Step 4 - The NeuralProphet pipeline¶

This is one of three step4_model_* notebooks, one per model architecture. Read them after step 1 (data splitting), step 2 (cross-validation), and step 3 (hyperparameter tuning -- demoed on prophet_xgb). Step 3 shows where a final set of hyperparameters comes from; here we fit the model with the package defaults from params.yaml and inspect its behavior. To run this notebook with tuned parameters instead, pass config_overrides={"models": {"neuralprophet": best_params}} into run_single_its().

Goal: walk through run_single_its() using the NeuralProphetModel.

Sections:

4a. Load the pre-built dummy data.
4b. Fit NeuralProphetModel manually and inspect the FitResult.
4c. Inside NeuralProphet -- AR terms and the warmup period.
4d. Run the full pipeline via run_single_its().
4e. Inspect PipelineResult: metrics, excess table, ATE.
4f. Reproduce the counterfactual plot with annotations.

In [1]:

Copied!





%matplotlib inline

from IPython.display import display
import logging
import warnings
from pathlib import Path

import matplotlib.dates as mdates
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

warnings.filterwarnings("ignore")
logging.basicConfig(
    level=logging.INFO,
    format="%(asctime)s  %(levelname)s  %(name)s - %(message)s",
    datefmt="%H:%M:%S",
)
logging.getLogger("cmdstanpy").setLevel(logging.WARNING)
logging.getLogger("its2s").setLevel(logging.WARNING)
logging.getLogger("NP").setLevel(logging.CRITICAL)

OUT_DIR = Path.cwd() / "figures"
OUT_DIR.mkdir(exist_ok=True)
INTERVENTION = "2022-03-15"
TEST_DAYS    = 365
HOLDOUT_DAYS = 42
%matplotlib inline

from IPython.display import display
import logging
import warnings
from pathlib import Path

import matplotlib.dates as mdates
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

warnings.filterwarnings("ignore")
logging.basicConfig(
    level=logging.INFO,
    format="%(asctime)s  %(levelname)s  %(name)s - %(message)s",
    datefmt="%H:%M:%S",
)
logging.getLogger("cmdstanpy").setLevel(logging.WARNING)
logging.getLogger("its2s").setLevel(logging.WARNING)
logging.getLogger("NP").setLevel(logging.CRITICAL)

OUT_DIR = Path.cwd() / "figures"
OUT_DIR.mkdir(exist_ok=True)
INTERVENTION = "2022-03-15"
TEST_DAYS    = 365
HOLDOUT_DAYS = 42

4a. Load the pre-built dummy data¶

The series has a +8/day intervention effect baked in for 42 days after 2022-03-15.

In [2]:

Copied!





df = pd.read_csv("data/dummy_data.csv", parse_dates=["ds"])

print("=" * 60)
print("Dummy dataset (with +8/day intervention effect)")
print("=" * 60)
print(df.tail())
df = pd.read_csv("data/dummy_data.csv", parse_dates=["ds"])

print("=" * 60)
print("Dummy dataset (with +8/day intervention effect)")
print("=" * 60)
print(df.tail())

============================================================
Dummy dataset (with +8/day intervention effect)
============================================================
             ds          y  covar_linear  covar_dow  covar_noise
1571 2022-04-21  71.975041      0.995182        3.0    -0.356611
1572 2022-04-22  73.042307      1.033630        4.0     0.247103
1573 2022-04-23  72.275013      1.003513        5.0     1.129482
1574 2022-04-24  69.332534      1.004775        6.0    -0.321536
1575 2022-04-25  69.651058      0.981820        0.0    -1.057655

4b. Manual fit¶

This replicates what run_single_its does internally, so we can inspect the FitResult.

NeuralProphet trains a neural network, so this cell takes longer than ARIMA or Prophet-based models.

In [3]:

Copied!





from its2s.data_prep import prepare_splits
from its2s.models.neuralprophet import NeuralProphetModel
from its2s.settings import get_model_config, load_config

config = load_config()
splits = prepare_splits(df, INTERVENTION, split_method="days", test_days=TEST_DAYS, holdout_days=HOLDOUT_DAYS)

model_params = get_model_config(config, "neuralprophet")
model = NeuralProphetModel(params=model_params)

print("Fitting NeuralProphetModel on training data ...")
print(f"  Training rows : {len(splits.train_df)}")
print(f"  Training range: {splits.train_df['ds'].min().date()} -> {splits.train_df['ds'].max().date()}")
print(f"  n_lags        : {model_params.get('n_lags', 14)}")
print(f"  epochs        : {model_params.get('epochs', 100)}")

fit_result = model.fit(splits.train_df, target_col="y", date_col="ds")

n_nan = int(np.isnan(fit_result.fitted_values).sum())
n_lags = model_params.get("n_lags", 14)
print("\nFitResult fields:")
print(f"  fitted_values  shape = {fit_result.fitted_values.shape}")
print(f"  residuals      shape = {fit_result.residuals.shape}")
print(f"  NaN fitted values   = {n_nan}  (AR warmup = n_lags = {n_lags})")
print(f"  residuals  mean={np.nanmean(fit_result.residuals):.4f}  std={np.nanstd(fit_result.residuals):.4f}  (NaNs excluded)")
from its2s.data_prep import prepare_splits
from its2s.models.neuralprophet import NeuralProphetModel
from its2s.settings import get_model_config, load_config

config = load_config()
splits = prepare_splits(df, INTERVENTION, split_method="days", test_days=TEST_DAYS, holdout_days=HOLDOUT_DAYS)

model_params = get_model_config(config, "neuralprophet")
model = NeuralProphetModel(params=model_params)

print("Fitting NeuralProphetModel on training data ...")
print(f"  Training rows : {len(splits.train_df)}")
print(f"  Training range: {splits.train_df['ds'].min().date()} -> {splits.train_df['ds'].max().date()}")
print(f"  n_lags        : {model_params.get('n_lags', 14)}")
print(f"  epochs        : {model_params.get('epochs', 100)}")

fit_result = model.fit(splits.train_df, target_col="y", date_col="ds")

n_nan = int(np.isnan(fit_result.fitted_values).sum())
n_lags = model_params.get("n_lags", 14)
print("\nFitResult fields:")
print(f"  fitted_values  shape = {fit_result.fitted_values.shape}")
print(f"  residuals      shape = {fit_result.residuals.shape}")
print(f"  NaN fitted values   = {n_nan}  (AR warmup = n_lags = {n_lags})")
print(f"  residuals  mean={np.nanmean(fit_result.residuals):.4f}  std={np.nanstd(fit_result.residuals):.4f}  (NaNs excluded)")

Fitting NeuralProphetModel on training data ...
  Training rows : 1169
  Training range: 2018-01-01 -> 2021-03-14
  n_lags        : 14
  epochs        : 100

Training: 0it [00:00, ?it/s]

Training:   0%|          | 0/100 [00:00<?, ?it/s]

Epoch 1:   0%|          | 0/100 [00:00<?, ?it/s]

Epoch 1:   1%|          | 1/100 [00:00<00:00, 1829.18it/s]

Epoch 1:   1%|          | 1/100 [00:00<00:06, 15.30it/s, loss=1.05, v_num=16, MAE=40.40, RMSE=53.80, Loss=1.050, RegLoss=0.000]

Epoch 1:   0%|          | 0/100 [00:00<?, ?it/s, loss=1.05, v_num=16, MAE=40.40, RMSE=53.80, Loss=1.050, RegLoss=0.000]

Epoch 2:   0%|          | 0/100 [00:00<?, ?it/s, loss=1.05, v_num=16, MAE=40.40, RMSE=53.80, Loss=1.050, RegLoss=0.000]

Epoch 2:   2%|▏         | 2/100 [00:00<00:00, 4840.51it/s, loss=1.05, v_num=16, MAE=40.40, RMSE=53.80, Loss=1.050, RegLoss=0.000]

Epoch 2:   2%|▏         | 2/100 [00:00<00:01, 82.29it/s, loss=0.943, v_num=16, MAE=36.40, RMSE=49.20, Loss=0.950, RegLoss=0.000]

Epoch 2:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.943, v_num=16, MAE=36.40, RMSE=49.20, Loss=0.950, RegLoss=0.000]

Epoch 3:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.943, v_num=16, MAE=36.40, RMSE=49.20, Loss=0.950, RegLoss=0.000]

Epoch 3:   3%|▎         | 3/100 [00:00<00:00, 8202.68it/s, loss=0.943, v_num=16, MAE=36.40, RMSE=49.20, Loss=0.950, RegLoss=0.000]

Epoch 3:   3%|▎         | 3/100 [00:00<00:00, 126.27it/s, loss=0.885, v_num=16, MAE=35.30, RMSE=48.20, Loss=0.901, RegLoss=0.000]

Epoch 3:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.885, v_num=16, MAE=35.30, RMSE=48.20, Loss=0.901, RegLoss=0.000]

Epoch 4:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.885, v_num=16, MAE=35.30, RMSE=48.20, Loss=0.901, RegLoss=0.000]

Epoch 4:   4%|▍         | 4/100 [00:00<00:00, 11358.98it/s, loss=0.885, v_num=16, MAE=35.30, RMSE=48.20, Loss=0.901, RegLoss=0.000]

Epoch 4:   4%|▍         | 4/100 [00:00<00:00, 175.40it/s, loss=0.809, v_num=16, MAE=30.80, RMSE=42.80, Loss=0.787, RegLoss=0.000]

Epoch 4:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.809, v_num=16, MAE=30.80, RMSE=42.80, Loss=0.787, RegLoss=0.000]

Epoch 5:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.809, v_num=16, MAE=30.80, RMSE=42.80, Loss=0.787, RegLoss=0.000]

Epoch 5:   5%|▌         | 5/100 [00:00<00:00, 13400.33it/s, loss=0.809, v_num=16, MAE=30.80, RMSE=42.80, Loss=0.787, RegLoss=0.000]

Epoch 5:   5%|▌         | 5/100 [00:00<00:00, 215.80it/s, loss=0.725, v_num=16, MAE=29.20, RMSE=41.80, Loss=0.739, RegLoss=0.000]

Epoch 5:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.725, v_num=16, MAE=29.20, RMSE=41.80, Loss=0.739, RegLoss=0.000]

Epoch 6:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.725, v_num=16, MAE=29.20, RMSE=41.80, Loss=0.739, RegLoss=0.000]

