33, train_loss_step=1.890, train_loss_epoch=1.890]Epoch 0: 0%| | 0/1 [00:00<?, ?it/s, v_num=33, train_loss_step=1.890, train_loss_epoch=1.890] Epoch 1: 0%| | 0/1 [00:00<?, ?it/s, v_num=33, train_loss_step=1.890, train_loss_epoch=1.890]Epoch 1: 100%|██████████| 1/1 [00:00<00:00, 20.04it/s, v_num=33, train_loss_step=1.890, train_loss_epoch=1.890]Epoch 1: 100%|██████████| 1/1 [00:00<00:00, 19.86it/s, v_num=33, train_loss_step=1.800, train_loss_epoch=1.890]Epoch 1: 100%|██████████| 1/1 [00:00<00:00, 19.68it/s, v_num=33, train_loss_step=1.800, train_loss_epoch=1.800]Epoch 1: 0%| | 0/1 [00:00<?, ?it/s, v_num=33, train_loss_step=1.800, train_loss_epoch=1.800] Epoch 2: 0%| | 0/1 [00:00<?, ?it/s, v_num=33, train_loss_step=1.800, train_loss_epoch=1.800]Epoch 2: 100%|██████████| 1/1 [00:00<00:00, 122.15it/s, v_num=33, train_loss_step=1.800, train_loss_epoch=1.800]Epoch 2: 100%|██████████| 1/1 [00:00<00:00, 117.92it/s, v_num=33, train_loss_step=1.730, train_loss_epoch=1.800]Epoch 2: 100%|██████████| 1/1 [00:00<00:00, 113.71it/s, v_num=33, train_loss_step=1.730, train_loss_epoch=1.730]Epoch 2: 0%| | 0/1 [00:00<?, ?it/s, v_num=33, train_loss_step=1.730, train_loss_epoch=1.730] Epoch 3: 0%| | 0/1 [00:00<?, ?it/s, v_num=33, train_loss_step=1.730, train_loss_epoch=1.730]Epoch 3: 100%|██████████| 1/1 [00:00<00:00, 91.15it/s, v_num=33, train_loss_step=1.730, train_loss_epoch=1.730]Epoch 3: 100%|██████████| 1/1 [00:00<00:00, 88.18it/s, v_num=33, train_loss_step=1.700, train_loss_epoch=1.730]Epoch 3: 100%|██████████| 1/1 [00:00<00:00, 84.90it/s, v_num=33, train_loss_step=1.700, train_loss_epoch=1.700]Epoch 3: 0%| | 0/1 [00:00<?, ?it/s, v_num=33, train_loss_step=1.700, train_loss_epoch=1.700] Epoch 4: 0%| | 0/1 [00:00<?, ?it/s, v_num=33, train_loss_step=1.700, train_loss_epoch=1.700]Epoch 4: 100%|██████████| 1/1 [00:00<00:00, 96.21it/s, v_num=33, train_loss_step=1.700, train_loss_epoch=1.700]Epoch 4: 100%|██████████| 1/1 [00:00<00:00, 93.51it/s, v_num=33, train_loss_step=1.740, train_loss_epoch=1.700]Epoch 4: 100%|██████████| 1/1 [00:00<00:00, 89.97it/s, v_num=33, train_loss_step=1.740, train_loss_epoch=1.740]Epoch 4: 0%| | 0/1 [00:00<?, ?it/s, v_num=33, train_loss_step=1.740, train_loss_epoch=1.740] Epoch 5: 0%| | 0/1 [00:00<?, ?it/s, v_num=33, train_loss_step=1.740, train_loss_epoch=1.740]Epoch 5: 100%|██████████| 1/1 [00:00<00:00, 81.74it/s, v_num=33, train_loss_step=1.740, train_loss_epoch=1.740]Epoch 5: 100%|██████████| 1/1 [00:00<00:00, 79.37it/s, v_num=33, train_loss_step=1.620, train_loss_epoch=1.740]Epoch 5: 100%|██████████| 1/1 [00:00<00:00, 76.55it/s, v_num=33, train_loss_step=1.620, train_loss_epoch=1.620]Epoch 5: 0%| | 0/1 [00:00<?, ?it/s, v_num=33, train_loss_step=1.620, train_loss_epoch=1.620] Epoch 6: 0%| | 0/1 [00:00<?, ?it/s, v_num=33, train_loss_step=1.620, train_loss_epoch=1.620]Epoch 6: 100%|██████████| 1/1 [00:00<00:00, 17.36it/s, v_num=33, train_loss_step=1.620, train_loss_epoch=1.620]Epoch 