Epoch 6:   6%|▌         | 6/100 [00:00<00:00, 17598.48it/s, loss=0.725, v_num=16, MAE=29.20, RMSE=41.80, Loss=0.739, RegLoss=0.000]

Epoch 6:   6%|▌         | 6/100 [00:00<00:00, 264.49it/s, loss=0.685, v_num=16, MAE=26.40, RMSE=38.70, Loss=0.666, RegLoss=0.000]

Epoch 6:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.685, v_num=16, MAE=26.40, RMSE=38.70, Loss=0.666, RegLoss=0.000]

Epoch 7:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.685, v_num=16, MAE=26.40, RMSE=38.70, Loss=0.666, RegLoss=0.000]

Epoch 7:   7%|▋         | 7/100 [00:00<00:00, 14322.01it/s, loss=0.685, v_num=16, MAE=26.40, RMSE=38.70, Loss=0.666, RegLoss=0.000]

Epoch 7:   7%|▋         | 7/100 [00:00<00:00, 288.97it/s, loss=0.566, v_num=16, MAE=21.90, RMSE=33.50, Loss=0.534, RegLoss=0.000]

Epoch 7:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.566, v_num=16, MAE=21.90, RMSE=33.50, Loss=0.534, RegLoss=0.000]

Epoch 8:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.566, v_num=16, MAE=21.90, RMSE=33.50, Loss=0.534, RegLoss=0.000]

Epoch 8:   8%|▊         | 8/100 [00:00<00:00, 23530.46it/s, loss=0.566, v_num=16, MAE=21.90, RMSE=33.50, Loss=0.534, RegLoss=0.000]

Epoch 8:   8%|▊         | 8/100 [00:00<00:00, 353.14it/s, loss=0.437, v_num=16, MAE=18.90, RMSE=30.10, Loss=0.452, RegLoss=0.000]

Epoch 8:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.437, v_num=16, MAE=18.90, RMSE=30.10, Loss=0.452, RegLoss=0.000]

Epoch 9:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.437, v_num=16, MAE=18.90, RMSE=30.10, Loss=0.452, RegLoss=0.000]

Epoch 9:   9%|▉         | 9/100 [00:00<00:00, 24136.02it/s, loss=0.437, v_num=16, MAE=18.90, RMSE=30.10, Loss=0.452, RegLoss=0.000]

Epoch 9:   9%|▉         | 9/100 [00:00<00:00, 392.19it/s, loss=0.365, v_num=16, MAE=16.40, RMSE=27.40, Loss=0.382, RegLoss=0.000]

Epoch 9:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.365, v_num=16, MAE=16.40, RMSE=27.40, Loss=0.382, RegLoss=0.000]

Epoch 10:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.365, v_num=16, MAE=16.40, RMSE=27.40, Loss=0.382, RegLoss=0.000]

Epoch 10:  10%|█         | 10/100 [00:00<00:00, 23198.58it/s, loss=0.365, v_num=16, MAE=16.40, RMSE=27.40, Loss=0.382, RegLoss=0.000]

Epoch 10:  10%|█         | 10/100 [00:00<00:00, 419.99it/s, loss=0.331, v_num=16, MAE=14.50, RMSE=25.50, Loss=0.341, RegLoss=0.000]

Epoch 10:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.331, v_num=16, MAE=14.50, RMSE=25.50, Loss=0.341, RegLoss=0.000]

Epoch 11:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.331, v_num=16, MAE=14.50, RMSE=25.50, Loss=0.341, RegLoss=0.000]

Epoch 11:  11%|█         | 11/100 [00:00<00:00, 30615.36it/s, loss=0.331, v_num=16, MAE=14.50, RMSE=25.50, Loss=0.341, RegLoss=0.000]

Epoch 11:  11%|█         | 11/100 [00:00<00:00, 479.30it/s, loss=0.268, v_num=16, MAE=11.20, RMSE=20.70, Loss=0.245, RegLoss=0.000]

Epoch 11:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.268, v_num=16, MAE=11.20, RMSE=20.70, Loss=0.245, RegLoss=0.000]

Epoch 12:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.268, v_num=16, MAE=11.20, RMSE=20.70, Loss=0.245, RegLoss=0.000]

Epoch 12:  12%|█▏        | 12/100 [00:00<00:00, 31916.07it/s, loss=0.268, v_num=16, MAE=11.20, RMSE=20.70, Loss=0.245, RegLoss=0.000]

Epoch 12:  12%|█▏        | 12/100 [00:00<00:00, 526.16it/s, loss=0.187, v_num=16, MAE=9.140, RMSE=17.30, Loss=0.195, RegLoss=0.000]

Epoch 12:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.187, v_num=16, MAE=9.140, RMSE=17.30, Loss=0.195, RegLoss=0.000]

Epoch 13:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.187, v_num=16, MAE=9.140, RMSE=17.30, Loss=0.195, RegLoss=0.000]

Epoch 13:  13%|█▎        | 13/100 [00:00<00:00, 28502.85it/s, loss=0.187, v_num=16, MAE=9.140, RMSE=17.30, Loss=0.195, RegLoss=0.000]

Epoch 13:  13%|█▎        | 13/100 [00:00<00:00, 558.68it/s, loss=0.147, v_num=16, MAE=7.440, RMSE=14.20, Loss=0.154, RegLoss=0.000]

Epoch 13:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.147, v_num=16, MAE=7.440, RMSE=14.20, Loss=0.154, RegLoss=0.000]

Epoch 14:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.147, v_num=16, MAE=7.440, RMSE=14.20, Loss=0.154, RegLoss=0.000]

Epoch 14:  14%|█▍        | 14/100 [00:00<00:00, 43304.02it/s, loss=0.147, v_num=16, MAE=7.440, RMSE=14.20, Loss=0.154, RegLoss=0.000]

Epoch 14:  14%|█▍        | 14/100 [00:00<00:00, 586.23it/s, loss=0.12, v_num=16, MAE=6.520, RMSE=11.70, Loss=0.125, RegLoss=0.000]

Epoch 14:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.12, v_num=16, MAE=6.520, RMSE=11.70, Loss=0.125, RegLoss=0.000]

Epoch 15:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.12, v_num=16, MAE=6.520, RMSE=11.70, Loss=0.125, RegLoss=0.000]

Epoch 15:  15%|█▌        | 15/100 [00:00<00:00, 41775.94it/s, loss=0.12, v_num=16, MAE=6.520, RMSE=11.70, Loss=0.125, RegLoss=0.000]

Epoch 15:  15%|█▌        | 15/100 [00:00<00:00, 637.34it/s, loss=0.0937, v_num=16, MAE=5.230, RMSE=8.630, Loss=0.0901, RegLoss=0.000]

Epoch 15:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0937, v_num=16, MAE=5.230, RMSE=8.630, Loss=0.0901, RegLoss=0.000]

Epoch 16:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0937, v_num=16, MAE=5.230, RMSE=8.630, Loss=0.0901, RegLoss=0.000]

Epoch 16:  16%|█▌        | 16/100 [00:00<00:00, 42313.28it/s, loss=0.0937, v_num=16, MAE=5.230, RMSE=8.630, Loss=0.0901, RegLoss=0.000]

Epoch 16:  16%|█▌        | 16/100 [00:00<00:00, 674.68it/s, loss=0.054, v_num=16, MAE=3.760, RMSE=5.440, Loss=0.0479, RegLoss=0.000]

Epoch 16:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.054, v_num=16, MAE=3.760, RMSE=5.440, Loss=0.0479, RegLoss=0.000]

Epoch 17:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.054, v_num=16, MAE=3.760, RMSE=5.440, Loss=0.0479, RegLoss=0.000]

Epoch 17:  17%|█▋        | 17/100 [00:00<00:00, 49140.71it/s, loss=0.054, v_num=16, MAE=3.760, RMSE=5.440, Loss=0.0479, RegLoss=0.000]

Epoch 17:  17%|█▋        | 17/100 [00:00<00:00, 729.24it/s, loss=0.0247, v_num=16, MAE=2.900, RMSE=3.680, Loss=0.0255, RegLoss=0.000]

Epoch 17:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0247, v_num=16, MAE=2.900, RMSE=3.680, Loss=0.0255, RegLoss=0.000]

Epoch 18:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0247, v_num=16, MAE=2.900, RMSE=3.680, Loss=0.0255, RegLoss=0.000]

Epoch 18:  18%|█▊        | 18/100 [00:00<00:00, 49965.24it/s, loss=0.0247, v_num=16, MAE=2.900, RMSE=3.680, Loss=0.0255, RegLoss=0.000]

Epoch 18:  18%|█▊        | 18/100 [00:00<00:00, 765.40it/s, loss=0.0179, v_num=16, MAE=2.360, RMSE=2.900, Loss=0.0158, RegLoss=0.000]

Epoch 18:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0179, v_num=16, MAE=2.360, RMSE=2.900, Loss=0.0158, RegLoss=0.000]

Epoch 19:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0179, v_num=16, MAE=2.360, RMSE=2.900, Loss=0.0158, RegLoss=0.000]

Epoch 19:  19%|█▉        | 19/100 [00:00<00:00, 60099.38it/s, loss=0.0179, v_num=16, MAE=2.360, RMSE=2.900, Loss=0.0158, RegLoss=0.000]

Epoch 19:  19%|█▉        | 19/100 [00:00<00:00, 832.72it/s, loss=0.0138, v_num=16, MAE=2.210, RMSE=2.770, Loss=0.0143, RegLoss=0.000]

Epoch 19:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0138, v_num=16, MAE=2.210, RMSE=2.770, Loss=0.0143, RegLoss=0.000]

Epoch 20:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0138, v_num=16, MAE=2.210, RMSE=2.770, Loss=0.0143, RegLoss=0.000]

Epoch 20:  20%|██        | 20/100 [00:00<00:00, 68478.43it/s, loss=0.0138, v_num=16, MAE=2.210, RMSE=2.770, Loss=0.0143, RegLoss=0.000]

Epoch 20:  20%|██        | 20/100 [00:00<00:00, 857.67it/s, loss=0.0132, v_num=16, MAE=2.100, RMSE=2.610, Loss=0.0129, RegLoss=0.000]

Epoch 20:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0132, v_num=16, MAE=2.100, RMSE=2.610, Loss=0.0129, RegLoss=0.000]

Epoch 21:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0132, v_num=16, MAE=2.100, RMSE=2.610, Loss=0.0129, RegLoss=0.000]

Epoch 21:  21%|██        | 21/100 [00:00<00:00, 59154.05it/s, loss=0.0132, v_num=16, MAE=2.100, RMSE=2.610, Loss=0.0129, RegLoss=0.000]

Epoch 21:  21%|██        | 21/100 [00:00<00:00, 895.48it/s, loss=0.0117, v_num=16, MAE=2.030, RMSE=2.520, Loss=0.0122, RegLoss=0.000]

Epoch 21:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0117, v_num=16, MAE=2.030, RMSE=2.520, Loss=0.0122, RegLoss=0.000]

Epoch 22:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0117, v_num=16, MAE=2.030, RMSE=2.520, Loss=0.0122, RegLoss=0.000]

Epoch 22:  22%|██▏       | 22/100 [00:00<00:00, 60787.01it/s, loss=0.0117, v_num=16, MAE=2.030, RMSE=2.520, Loss=0.0122, RegLoss=0.000]

Epoch 22:  22%|██▏       | 22/100 [00:00<00:00, 937.88it/s, loss=0.012, v_num=16, MAE=2.120, RMSE=2.620, Loss=0.0126, RegLoss=0.000]

Epoch 22:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.012, v_num=16, MAE=2.120, RMSE=2.620, Loss=0.0126, RegLoss=0.000]

Epoch 23:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.012, v_num=16, MAE=2.120, RMSE=2.620, Loss=0.0126, RegLoss=0.000]

Epoch 23:  23%|██▎       | 23/100 [00:00<00:00, 66714.38it/s, loss=0.012, v_num=16, MAE=2.120, RMSE=2.620, Loss=0.0126, RegLoss=0.000]

Epoch 23:  23%|██▎       | 23/100 [00:00<00:00, 976.02it/s, loss=0.0142, v_num=16, MAE=2.200, RMSE=2.710, Loss=0.0143, RegLoss=0.000]

Epoch 23:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0142, v_num=16, MAE=2.200, RMSE=2.710, Loss=0.0143, RegLoss=0.000]

Epoch 24:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0142, v_num=16, MAE=2.200, RMSE=2.710, Loss=0.0143, RegLoss=0.000]

Epoch 24:  24%|██▍       | 24/100 [00:00<00:00, 64198.53it/s, loss=0.0142, v_num=16, MAE=2.200, RMSE=2.710, Loss=0.0143, RegLoss=0.000]

Epoch 24:  24%|██▍       | 24/100 [00:00<00:00, 1028.94it/s, loss=0.015, v_num=16, MAE=2.140, RMSE=2.660, Loss=0.0133, RegLoss=0.000]

Epoch 24:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.015, v_num=16, MAE=2.140, RMSE=2.660, Loss=0.0133, RegLoss=0.000]

Epoch 25:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.015, v_num=16, MAE=2.140, RMSE=2.660, Loss=0.0133, RegLoss=0.000]

Epoch 25:  25%|██▌       | 25/100 [00:00<00:00, 77614.80it/s, loss=0.015, v_num=16, MAE=2.140, RMSE=2.660, Loss=0.0133, RegLoss=0.000]

Epoch 25:  25%|██▌       | 25/100 [00:00<00:00, 1093.18it/s, loss=0.0132, v_num=16, MAE=2.110, RMSE=2.670, Loss=0.0132, RegLoss=0.000]

Epoch 25:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0132, v_num=16, MAE=2.110, RMSE=2.670, Loss=0.0132, RegLoss=0.000]

Epoch 26:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0132, v_num=16, MAE=2.110, RMSE=2.670, Loss=0.0132, RegLoss=0.000]

Epoch 26:  26%|██▌       | 26/100 [00:00<00:00, 69504.08it/s, loss=0.0132, v_num=16, MAE=2.110, RMSE=2.670, Loss=0.0132, RegLoss=0.000]

Epoch 26:  26%|██▌       | 26/100 [00:00<00:00, 1110.02it/s, loss=0.0122, v_num=16, MAE=2.070, RMSE=2.570, Loss=0.0122, RegLoss=0.000]

Epoch 26:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0122, v_num=16, MAE=2.070, RMSE=2.570, Loss=0.0122, RegLoss=0.000]

Epoch 27:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0122, v_num=16, MAE=2.070, RMSE=2.570, Loss=0.0122, RegLoss=0.000]

Epoch 27:  27%|██▋       | 27/100 [00:00<00:00, 76004.17it/s, loss=0.0122, v_num=16, MAE=2.070, RMSE=2.570, Loss=0.0122, RegLoss=0.000]

Epoch 27:  27%|██▋       | 27/100 [00:00<00:00, 1179.91it/s, loss=0.0121, v_num=16, MAE=2.050, RMSE=2.560, Loss=0.012, RegLoss=0.000]

Epoch 27:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0121, v_num=16, MAE=2.050, RMSE=2.560, Loss=0.012, RegLoss=0.000]

Epoch 28:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0121, v_num=16, MAE=2.050, RMSE=2.560, Loss=0.012, RegLoss=0.000]

Epoch 28:  28%|██▊       | 28/100 [00:00<00:00, 78925.08it/s, loss=0.0121, v_num=16, MAE=2.050, RMSE=2.560, Loss=0.012, RegLoss=0.000]

Epoch 28:  28%|██▊       | 28/100 [00:00<00:00, 1188.86it/s, loss=0.0123, v_num=16, MAE=2.080, RMSE=2.570, Loss=0.0126, RegLoss=0.000]

Epoch 28:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0123, v_num=16, MAE=2.080, RMSE=2.570, Loss=0.0126, RegLoss=0.000]

Epoch 29:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0123, v_num=16, MAE=2.080, RMSE=2.570, Loss=0.0126, RegLoss=0.000]

Epoch 29:  29%|██▉       | 29/100 [00:00<00:00, 63450.61it/s, loss=0.0123, v_num=16, MAE=2.080, RMSE=2.570, Loss=0.0126, RegLoss=0.000]

Epoch 29:  29%|██▉       | 29/100 [00:00<00:00, 1212.88it/s, loss=0.0125, v_num=16, MAE=2.020, RMSE=2.490, Loss=0.0118, RegLoss=0.000]

Epoch 29:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0125, v_num=16, MAE=2.020, RMSE=2.490, Loss=0.0118, RegLoss=0.000]

Epoch 30:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0125, v_num=16, MAE=2.020, RMSE=2.490, Loss=0.0118, RegLoss=0.000]

Epoch 30:  30%|███       | 30/100 [00:00<00:00, 78349.39it/s, loss=0.0125, v_num=16, MAE=2.020, RMSE=2.490, Loss=0.0118, RegLoss=0.000]

Epoch 30:  30%|███       | 30/100 [00:00<00:00, 1286.62it/s, loss=0.0112, v_num=16, MAE=1.990, RMSE=2.500, Loss=0.0116, RegLoss=0.000]

Epoch 30:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0112, v_num=16, MAE=1.990, RMSE=2.500, Loss=0.0116, RegLoss=0.000]

Epoch 31:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0112, v_num=16, MAE=1.990, RMSE=2.500, Loss=0.0116, RegLoss=0.000]

Epoch 31:  31%|███       | 31/100 [00:00<00:00, 90925.47it/s, loss=0.0112, v_num=16, MAE=1.990, RMSE=2.500, Loss=0.0116, RegLoss=0.000]

Epoch 31:  31%|███       | 31/100 [00:00<00:00, 1341.28it/s, loss=0.0116, v_num=16, MAE=2.050, RMSE=2.510, Loss=0.0116, RegLoss=0.000]

Epoch 31:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0116, v_num=16, MAE=2.050, RMSE=2.510, Loss=0.0116, RegLoss=0.000]

Epoch 32:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0116, v_num=16, MAE=2.050, RMSE=2.510, Loss=0.0116, RegLoss=0.000]

Epoch 32:  32%|███▏      | 32/100 [00:00<00:00, 88885.91it/s, loss=0.0116, v_num=16, MAE=2.050, RMSE=2.510, Loss=0.0116, RegLoss=0.000]

Epoch 32:  32%|███▏      | 32/100 [00:00<00:00, 1354.20it/s, loss=0.0109, v_num=16, MAE=1.920, RMSE=2.390, Loss=0.0109, RegLoss=0.000]

Epoch 32:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0109, v_num=16, MAE=1.920, RMSE=2.390, Loss=0.0109, RegLoss=0.000]

Epoch 33:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0109, v_num=16, MAE=1.920, RMSE=2.390, Loss=0.0109, RegLoss=0.000]

Epoch 33:  33%|███▎      | 33/100 [00:00<00:00, 85704.04it/s, loss=0.0109, v_num=16, MAE=1.920, RMSE=2.390, Loss=0.0109, RegLoss=0.000]

Epoch 33:  33%|███▎      | 33/100 [00:00<00:00, 1376.02it/s, loss=0.0107, v_num=16, MAE=1.990, RMSE=2.460, Loss=0.0112, RegLoss=0.000]

Epoch 33:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0107, v_num=16, MAE=1.990, RMSE=2.460, Loss=0.0112, RegLoss=0.000]

Epoch 34:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0107, v_num=16, MAE=1.990, RMSE=2.460, Loss=0.0112, RegLoss=0.000]

Epoch 34:  34%|███▍      | 34/100 [00:00<00:00, 94629.29it/s, loss=0.0107, v_num=16, MAE=1.990, RMSE=2.460, Loss=0.0112, RegLoss=0.000]

Epoch 34:  34%|███▍      | 34/100 [00:00<00:00, 1468.31it/s, loss=0.0114, v_num=16, MAE=1.960, RMSE=2.430, Loss=0.0113, RegLoss=0.000]

Epoch 34:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0114, v_num=16, MAE=1.960, RMSE=2.430, Loss=0.0113, RegLoss=0.000]

Epoch 35:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0114, v_num=16, MAE=1.960, RMSE=2.430, Loss=0.0113, RegLoss=0.000]

Epoch 35:  35%|███▌      | 35/100 [00:00<00:00, 98523.92it/s, loss=0.0114, v_num=16, MAE=1.960, RMSE=2.430, Loss=0.0113, RegLoss=0.000]

Epoch 35:  35%|███▌      | 35/100 [00:00<00:00, 1531.18it/s, loss=0.0103, v_num=16, MAE=1.930, RMSE=2.400, Loss=0.0108, RegLoss=0.000]

Epoch 35:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0103, v_num=16, MAE=1.930, RMSE=2.400, Loss=0.0108, RegLoss=0.000]