6: 100%|██████████| 1/1 [00:00<00:00, 17.23it/s, v_num=33, train_loss_step=1.600, train_loss_epoch=1.620]Epoch 6: 100%|██████████| 1/1 [00:00<00:00, 17.09it/s, v_num=33, train_loss_step=1.600, train_loss_epoch=1.600]Epoch 6: 0%| | 0/1 [00:00<?, ?it/s, v_num=33, train_loss_step=1.600, train_loss_epoch=1.600] Epoch 7: 0%| | 0/1 [00:00<?, ?it/s, v_num=33, train_loss_step=1.600, train_loss_epoch=1.600]Epoch 7: 100%|██████████| 1/1 [00:00<00:00, 87.76it/s, v_num=33, train_loss_step=1.600, train_loss_epoch=1.600]Epoch 7: 100%|██████████| 1/1 [00:00<00:00, 85.41it/s, v_num=33, train_loss_step=1.550, train_loss_epoch=1.600]Epoch 7: 100%|██████████| 1/1 [00:00<00:00, 82.43it/s, v_num=33, train_loss_step=1.550, train_loss_epoch=1.550]Epoch 7: 0%| | 0/1 [00:00<?, ?it/s, v_num=33, train_loss_step=1.550, train_loss_epoch=1.550] Epoch 8: 0%| | 0/1 [00:00<?, ?it/s, v_num=33, train_loss_step=1.550, train_loss_epoch=1.550]Epoch 8: 100%|██████████| 1/1 [00:00<00:00, 16.36it/s, v_num=33, train_loss_step=1.550, train_loss_epoch=1.550]Epoch 8: 100%|██████████| 1/1 [00:00<00:00, 16.24it/s, v_num=33, train_loss_step=1.510, train_loss_epoch=1.550]Epoch 8: 100%|██████████| 1/1 [00:00<00:00, 16.12it/s, v_num=33, train_loss_step=1.510, train_loss_epoch=1.510]Epoch 8: 0%| | 0/1 [00:00<?, ?it/s, v_num=33, train_loss_step=1.510, train_loss_epoch=1.510] Epoch 9: 0%| | 0/1 [00:00<?, ?it/s, v_num=33, train_loss_step=1.510, train_loss_epoch=1.510]Epoch 9: 100%|██████████| 1/1 [00:00<00:00, 23.78it/s, v_num=33, train_loss_step=1.510, train_loss_epoch=1.510]Epoch 9: 100%|██████████| 1/1 [00:00<00:00, 23.54it/s, v_num=33, train_loss_step=1.490, train_loss_epoch=1.510]
#> Validation: | | 0/? [00:00<?, ?it/s][A
#> Validation: | | 0/? [00:00<?, ?it/s][A
#> Validation DataLoader 0: 0%| | 0/1 [00:00<?, ?it/s][A
#> Validation DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 1417.95it/s][A
#> [AEpoch 9: 100%|██████████| 1/1 [00:00<00:00, 21.68it/s, v_num=33, train_loss_step=1.490, train_loss_epoch=1.510]Epoch 9: 100%|██████████| 1/1 [00:00<00:00, 21.46it/s, v_num=33, train_loss_step=1.490, train_loss_epoch=1.490]Epoch 9: 100%|██████████| 1/1 [00:00<00:00, 18.01it/s, v_num=33, train_loss_step=1.490, train_loss_epoch=1.490]
#> Predicting: | | 0/? [00:00<?, ?it/s]Predicting: | | 0/? [00:00<?, ?it/s]Predicting DataLoader 0: 0%| | 0/1 [00:00<?, ?it/s]Predicting DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 560.06it/s]Predicting DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 474.84it/s]