Epoch 36:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0103, v_num=16, MAE=1.930, RMSE=2.400, Loss=0.0108, RegLoss=0.000]

Epoch 36:  36%|███▌      | 36/100 [00:00<00:00, 101680.10it/s, loss=0.0103, v_num=16, MAE=1.930, RMSE=2.400, Loss=0.0108, RegLoss=0.000]

Epoch 36:  36%|███▌      | 36/100 [00:00<00:00, 1534.97it/s, loss=0.0107, v_num=16, MAE=1.950, RMSE=2.420, Loss=0.0109, RegLoss=0.000]

Epoch 36:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0107, v_num=16, MAE=1.950, RMSE=2.420, Loss=0.0109, RegLoss=0.000]

Epoch 37:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0107, v_num=16, MAE=1.950, RMSE=2.420, Loss=0.0109, RegLoss=0.000]

Epoch 37:  37%|███▋      | 37/100 [00:00<00:00, 112455.98it/s, loss=0.0107, v_num=16, MAE=1.950, RMSE=2.420, Loss=0.0109, RegLoss=0.000]

Epoch 37:  37%|███▋      | 37/100 [00:00<00:00, 1579.65it/s, loss=0.011, v_num=16, MAE=1.970, RMSE=2.430, Loss=0.0109, RegLoss=0.000]

Epoch 37:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.011, v_num=16, MAE=1.970, RMSE=2.430, Loss=0.0109, RegLoss=0.000]

Epoch 38:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.011, v_num=16, MAE=1.970, RMSE=2.430, Loss=0.0109, RegLoss=0.000]

Epoch 38:  38%|███▊      | 38/100 [00:00<00:00, 110836.96it/s, loss=0.011, v_num=16, MAE=1.970, RMSE=2.430, Loss=0.0109, RegLoss=0.000]

Epoch 38:  38%|███▊      | 38/100 [00:00<00:00, 1599.47it/s, loss=0.0111, v_num=16, MAE=1.960, RMSE=2.410, Loss=0.0111, RegLoss=0.000]

Epoch 38:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0111, v_num=16, MAE=1.960, RMSE=2.410, Loss=0.0111, RegLoss=0.000]

Epoch 39:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0111, v_num=16, MAE=1.960, RMSE=2.410, Loss=0.0111, RegLoss=0.000]

Epoch 39:  39%|███▉      | 39/100 [00:00<00:00, 102108.52it/s, loss=0.0111, v_num=16, MAE=1.960, RMSE=2.410, Loss=0.0111, RegLoss=0.000]

Epoch 39:  39%|███▉      | 39/100 [00:00<00:00, 1683.07it/s, loss=0.0114, v_num=16, MAE=1.950, RMSE=2.430, Loss=0.0109, RegLoss=0.000]

Epoch 39:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0114, v_num=16, MAE=1.950, RMSE=2.430, Loss=0.0109, RegLoss=0.000]

Epoch 40:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0114, v_num=16, MAE=1.950, RMSE=2.430, Loss=0.0109, RegLoss=0.000]

Epoch 40:  40%|████      | 40/100 [00:00<00:00, 115545.56it/s, loss=0.0114, v_num=16, MAE=1.950, RMSE=2.430, Loss=0.0109, RegLoss=0.000]

Epoch 40:  40%|████      | 40/100 [00:00<00:00, 1732.41it/s, loss=0.011, v_num=16, MAE=1.950, RMSE=2.400, Loss=0.0108, RegLoss=0.000]

Epoch 40:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.011, v_num=16, MAE=1.950, RMSE=2.400, Loss=0.0108, RegLoss=0.000]

Epoch 41:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.011, v_num=16, MAE=1.950, RMSE=2.400, Loss=0.0108, RegLoss=0.000]

Epoch 41:  41%|████      | 41/100 [00:00<00:00, 113584.19it/s, loss=0.011, v_num=16, MAE=1.950, RMSE=2.400, Loss=0.0108, RegLoss=0.000]

Epoch 41:  41%|████      | 41/100 [00:00<00:00, 1786.72it/s, loss=0.0106, v_num=16, MAE=1.920, RMSE=2.380, Loss=0.0105, RegLoss=0.000]

Epoch 41:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0106, v_num=16, MAE=1.920, RMSE=2.380, Loss=0.0105, RegLoss=0.000]

Epoch 42:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0106, v_num=16, MAE=1.920, RMSE=2.380, Loss=0.0105, RegLoss=0.000]

Epoch 42:  42%|████▏     | 42/100 [00:00<00:00, 119269.31it/s, loss=0.0106, v_num=16, MAE=1.920, RMSE=2.380, Loss=0.0105, RegLoss=0.000]

Epoch 42:  42%|████▏     | 42/100 [00:00<00:00, 1842.18it/s, loss=0.0105, v_num=16, MAE=1.920, RMSE=2.370, Loss=0.0106, RegLoss=0.000]

Epoch 42:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0105, v_num=16, MAE=1.920, RMSE=2.370, Loss=0.0106, RegLoss=0.000]

Epoch 43:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0105, v_num=16, MAE=1.920, RMSE=2.370, Loss=0.0106, RegLoss=0.000]

Epoch 43:  43%|████▎     | 43/100 [00:00<00:00, 130692.08it/s, loss=0.0105, v_num=16, MAE=1.920, RMSE=2.370, Loss=0.0106, RegLoss=0.000]

Epoch 43:  43%|████▎     | 43/100 [00:00<00:00, 1824.33it/s, loss=0.0105, v_num=16, MAE=1.860, RMSE=2.330, Loss=0.0101, RegLoss=0.000]

Epoch 43:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0105, v_num=16, MAE=1.860, RMSE=2.330, Loss=0.0101, RegLoss=0.000]

Epoch 44:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0105, v_num=16, MAE=1.860, RMSE=2.330, Loss=0.0101, RegLoss=0.000]

Epoch 44:  44%|████▍     | 44/100 [00:00<00:00, 129690.36it/s, loss=0.0105, v_num=16, MAE=1.860, RMSE=2.330, Loss=0.0101, RegLoss=0.000]

Epoch 44:  44%|████▍     | 44/100 [00:00<00:00, 1880.74it/s, loss=0.0103, v_num=16, MAE=1.860, RMSE=2.340, Loss=0.0103, RegLoss=0.000]

Epoch 44:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0103, v_num=16, MAE=1.860, RMSE=2.340, Loss=0.0103, RegLoss=0.000]

Epoch 45:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0103, v_num=16, MAE=1.860, RMSE=2.340, Loss=0.0103, RegLoss=0.000]

Epoch 45:  45%|████▌     | 45/100 [00:00<00:00, 128659.63it/s, loss=0.0103, v_num=16, MAE=1.860, RMSE=2.340, Loss=0.0103, RegLoss=0.000]

Epoch 45:  45%|████▌     | 45/100 [00:00<00:00, 1962.01it/s, loss=0.0106, v_num=16, MAE=1.890, RMSE=2.360, Loss=0.0104, RegLoss=0.000]

Epoch 45:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0106, v_num=16, MAE=1.890, RMSE=2.360, Loss=0.0104, RegLoss=0.000]

Epoch 46:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0106, v_num=16, MAE=1.890, RMSE=2.360, Loss=0.0104, RegLoss=0.000]

Epoch 46:  46%|████▌     | 46/100 [00:00<00:00, 141139.71it/s, loss=0.0106, v_num=16, MAE=1.890, RMSE=2.360, Loss=0.0104, RegLoss=0.000]

Epoch 46:  46%|████▌     | 46/100 [00:00<00:00, 1967.09it/s, loss=0.0101, v_num=16, MAE=1.860, RMSE=2.320, Loss=0.010, RegLoss=0.000]

Epoch 46:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0101, v_num=16, MAE=1.860, RMSE=2.320, Loss=0.010, RegLoss=0.000]

Epoch 47:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0101, v_num=16, MAE=1.860, RMSE=2.320, Loss=0.010, RegLoss=0.000]

Epoch 47:  47%|████▋     | 47/100 [00:00<00:00, 127018.23it/s, loss=0.0101, v_num=16, MAE=1.860, RMSE=2.320, Loss=0.010, RegLoss=0.000]

Epoch 47:  47%|████▋     | 47/100 [00:00<00:00, 1984.38it/s, loss=0.00983, v_num=16, MAE=1.880, RMSE=2.330, Loss=0.0102, RegLoss=0.000]

Epoch 47:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00983, v_num=16, MAE=1.880, RMSE=2.330, Loss=0.0102, RegLoss=0.000]

Epoch 48:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00983, v_num=16, MAE=1.880, RMSE=2.330, Loss=0.0102, RegLoss=0.000]

Epoch 48:  48%|████▊     | 48/100 [00:00<00:00, 117941.76it/s, loss=0.00983, v_num=16, MAE=1.880, RMSE=2.330, Loss=0.0102, RegLoss=0.000]

Epoch 48:  48%|████▊     | 48/100 [00:00<00:00, 2075.83it/s, loss=0.00973, v_num=16, MAE=1.770, RMSE=2.210, Loss=0.00931, RegLoss=0.000]

Epoch 48:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00973, v_num=16, MAE=1.770, RMSE=2.210, Loss=0.00931, RegLoss=0.000]

Epoch 49:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00973, v_num=16, MAE=1.770, RMSE=2.210, Loss=0.00931, RegLoss=0.000]

Epoch 49:  49%|████▉     | 49/100 [00:00<00:00, 144124.05it/s, loss=0.00973, v_num=16, MAE=1.770, RMSE=2.210, Loss=0.00931, RegLoss=0.000]

Epoch 49:  49%|████▉     | 49/100 [00:00<00:00, 2152.91it/s, loss=0.00908, v_num=16, MAE=1.840, RMSE=2.270, Loss=0.00949, RegLoss=0.000]

Epoch 49:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00908, v_num=16, MAE=1.840, RMSE=2.270, Loss=0.00949, RegLoss=0.000]

Epoch 50:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00908, v_num=16, MAE=1.840, RMSE=2.270, Loss=0.00949, RegLoss=0.000]

Epoch 50:  50%|█████     | 50/100 [00:00<00:00, 148734.18it/s, loss=0.00908, v_num=16, MAE=1.840, RMSE=2.270, Loss=0.00949, RegLoss=0.000]

Epoch 50:  50%|█████     | 50/100 [00:00<00:00, 2144.90it/s, loss=0.00907, v_num=16, MAE=1.750, RMSE=2.190, Loss=0.00922, RegLoss=0.000]

Epoch 50:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00907, v_num=16, MAE=1.750, RMSE=2.190, Loss=0.00922, RegLoss=0.000]

Epoch 51:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00907, v_num=16, MAE=1.750, RMSE=2.190, Loss=0.00922, RegLoss=0.000]

Epoch 51:  51%|█████     | 51/100 [00:00<00:00, 142796.73it/s, loss=0.00907, v_num=16, MAE=1.750, RMSE=2.190, Loss=0.00922, RegLoss=0.000]

Epoch 51:  51%|█████     | 51/100 [00:00<00:00, 2229.20it/s, loss=0.00873, v_num=16, MAE=1.770, RMSE=2.210, Loss=0.00917, RegLoss=0.000]

Epoch 51:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00873, v_num=16, MAE=1.770, RMSE=2.210, Loss=0.00917, RegLoss=0.000]

Epoch 52:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00873, v_num=16, MAE=1.770, RMSE=2.210, Loss=0.00917, RegLoss=0.000]

Epoch 52:  52%|█████▏    | 52/100 [00:00<00:00, 148067.76it/s, loss=0.00873, v_num=16, MAE=1.770, RMSE=2.210, Loss=0.00917, RegLoss=0.000]

Epoch 52:  52%|█████▏    | 52/100 [00:00<00:00, 2212.01it/s, loss=0.00983, v_num=16, MAE=1.900, RMSE=2.330, Loss=0.0103, RegLoss=0.000]

Epoch 52:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00983, v_num=16, MAE=1.900, RMSE=2.330, Loss=0.0103, RegLoss=0.000]

Epoch 53:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00983, v_num=16, MAE=1.900, RMSE=2.330, Loss=0.0103, RegLoss=0.000]

Epoch 53:  53%|█████▎    | 53/100 [00:00<00:00, 166765.28it/s, loss=0.00983, v_num=16, MAE=1.900, RMSE=2.330, Loss=0.0103, RegLoss=0.000]

Epoch 53:  53%|█████▎    | 53/100 [00:00<00:00, 2343.48it/s, loss=0.00995, v_num=16, MAE=1.770, RMSE=2.210, Loss=0.00912, RegLoss=0.000]

Epoch 53:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00995, v_num=16, MAE=1.770, RMSE=2.210, Loss=0.00912, RegLoss=0.000]

Epoch 54:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00995, v_num=16, MAE=1.770, RMSE=2.210, Loss=0.00912, RegLoss=0.000]

Epoch 54:  54%|█████▍    | 54/100 [00:00<00:00, 192268.60it/s, loss=0.00995, v_num=16, MAE=1.770, RMSE=2.210, Loss=0.00912, RegLoss=0.000]

Epoch 54:  54%|█████▍    | 54/100 [00:00<00:00, 2332.81it/s, loss=0.00883, v_num=16, MAE=1.760, RMSE=2.200, Loss=0.00909, RegLoss=0.000]

Epoch 54:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00883, v_num=16, MAE=1.760, RMSE=2.200, Loss=0.00909, RegLoss=0.000]

Epoch 55:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00883, v_num=16, MAE=1.760, RMSE=2.200, Loss=0.00909, RegLoss=0.000]

Epoch 55:  55%|█████▌    | 55/100 [00:00<00:00, 155031.40it/s, loss=0.00883, v_num=16, MAE=1.760, RMSE=2.200, Loss=0.00909, RegLoss=0.000]

Epoch 55:  55%|█████▌    | 55/100 [00:00<00:00, 2367.74it/s, loss=0.00895, v_num=16, MAE=1.790, RMSE=2.220, Loss=0.00925, RegLoss=0.000]

Epoch 55:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00895, v_num=16, MAE=1.790, RMSE=2.220, Loss=0.00925, RegLoss=0.000]

Epoch 56:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00895, v_num=16, MAE=1.790, RMSE=2.220, Loss=0.00925, RegLoss=0.000]

Epoch 56:  56%|█████▌    | 56/100 [00:00<00:00, 170822.56it/s, loss=0.00895, v_num=16, MAE=1.790, RMSE=2.220, Loss=0.00925, RegLoss=0.000]

Epoch 56:  56%|█████▌    | 56/100 [00:00<00:00, 2386.74it/s, loss=0.00918, v_num=16, MAE=1.810, RMSE=2.240, Loss=0.00927, RegLoss=0.000]

Epoch 56:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00918, v_num=16, MAE=1.810, RMSE=2.240, Loss=0.00927, RegLoss=0.000]

Epoch 57:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00918, v_num=16, MAE=1.810, RMSE=2.240, Loss=0.00927, RegLoss=0.000]

Epoch 57:  57%|█████▋    | 57/100 [00:00<00:00, 166255.44it/s, loss=0.00918, v_num=16, MAE=1.810, RMSE=2.240, Loss=0.00927, RegLoss=0.000]

Epoch 57:  57%|█████▋    | 57/100 [00:00<00:00, 2438.50it/s, loss=0.00897, v_num=16, MAE=1.760, RMSE=2.190, Loss=0.00904, RegLoss=0.000]

Epoch 57:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00897, v_num=16, MAE=1.760, RMSE=2.190, Loss=0.00904, RegLoss=0.000]

Epoch 58:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00897, v_num=16, MAE=1.760, RMSE=2.190, Loss=0.00904, RegLoss=0.000]

Epoch 58:  58%|█████▊    | 58/100 [00:00<00:00, 158172.71it/s, loss=0.00897, v_num=16, MAE=1.760, RMSE=2.190, Loss=0.00904, RegLoss=0.000]

Epoch 58:  58%|█████▊    | 58/100 [00:00<00:00, 2478.98it/s, loss=0.00907, v_num=16, MAE=1.770, RMSE=2.240, Loss=0.00935, RegLoss=0.000]

Epoch 58:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00907, v_num=16, MAE=1.770, RMSE=2.240, Loss=0.00935, RegLoss=0.000]

Epoch 59:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00907, v_num=16, MAE=1.770, RMSE=2.240, Loss=0.00935, RegLoss=0.000]

Epoch 59:  59%|█████▉    | 59/100 [00:00<00:00, 168457.41it/s, loss=0.00907, v_num=16, MAE=1.770, RMSE=2.240, Loss=0.00935, RegLoss=0.000]

Epoch 59:  59%|█████▉    | 59/100 [00:00<00:00, 2530.46it/s, loss=0.0109, v_num=16, MAE=1.910, RMSE=2.330, Loss=0.0109, RegLoss=0.000]

Epoch 59:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0109, v_num=16, MAE=1.910, RMSE=2.330, Loss=0.0109, RegLoss=0.000]

Epoch 60:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0109, v_num=16, MAE=1.910, RMSE=2.330, Loss=0.0109, RegLoss=0.000]

Epoch 60:  60%|██████    | 60/100 [00:00<00:00, 178102.08it/s, loss=0.0109, v_num=16, MAE=1.910, RMSE=2.330, Loss=0.0109, RegLoss=0.000]

Epoch 60:  60%|██████    | 60/100 [00:00<00:00, 2657.82it/s, loss=0.011, v_num=16, MAE=1.830, RMSE=2.250, Loss=0.00934, RegLoss=0.000]

Epoch 60:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.011, v_num=16, MAE=1.830, RMSE=2.250, Loss=0.00934, RegLoss=0.000]

Epoch 61:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.011, v_num=16, MAE=1.830, RMSE=2.250, Loss=0.00934, RegLoss=0.000]

Epoch 61:  61%|██████    | 61/100 [00:00<00:00, 176328.42it/s, loss=0.011, v_num=16, MAE=1.830, RMSE=2.250, Loss=0.00934, RegLoss=0.000]

Epoch 61:  61%|██████    | 61/100 [00:00<00:00, 2603.94it/s, loss=0.00937, v_num=16, MAE=1.800, RMSE=2.230, Loss=0.00933, RegLoss=0.000]

Epoch 61:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00937, v_num=16, MAE=1.800, RMSE=2.230, Loss=0.00933, RegLoss=0.000]

Epoch 62:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00937, v_num=16, MAE=1.800, RMSE=2.230, Loss=0.00933, RegLoss=0.000]

Epoch 62:  62%|██████▏   | 62/100 [00:00<00:00, 205246.13it/s, loss=0.00937, v_num=16, MAE=1.800, RMSE=2.230, Loss=0.00933, RegLoss=0.000]

Epoch 62:  62%|██████▏   | 62/100 [00:00<00:00, 2705.16it/s, loss=0.00999, v_num=16, MAE=1.820, RMSE=2.280, Loss=0.00998, RegLoss=0.000]

Epoch 62:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00999, v_num=16, MAE=1.820, RMSE=2.280, Loss=0.00998, RegLoss=0.000]

Epoch 63:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00999, v_num=16, MAE=1.820, RMSE=2.280, Loss=0.00998, RegLoss=0.000]

Epoch 63:  63%|██████▎   | 63/100 [00:00<00:00, 176868.24it/s, loss=0.00999, v_num=16, MAE=1.820, RMSE=2.280, Loss=0.00998, RegLoss=0.000]

Epoch 63:  63%|██████▎   | 63/100 [00:00<00:00, 2640.83it/s, loss=0.00975, v_num=16, MAE=1.770, RMSE=2.190, Loss=0.00904, RegLoss=0.000]

Epoch 63:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00975, v_num=16, MAE=1.770, RMSE=2.190, Loss=0.00904, RegLoss=0.000]

Epoch 64:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00975, v_num=16, MAE=1.770, RMSE=2.190, Loss=0.00904, RegLoss=0.000]

Epoch 64:  64%|██████▍   | 64/100 [00:00<00:00, 166213.90it/s, loss=0.00975, v_num=16, MAE=1.770, RMSE=2.190, Loss=0.00904, RegLoss=0.000]

Epoch 64:  64%|██████▍   | 64/100 [00:00<00:00, 2782.15it/s, loss=0.00873, v_num=16, MAE=1.760, RMSE=2.180, Loss=0.00897, RegLoss=0.000]

Epoch 64:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00873, v_num=16, MAE=1.760, RMSE=2.180, Loss=0.00897, RegLoss=0.000]

Epoch 65:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00873, v_num=16, MAE=1.760, RMSE=2.180, Loss=0.00897, RegLoss=0.000]