#> Sanity Checking: | | 0/? [00:00<?, ?it/s]Sanity Checking: | | 0/? [00:00<?, ?it/s]Sanity Checking DataLoader 0: 0%| | 0/1 [00:00<?, ?it/s]Sanity Checking DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 3495.25it/s] Training: | | 0/? [00:00<?, ?it/s]Training: | | 0/? [00:00<?, ?it/s]Epoch 0: 0%| | 0/1 [00:00<?, ?it/s]Epoch 0: 100%|██████████| 1/1 [00:00<00:00, 90.34it/s]Epoch 0: 100%|██████████| 1/1 [00:00<00:00, 87.86it/s, v_num=35, train_loss_step=1.400]Epoch 0: 100%|██████████| 1/1 [00:00<00:00, 85.49it/s, v_num=35, train_loss_step=1.400, train_loss_epoch=1.400]Epoch 0: 0%| | 0/1 [00:00<?, ?it/s, v_num=35, train_loss_step=1.400, train_loss_epoch=1.400] Epoch 1: 0%| | 0/1 [00:00<?, ?it/s, v_num=35, train_loss_step=1.400, train_loss_epoch=1.400]Epoch 1: 100%|██████████| 1/1 [00:00<00:00, 82.10it/s, v_num=35, train_loss_step=1.400, train_loss_epoch=1.400]Epoch 1: 100%|██████████| 1/1 [00:00<00:00, 80.10it/s, v_num=35, train_loss_step=1.400, train_loss_epoch=1.400]Epoch 1: 100%|██████████| 1/1 [00:00<00:00, 77.60it/s, v_num=35, train_loss_step=1.400, train_loss_epoch=1.400]Epoch 1: 0%| | 0/1 [00:00<?, ?it/s, v_num=35, train_loss_step=1.400, train_loss_epoch=1.400] Epoch 2: 0%| | 0/1 [00:00<?, ?it/s, v_num=35, train_loss_step=1.400, train_loss_epoch=1.400]Epoch 2: 100%|██████████| 1/1 [00:00<00:00, 90.16it/s, v_num=35, train_loss_step=1.400, train_loss_epoch=1.400]Epoch 2: 100%|██████████| 1/1 [00:00<00:00, 87.48it/s, v_num=35, train_loss_step=1.280, train_loss_epoch=1.400]Epoch 2: 100%|██████████| 1/1 [00:00<00:00, 84.41it/s, v_num=35, train_loss_step=1.280, train_loss_epoch=1.280]Epoch 2: 0%| | 0/1 [00:00<?, ?it/s, v_num=35, train_loss_step=1.280, train_loss_epoch=1.280] Epoch 3: 0%| | 0/1 [00:00<?, ?it/s, v_num=35, train_loss_step=1.280, train_loss_epoch=1.280]Epoch 3: 100%|██████████| 1/1 [00:00<00:00, 76.49it/s, v_num=35, train_loss_step=1.280, train_loss_epoch=1.280]Epoch 3: 100%|██████████| 1/1 [00:00<00:00, 74.51it/s, v_num=35, train_loss_step=1.290, train_loss_epoch=1.280]Epoch 3: 100%|██████████| 1/1 [00:00<00:00, 72.22it/s, v_num=35, train_loss_step=1.290, train_loss_epoch=1.290]Epoch 3: 0%| | 0/1 [00:00<?, ?it/s, v_num=35, train_loss_step=1.290, train_loss_epoch=1.290] Epoch 4: 0%| | 0/1 [00:00<?, ?it/s, v_num=35, train_loss_step=1.290, train_loss_epoch=1.290]Epoch 4: 100%|██████████| 1/1 [00:00<00:00, 18.01it/s, v_num=35, train_loss_step=1.290, train_loss_epoch=1.290]Epoch 4: 100%|██████████| 1/1 [00:00<00:00, 17.87it/s, v_num=35, train_loss_step=1.300, train_loss_epoch=1.290]Epoch 4: 100%|██████████| 1/1 [00:00<00:00, 17.73it/s, v_num=35, train_loss_step=1.300, train_loss_epoch=1.300]Epoch 4: 0%| | 0/1 [00:00<?, ?it/s, v_num=35, train_loss_step=1.300, train_loss_epoch=1.300] Epoch 5: 0%| | 0/1 [00:00<?, ?it/s, v_num=35, train_loss_step=1.300, train_loss_epoch=1.300]Epoch 5: 100%|██████████| 1/1 [00:00<00:00, 81.99it/s, v_num=35, train_loss_step=1.300, train_loss_epoch=1.300]Epoch 5: 100%|██████████| 1/1 [00:00<00:00, 79.84it/s, v_num=35, train_loss_step=1.270, train_loss_epoch=1.300]Epoch 5: 100%|██████████| 1/1 [00:00<00:00, 77.25it/s, v_num=35, train_loss_step=1.270, train_loss_epoch=1.270]Epoch 5: 0%| | 0/1 [00:00<?, ?it/s, v_num=35, train_loss_step=1.270, train_loss_epoch=1.270] Epoch 6: 0%| | 0/1 [00:00<?, ?it/s, v_num=35, train_loss_step=1.270, train_loss_epoch=1.270]Epoch 6: 100%|██████████| 1/1 [00:00<00:00, 86.91it/s, v_num=35, train_loss_step=1.270, train_loss_epoch=1.270]Epoch 6: 100%|██████████| 1/1 [00:00<00:00, 84.56it/s, v_num=35, train_loss_step=1.220, train_loss_epoch=1.270]Epoch 6: 100%|██████████| 1/1 [00:00<00:00, 81.58it/s, v_num=35, train_loss_step=1.220, train_loss_epoch=1.220]Epoch 6: 0%| | 0/1 [00:00<?, ?it/s, v_num=35, train_loss_step=1.220, train_loss_epoch=1.220] Epoch 7: 0%| | 0/1 [00:00<?, ?it/s, v_num=35, train_loss_step=1.220, train_loss_epoch=1.220]Epoch 7: 100%|██████████| 1/1 [00:00<00:00, 98.22it/s, v_num=35, train_loss_step=1.220, train_loss_epoch=1.220]Epoch 7: 100%|██████████| 1/1 [00:00<00:00, 95.39it/s, v_num=35, train_loss_step=1.060, train_loss_epoch=1.220]Epoch 7: 100%|██████████| 1/1 [00:00<00:00, 92.48it/s, v_num=35, train_loss_step=1.060, train_loss_epoch=1.060]Epoch 7: 0%| | 0/1 [00:00<?, ?it/s, v_num=35, train_loss_step=1.060, train_loss_epoch=1.060] Epoch 8: 0%| | 0/1 [00:00<?, ?it/s, v_num=35, train_loss_step=1.060, train_loss_epoch=1.060]Epoch 8: 100%|██████████| 1/1 [00:00<00:00, 75.67it/s, v_num=35, train_loss_step=1.060, train_loss_epoch=1.060]Epoch 8: 100%|██████████| 1/1 [00:00<00:00, 73.84it/s, v_num=35, train_loss_step=1.150, train_loss_epoch=1.060]Epoch 8: 100%|██████████| 1/1 [00:00<00:00, 71.59it/s, v_num=35, train_loss_step=1.150, train_loss_epoch=1.150]Epoch 8: 0%| | 0/1 [00:00<?, ?it/s, v_num=35, train_loss_step=1.150, train_loss_epoch=1.150] Epoch 9: 0%| | 0/1 [00:00<?, ?it/s, v_num=35, train_loss_step=1.150, train_loss_epoch=1.150]Epoch 9: 100%|██████████| 1/1 [00:00<00:00, 17.39it/s, v_num=35, train_loss_step=1.150, train_loss_epoch=1.150]Epoch 9: 100%|██████████| 1/1 [00:00<00:00, 17.26it/s, v_num=35, train_loss_step=1.090, train_loss_epoch=1.150]
#> Validation: | | 0/? [00:00<?, ?it/s][A
#> Validation: | | 0/? [00:00<?, ?it/s][A
#> Validation DataLoader 0: 0%| | 0/1 [00:00<?, ?it/s][A
#> Validation DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 1391.15it/s][A
#> [AEpoch 9: 100%|██████████| 1/1 [00:00<00:00, 16.32it/s, v_num=35, train_loss_step=1.090, train_loss_epoch=1.150]Epoch 9: 100%|██████████| 1/1 [00:00<00:00, 16.19it/s, v_num=35, train_loss_step=1.090, train_loss_epoch=1.090]Epoch 9: 100%|██████████| 1/1 [00:00<00:00, 14.13it/s, v_num=35, train_loss_step=1.090, train_loss_epoch=1.090]
# rmok predict
rmok_point_predictions <-
rmok_fit |>
modeltime_table() |>
modeltime_forecast(actual_data = training, new_data = testing)
#> Predicting: | | 0/? [00:00<?, ?it/s]Predicting: | | 0/? [00:00<?, ?it/s]Predicting DataLoader 0: 0%| | 0/1 [00:00<?, ?it/s]Predicting DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 552.17it/s]Predicting DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 468.79it/s]
rmok_point_predictions |>
plot_modeltime_forecast(.conf_interval_show = FALSE, .interactive = FALSE)