Epoch 65:  65%|██████▌   | 65/100 [00:00<00:00, 179125.99it/s, loss=0.00873, v_num=16, MAE=1.760, RMSE=2.180, Loss=0.00897, RegLoss=0.000]

Epoch 65:  65%|██████▌   | 65/100 [00:00<00:00, 2827.43it/s, loss=0.00885, v_num=16, MAE=1.790, RMSE=2.210, Loss=0.00913, RegLoss=0.000]

Epoch 65:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00885, v_num=16, MAE=1.790, RMSE=2.210, Loss=0.00913, RegLoss=0.000]

Epoch 66:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00885, v_num=16, MAE=1.790, RMSE=2.210, Loss=0.00913, RegLoss=0.000]

Epoch 66:  66%|██████▌   | 66/100 [00:00<00:00, 192372.53it/s, loss=0.00885, v_num=16, MAE=1.790, RMSE=2.210, Loss=0.00913, RegLoss=0.000]

Epoch 66:  66%|██████▌   | 66/100 [00:00<00:00, 2791.18it/s, loss=0.00919, v_num=16, MAE=1.770, RMSE=2.210, Loss=0.00931, RegLoss=0.000]

Epoch 66:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00919, v_num=16, MAE=1.770, RMSE=2.210, Loss=0.00931, RegLoss=0.000]

Epoch 67:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00919, v_num=16, MAE=1.770, RMSE=2.210, Loss=0.00931, RegLoss=0.000]

Epoch 67:  67%|██████▋   | 67/100 [00:00<00:00, 195287.26it/s, loss=0.00919, v_num=16, MAE=1.770, RMSE=2.210, Loss=0.00931, RegLoss=0.000]

Epoch 67:  67%|██████▋   | 67/100 [00:00<00:00, 2942.20it/s, loss=0.00905, v_num=16, MAE=1.750, RMSE=2.170, Loss=0.00895, RegLoss=0.000]

Epoch 67:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00905, v_num=16, MAE=1.750, RMSE=2.170, Loss=0.00895, RegLoss=0.000]

Epoch 68:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00905, v_num=16, MAE=1.750, RMSE=2.170, Loss=0.00895, RegLoss=0.000]

Epoch 68:  68%|██████▊   | 68/100 [00:00<00:00, 191675.18it/s, loss=0.00905, v_num=16, MAE=1.750, RMSE=2.170, Loss=0.00895, RegLoss=0.000]

Epoch 68:  68%|██████▊   | 68/100 [00:00<00:00, 2897.09it/s, loss=0.0098, v_num=16, MAE=1.900, RMSE=2.310, Loss=0.0101, RegLoss=0.000]

Epoch 68:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0098, v_num=16, MAE=1.900, RMSE=2.310, Loss=0.0101, RegLoss=0.000]

Epoch 69:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0098, v_num=16, MAE=1.900, RMSE=2.310, Loss=0.0101, RegLoss=0.000]

Epoch 69:  69%|██████▉   | 69/100 [00:00<00:00, 211554.81it/s, loss=0.0098, v_num=16, MAE=1.900, RMSE=2.310, Loss=0.0101, RegLoss=0.000]

Epoch 69:  69%|██████▉   | 69/100 [00:00<00:00, 3019.28it/s, loss=0.0107, v_num=16, MAE=1.820, RMSE=2.250, Loss=0.00977, RegLoss=0.000]

Epoch 69:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0107, v_num=16, MAE=1.820, RMSE=2.250, Loss=0.00977, RegLoss=0.000]

Epoch 70:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0107, v_num=16, MAE=1.820, RMSE=2.250, Loss=0.00977, RegLoss=0.000]

Epoch 70:  70%|███████   | 70/100 [00:00<00:00, 234131.80it/s, loss=0.0107, v_num=16, MAE=1.820, RMSE=2.250, Loss=0.00977, RegLoss=0.000]

Epoch 70:  70%|███████   | 70/100 [00:00<00:00, 3088.98it/s, loss=0.0096, v_num=16, MAE=1.760, RMSE=2.190, Loss=0.00901, RegLoss=0.000]

Epoch 70:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0096, v_num=16, MAE=1.760, RMSE=2.190, Loss=0.00901, RegLoss=0.000]

Epoch 71:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0096, v_num=16, MAE=1.760, RMSE=2.190, Loss=0.00901, RegLoss=0.000]

Epoch 71:  71%|███████   | 71/100 [00:00<00:00, 231387.40it/s, loss=0.0096, v_num=16, MAE=1.760, RMSE=2.190, Loss=0.00901, RegLoss=0.000]

Epoch 71:  71%|███████   | 71/100 [00:00<00:00, 3148.01it/s, loss=0.00867, v_num=16, MAE=1.740, RMSE=2.160, Loss=0.00886, RegLoss=0.000]

Epoch 71:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00867, v_num=16, MAE=1.740, RMSE=2.160, Loss=0.00886, RegLoss=0.000]

Epoch 72:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00867, v_num=16, MAE=1.740, RMSE=2.160, Loss=0.00886, RegLoss=0.000]

Epoch 72:  72%|███████▏  | 72/100 [00:00<00:00, 230702.74it/s, loss=0.00867, v_num=16, MAE=1.740, RMSE=2.160, Loss=0.00886, RegLoss=0.000]

Epoch 72:  72%|███████▏  | 72/100 [00:00<00:00, 3096.63it/s, loss=0.00843, v_num=16, MAE=1.720, RMSE=2.150, Loss=0.00876, RegLoss=0.000]

Epoch 72:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00843, v_num=16, MAE=1.720, RMSE=2.150, Loss=0.00876, RegLoss=0.000]

Epoch 73:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00843, v_num=16, MAE=1.720, RMSE=2.150, Loss=0.00876, RegLoss=0.000]

Epoch 73:  73%|███████▎  | 73/100 [00:00<00:00, 213071.81it/s, loss=0.00843, v_num=16, MAE=1.720, RMSE=2.150, Loss=0.00876, RegLoss=0.000]

Epoch 73:  73%|███████▎  | 73/100 [00:00<00:00, 3151.60it/s, loss=0.00871, v_num=16, MAE=1.780, RMSE=2.210, Loss=0.00912, RegLoss=0.000]

Epoch 73:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00871, v_num=16, MAE=1.780, RMSE=2.210, Loss=0.00912, RegLoss=0.000]

Epoch 74:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00871, v_num=16, MAE=1.780, RMSE=2.210, Loss=0.00912, RegLoss=0.000]

Epoch 74:  74%|███████▍  | 74/100 [00:00<00:00, 220126.59it/s, loss=0.00871, v_num=16, MAE=1.780, RMSE=2.210, Loss=0.00912, RegLoss=0.000]

Epoch 74:  74%|███████▍  | 74/100 [00:00<00:00, 3135.33it/s, loss=0.009, v_num=16, MAE=1.780, RMSE=2.210, Loss=0.00904, RegLoss=0.000]

Epoch 74:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.009, v_num=16, MAE=1.780, RMSE=2.210, Loss=0.00904, RegLoss=0.000]

Epoch 75:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.009, v_num=16, MAE=1.780, RMSE=2.210, Loss=0.00904, RegLoss=0.000]

Epoch 75:  75%|███████▌  | 75/100 [00:00<00:00, 177724.75it/s, loss=0.009, v_num=16, MAE=1.780, RMSE=2.210, Loss=0.00904, RegLoss=0.000]

Epoch 75:  75%|███████▌  | 75/100 [00:00<00:00, 3186.48it/s, loss=0.00913, v_num=16, MAE=1.780, RMSE=2.210, Loss=0.00924, RegLoss=0.000]

Epoch 75:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00913, v_num=16, MAE=1.780, RMSE=2.210, Loss=0.00924, RegLoss=0.000]

Epoch 76:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00913, v_num=16, MAE=1.780, RMSE=2.210, Loss=0.00924, RegLoss=0.000]

Epoch 76:  76%|███████▌  | 76/100 [00:00<00:00, 229659.30it/s, loss=0.00913, v_num=16, MAE=1.780, RMSE=2.210, Loss=0.00924, RegLoss=0.000]

Epoch 76:  76%|███████▌  | 76/100 [00:00<00:00, 3288.46it/s, loss=0.00885, v_num=16, MAE=1.720, RMSE=2.140, Loss=0.00875, RegLoss=0.000]

Epoch 76:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00885, v_num=16, MAE=1.720, RMSE=2.140, Loss=0.00875, RegLoss=0.000]

Epoch 77:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00885, v_num=16, MAE=1.720, RMSE=2.140, Loss=0.00875, RegLoss=0.000]

Epoch 77:  77%|███████▋  | 77/100 [00:00<00:00, 221814.15it/s, loss=0.00885, v_num=16, MAE=1.720, RMSE=2.140, Loss=0.00875, RegLoss=0.000]

Epoch 77:  77%|███████▋  | 77/100 [00:00<00:00, 3322.68it/s, loss=0.00986, v_num=16, MAE=1.850, RMSE=2.290, Loss=0.0103, RegLoss=0.000]

Epoch 77:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00986, v_num=16, MAE=1.850, RMSE=2.290, Loss=0.0103, RegLoss=0.000]

Epoch 78:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00986, v_num=16, MAE=1.850, RMSE=2.290, Loss=0.0103, RegLoss=0.000]

Epoch 78:  78%|███████▊  | 78/100 [00:00<00:00, 233682.65it/s, loss=0.00986, v_num=16, MAE=1.850, RMSE=2.290, Loss=0.0103, RegLoss=0.000]

Epoch 78:  78%|███████▊  | 78/100 [00:00<00:00, 3423.03it/s, loss=0.01, v_num=16, MAE=1.750, RMSE=2.170, Loss=0.00882, RegLoss=0.000]

Epoch 78:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.01, v_num=16, MAE=1.750, RMSE=2.170, Loss=0.00882, RegLoss=0.000]

Epoch 79:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.01, v_num=16, MAE=1.750, RMSE=2.170, Loss=0.00882, RegLoss=0.000]

Epoch 79:  79%|███████▉  | 79/100 [00:00<00:00, 216427.18it/s, loss=0.01, v_num=16, MAE=1.750, RMSE=2.170, Loss=0.00882, RegLoss=0.000]

Epoch 79:  79%|███████▉  | 79/100 [00:00<00:00, 3366.15it/s, loss=0.00856, v_num=16, MAE=1.740, RMSE=2.170, Loss=0.00883, RegLoss=0.000]

Epoch 79:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00856, v_num=16, MAE=1.740, RMSE=2.170, Loss=0.00883, RegLoss=0.000]

Epoch 80:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00856, v_num=16, MAE=1.740, RMSE=2.170, Loss=0.00883, RegLoss=0.000]

Epoch 80:  80%|████████  | 80/100 [00:00<00:00, 242445.32it/s, loss=0.00856, v_num=16, MAE=1.740, RMSE=2.170, Loss=0.00883, RegLoss=0.000]

Epoch 80:  80%|████████  | 80/100 [00:00<00:00, 3462.90it/s, loss=0.00932, v_num=16, MAE=1.840, RMSE=2.260, Loss=0.00961, RegLoss=0.000]

Epoch 80:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00932, v_num=16, MAE=1.840, RMSE=2.260, Loss=0.00961, RegLoss=0.000]

Epoch 81:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00932, v_num=16, MAE=1.840, RMSE=2.260, Loss=0.00961, RegLoss=0.000]

Epoch 81:  81%|████████  | 81/100 [00:00<00:00, 226190.83it/s, loss=0.00932, v_num=16, MAE=1.840, RMSE=2.260, Loss=0.00961, RegLoss=0.000]

Epoch 81:  81%|████████  | 81/100 [00:00<00:00, 3492.41it/s, loss=0.00956, v_num=16, MAE=1.770, RMSE=2.190, Loss=0.00904, RegLoss=0.000]

Epoch 81:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00956, v_num=16, MAE=1.770, RMSE=2.190, Loss=0.00904, RegLoss=0.000]

Epoch 82:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00956, v_num=16, MAE=1.770, RMSE=2.190, Loss=0.00904, RegLoss=0.000]

Epoch 82:  82%|████████▏ | 82/100 [00:00<00:00, 227018.43it/s, loss=0.00956, v_num=16, MAE=1.770, RMSE=2.190, Loss=0.00904, RegLoss=0.000]

Epoch 82:  82%|████████▏ | 82/100 [00:00<00:00, 3534.91it/s, loss=0.00875, v_num=16, MAE=1.740, RMSE=2.180, Loss=0.00881, RegLoss=0.000]

Epoch 82:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00875, v_num=16, MAE=1.740, RMSE=2.180, Loss=0.00881, RegLoss=0.000]

Epoch 83:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00875, v_num=16, MAE=1.740, RMSE=2.180, Loss=0.00881, RegLoss=0.000]

Epoch 83:  83%|████████▎ | 83/100 [00:00<00:00, 247953.87it/s, loss=0.00875, v_num=16, MAE=1.740, RMSE=2.180, Loss=0.00881, RegLoss=0.000]

Epoch 83:  83%|████████▎ | 83/100 [00:00<00:00, 3551.12it/s, loss=0.00849, v_num=16, MAE=1.730, RMSE=2.160, Loss=0.00877, RegLoss=0.000]

Epoch 83:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00849, v_num=16, MAE=1.730, RMSE=2.160, Loss=0.00877, RegLoss=0.000]

Epoch 84:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00849, v_num=16, MAE=1.730, RMSE=2.160, Loss=0.00877, RegLoss=0.000]

Epoch 84:  84%|████████▍ | 84/100 [00:00<00:00, 223412.51it/s, loss=0.00849, v_num=16, MAE=1.730, RMSE=2.160, Loss=0.00877, RegLoss=0.000]

Epoch 84:  84%|████████▍ | 84/100 [00:00<00:00, 3612.11it/s, loss=0.00877, v_num=16, MAE=1.770, RMSE=2.200, Loss=0.00907, RegLoss=0.000]

Epoch 84:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00877, v_num=16, MAE=1.770, RMSE=2.200, Loss=0.00907, RegLoss=0.000]

Epoch 85:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00877, v_num=16, MAE=1.770, RMSE=2.200, Loss=0.00907, RegLoss=0.000]

Epoch 85:  85%|████████▌ | 85/100 [00:00<00:00, 241541.90it/s, loss=0.00877, v_num=16, MAE=1.770, RMSE=2.200, Loss=0.00907, RegLoss=0.000]

Epoch 85:  85%|████████▌ | 85/100 [00:00<00:00, 3686.18it/s, loss=0.0101, v_num=16, MAE=1.840, RMSE=2.280, Loss=0.0102, RegLoss=0.000]

Epoch 85:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0101, v_num=16, MAE=1.840, RMSE=2.280, Loss=0.0102, RegLoss=0.000]

Epoch 86:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0101, v_num=16, MAE=1.840, RMSE=2.280, Loss=0.0102, RegLoss=0.000]

Epoch 86:  86%|████████▌ | 86/100 [00:00<00:00, 228731.86it/s, loss=0.0101, v_num=16, MAE=1.840, RMSE=2.280, Loss=0.0102, RegLoss=0.000]

Epoch 86:  86%|████████▌ | 86/100 [00:00<00:00, 3681.99it/s, loss=0.0103, v_num=16, MAE=1.800, RMSE=2.230, Loss=0.00919, RegLoss=0.000]

Epoch 86:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0103, v_num=16, MAE=1.800, RMSE=2.230, Loss=0.00919, RegLoss=0.000]

Epoch 87:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0103, v_num=16, MAE=1.800, RMSE=2.230, Loss=0.00919, RegLoss=0.000]

Epoch 87:  87%|████████▋ | 87/100 [00:00<00:00, 248572.51it/s, loss=0.0103, v_num=16, MAE=1.800, RMSE=2.230, Loss=0.00919, RegLoss=0.000]

Epoch 87:  87%|████████▋ | 87/100 [00:00<00:00, 3701.04it/s, loss=0.00924, v_num=16, MAE=1.790, RMSE=2.220, Loss=0.00915, RegLoss=0.000]

Epoch 87:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00924, v_num=16, MAE=1.790, RMSE=2.220, Loss=0.00915, RegLoss=0.000]

Epoch 88:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00924, v_num=16, MAE=1.790, RMSE=2.220, Loss=0.00915, RegLoss=0.000]

Epoch 88:  88%|████████▊ | 88/100 [00:00<00:00, 236601.76it/s, loss=0.00924, v_num=16, MAE=1.790, RMSE=2.220, Loss=0.00915, RegLoss=0.000]

Epoch 88:  88%|████████▊ | 88/100 [00:00<00:00, 3763.24it/s, loss=0.00892, v_num=16, MAE=1.750, RMSE=2.170, Loss=0.00882, RegLoss=0.000]

Epoch 88:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00892, v_num=16, MAE=1.750, RMSE=2.170, Loss=0.00882, RegLoss=0.000]

Epoch 89:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00892, v_num=16, MAE=1.750, RMSE=2.170, Loss=0.00882, RegLoss=0.000]

Epoch 89:  89%|████████▉ | 89/100 [00:00<00:00, 241926.80it/s, loss=0.00892, v_num=16, MAE=1.750, RMSE=2.170, Loss=0.00882, RegLoss=0.000]

Epoch 89:  89%|████████▉ | 89/100 [00:00<00:00, 3862.52it/s, loss=0.0085, v_num=16, MAE=1.710, RMSE=2.140, Loss=0.00872, RegLoss=0.000]

Epoch 89:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0085, v_num=16, MAE=1.710, RMSE=2.140, Loss=0.00872, RegLoss=0.000]

Epoch 90:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0085, v_num=16, MAE=1.710, RMSE=2.140, Loss=0.00872, RegLoss=0.000]

Epoch 90:  90%|█████████ | 90/100 [00:00<00:00, 264717.64it/s, loss=0.0085, v_num=16, MAE=1.710, RMSE=2.140, Loss=0.00872, RegLoss=0.000]

Epoch 90:  90%|█████████ | 90/100 [00:00<00:00, 3848.61it/s, loss=0.0085, v_num=16, MAE=1.740, RMSE=2.160, Loss=0.0088, RegLoss=0.000]

Epoch 90:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0085, v_num=16, MAE=1.740, RMSE=2.160, Loss=0.0088, RegLoss=0.000]

Epoch 91:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0085, v_num=16, MAE=1.740, RMSE=2.160, Loss=0.0088, RegLoss=0.000]

Epoch 91:  91%|█████████ | 91/100 [00:00<00:00, 265980.25it/s, loss=0.0085, v_num=16, MAE=1.740, RMSE=2.160, Loss=0.0088, RegLoss=0.000]

Epoch 91:  91%|█████████ | 91/100 [00:00<00:00, 3910.27it/s, loss=0.00929, v_num=16, MAE=1.810, RMSE=2.240, Loss=0.00954, RegLoss=0.000]

Epoch 91:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00929, v_num=16, MAE=1.810, RMSE=2.240, Loss=0.00954, RegLoss=0.000]

Epoch 92:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00929, v_num=16, MAE=1.810, RMSE=2.240, Loss=0.00954, RegLoss=0.000]

Epoch 92:  92%|█████████▏| 92/100 [00:00<00:00, 257079.26it/s, loss=0.00929, v_num=16, MAE=1.810, RMSE=2.240, Loss=0.00954, RegLoss=0.000]

Epoch 92:  92%|█████████▏| 92/100 [00:00<00:00, 4001.95it/s, loss=0.00967, v_num=16, MAE=1.760, RMSE=2.200, Loss=0.00917, RegLoss=0.000]

Epoch 92:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00967, v_num=16, MAE=1.760, RMSE=2.200, Loss=0.00917, RegLoss=0.000]

Epoch 93:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00967, v_num=16, MAE=1.760, RMSE=2.200, Loss=0.00917, RegLoss=0.000]

Epoch 93:  93%|█████████▎| 93/100 [00:00<00:00, 281030.46it/s, loss=0.00967, v_num=16, MAE=1.760, RMSE=2.200, Loss=0.00917, RegLoss=0.000]

Epoch 93:  93%|█████████▎| 93/100 [00:00<00:00, 3957.97it/s, loss=0.00883, v_num=16, MAE=1.720, RMSE=2.140, Loss=0.00868, RegLoss=0.000]

Epoch 93:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00883, v_num=16, MAE=1.720, RMSE=2.140, Loss=0.00868, RegLoss=0.000]

Epoch 94:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00883, v_num=16, MAE=1.720, RMSE=2.140, Loss=0.00868, RegLoss=0.000]

Epoch 94:  94%|█████████▍| 94/100 [00:00<00:00, 279818.72it/s, loss=0.00883, v_num=16, MAE=1.720, RMSE=2.140, Loss=0.00868, RegLoss=0.000]

Epoch 94:  94%|█████████▍| 94/100 [00:00<00:00, 4038.35it/s, loss=0.009, v_num=16, MAE=1.780, RMSE=2.210, Loss=0.00936, RegLoss=0.000]

Epoch 94:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.009, v_num=16, MAE=1.780, RMSE=2.210, Loss=0.00936, RegLoss=0.000]

Epoch 95:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.009, v_num=16, MAE=1.780, RMSE=2.210, Loss=0.00936, RegLoss=0.000]

Epoch 95:  95%|█████████▌| 95/100 [00:00<00:00, 226783.65it/s, loss=0.009, v_num=16, MAE=1.780, RMSE=2.210, Loss=0.00936, RegLoss=0.000]

Epoch 95:  95%|█████████▌| 95/100 [00:00<00:00, 4018.99it/s, loss=0.00995, v_num=16, MAE=1.780, RMSE=2.240, Loss=0.00964, RegLoss=0.000]

Epoch 95:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00995, v_num=16, MAE=1.780, RMSE=2.240, Loss=0.00964, RegLoss=0.000]

Epoch 96:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00995, v_num=16, MAE=1.780, RMSE=2.240, Loss=0.00964, RegLoss=0.000]

Epoch 96:  96%|█████████▌| 96/100 [00:00<00:00, 254521.61it/s, loss=0.00995, v_num=16, MAE=1.780, RMSE=2.240, Loss=0.00964, RegLoss=0.000]

Epoch 96:  96%|█████████▌| 96/100 [00:00<00:00, 4158.14it/s, loss=0.00968, v_num=16, MAE=1.780, RMSE=2.210, Loss=0.00907, RegLoss=0.000]

Epoch 96:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00968, v_num=16, MAE=1.780, RMSE=2.210, Loss=0.00907, RegLoss=0.000]

Epoch 97:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00968, v_num=16, MAE=1.780, RMSE=2.210, Loss=0.00907, RegLoss=0.000]

Epoch 97:  97%|█████████▋| 97/100 [00:00<00:00, 253962.23it/s, loss=0.00968, v_num=16, MAE=1.780, RMSE=2.210, Loss=0.00907, RegLoss=0.000]

Epoch 97:  97%|█████████▋| 97/100 [00:00<00:00, 4159.19it/s, loss=0.00895, v_num=16, MAE=1.760, RMSE=2.170, Loss=0.00889, RegLoss=0.000]

Epoch 97:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00895, v_num=16, MAE=1.760, RMSE=2.170, Loss=0.00889, RegLoss=0.000]

Epoch 98:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00895, v_num=16, MAE=1.760, RMSE=2.170, Loss=0.00889, RegLoss=0.000]

Epoch 98:  98%|█████████▊| 98/100 [00:00<00:00, 315942.96it/s, loss=0.00895, v_num=16, MAE=1.760, RMSE=2.170, Loss=0.00889, RegLoss=0.000]

Epoch 98:  98%|█████████▊| 98/100 [00:00<00:00, 4238.02it/s, loss=0.00851, v_num=16, MAE=1.700, RMSE=2.120, Loss=0.00861, RegLoss=0.000]

Epoch 98:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00851, v_num=16, MAE=1.700, RMSE=2.120, Loss=0.00861, RegLoss=0.000]

Epoch 99:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.00851, v_num=16, MAE=1.700, RMSE=2.120, Loss=0.00861, RegLoss=0.000]

Epoch 99:  99%|█████████▉| 99/100 [00:00<00:00, 292008.51it/s, loss=0.00851, v_num=16, MAE=1.700, RMSE=2.120, Loss=0.00861, RegLoss=0.000]

Epoch 99:  99%|█████████▉| 99/100 [00:00<00:00, 4237.32it/s, loss=0.0085, v_num=16, MAE=1.750, RMSE=2.180, Loss=0.00889, RegLoss=0.000]

Epoch 99:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0085, v_num=16, MAE=1.750, RMSE=2.180, Loss=0.00889, RegLoss=0.000]

Epoch 100:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0085, v_num=16, MAE=1.750, RMSE=2.180, Loss=0.00889, RegLoss=0.000]

Epoch 100: 100%|██████████| 100/100 [00:00<00:00, 292285.99it/s, loss=0.0085, v_num=16, MAE=1.750, RMSE=2.180, Loss=0.00889, RegLoss=0.000]

Epoch 100: 100%|██████████| 100/100 [00:00<00:00, 4381.34it/s, loss=0.00856, v_num=16, MAE=1.720, RMSE=2.150, Loss=0.00867, RegLoss=0.000]

Epoch 100: 100%|██████████| 100/100 [00:00<00:00, 4205.57it/s, loss=0.00856, v_num=16, MAE=1.720, RMSE=2.150, Loss=0.00867, RegLoss=0.000]

Predicting: 19it [00:00, ?it/s]

Predicting:   0%|          | 0/2 [00:00<-00:00, -1695569.70it/s]

Predicting DataLoader 0:   0%|          | 0/2 [00:00<?, ?it/s]

Predicting DataLoader 0: 100%|██████████| 2/2 [00:00<00:00, 893.93it/s]

Predicting DataLoader 0: 100%|██████████| 2/2 [00:00<00:00, 758.40it/s]


FitResult fields:
  fitted_values  shape = (1169,)
  residuals      shape = (1169,)
  NaN fitted values   = 14  (AR warmup = n_lags = 14)
  residuals  mean=-0.0095  std=2.2023  (NaNs excluded)

4c. Inside NeuralProphet -- AR terms and the warmup period¶

NeuralProphet uses autoregressive (AR) lags: to predict observation t, it uses the observed values at observations t-1 through t-n_lags. n_lags counts observations of the series, not calendar days (the two coincide here because the dummy data is daily). This means the first n_lags rows cannot produce a prediction because there is no history yet -- they form the AR warmup period and produce NaN fitted values and residuals.

This has a practical implication for Moving Block Bootstrap: the bootstrap resamples only the non-NaN residuals. The available residuals are total_train_rows - n_lags.

In [4]:

Copied!





final_metrics = fit_result.metadata.get("final_metrics")
if final_metrics is not None:
    print("Training loss (last 5 epochs):")
    print(final_metrics.tail().to_string(index=False))

n_lags       = model_params.get("n_lags", 14)
n_train      = len(splits.train_df)
n_valid_resid = int(np.isfinite(fit_result.residuals).sum())
print(f"\n  Total training rows      : {n_train}")
print(f"  AR warmup rows (n_lags)  : {n_lags}")
print(f"  Valid (non-NaN) residuals: {n_valid_resid}")
print(f"  -> MBB will resample from {n_valid_resid} residuals")
final_metrics = fit_result.metadata.get("final_metrics")
if final_metrics is not None:
    print("Training loss (last 5 epochs):")
    print(final_metrics.tail().to_string(index=False))

n_lags       = model_params.get("n_lags", 14)
n_train      = len(splits.train_df)
n_valid_resid = int(np.isfinite(fit_result.residuals).sum())
print(f"\n  Total training rows      : {n_train}")
print(f"  AR warmup rows (n_lags)  : {n_lags}")
print(f"  Valid (non-NaN) residuals: {n_valid_resid}")
print(f"  -> MBB will resample from {n_valid_resid} residuals")

Training loss (last 5 epochs):
     MAE     RMSE     Loss  RegLoss  epoch
1.775751 2.211107 0.009065      0.0     95
1.759129 2.170924 0.008893      0.0     96
1.700649 2.118625 0.008615      0.0     97
1.754008 2.181051 0.008892      0.0     98
1.716331 2.146283 0.008666      0.0     99

  Total training rows      : 1169
  AR warmup rows (n_lags)  : 14
  Valid (non-NaN) residuals: 1155
  -> MBB will resample from 1155 residuals

In [5]:

Copied!





dates    = splits.train_df["ds"].values
residuals = fit_result.residuals
warmup_end = splits.train_df["ds"].iloc[n_lags - 1]

fig, ax = plt.subplots(figsize=(13, 3.5))
ax.plot(dates, residuals, linewidth=0.6, color="#4C72B0", alpha=0.7, label="Residuals")
ax.axhline(0, color="black", linewidth=0.8, linestyle="--")

# Shade the warmup region
ax.axvspan(dates[0], warmup_end, color="#CCCCCC", alpha=0.5, label=f"AR warmup (n_lags={n_lags}, NaN)")
ax.axvline(pd.Timestamp(warmup_end), color="#888888", linestyle=":", linewidth=1.2)
ax.annotate(
    f"Warmup ends\n(row {n_lags})",
    xy=(pd.Timestamp(warmup_end), 0),
    xytext=(pd.Timestamp(warmup_end) + pd.Timedelta(days=60), residuals[n_lags + 10]),
    arrowprops=dict(arrowstyle="->", color="#555555"),
    fontsize=8,
    color="#555555",
)

ax.set_title("NeuralProphet residuals (y - fitted) -- shaded region is AR warmup (NaN)", fontsize=10)
ax.set_ylabel("Residual")
ax.legend(fontsize=8)
ax.xaxis.set_major_formatter(mdates.DateFormatter("%Y"))
plt.tight_layout()
plt.savefig(OUT_DIR / "neuralprophet_residuals.png", dpi=150)
display(fig)
dates    = splits.train_df["ds"].values
residuals = fit_result.residuals
warmup_end = splits.train_df["ds"].iloc[n_lags - 1]

fig, ax = plt.subplots(figsize=(13, 3.5))
ax.plot(dates, residuals, linewidth=0.6, color="#4C72B0", alpha=0.7, label="Residuals")
ax.axhline(0, color="black", linewidth=0.8, linestyle="--")

# Shade the warmup region
ax.axvspan(dates[0], warmup_end, color="#CCCCCC", alpha=0.5, label=f"AR warmup (n_lags={n_lags}, NaN)")
ax.axvline(pd.Timestamp(warmup_end), color="#888888", linestyle=":", linewidth=1.2)
ax.annotate(
    f"Warmup ends\n(row {n_lags})",
    xy=(pd.Timestamp(warmup_end), 0),
    xytext=(pd.Timestamp(warmup_end) + pd.Timedelta(days=60), residuals[n_lags + 10]),
    arrowprops=dict(arrowstyle="->", color="#555555"),
    fontsize=8,
    color="#555555",
)

ax.set_title("NeuralProphet residuals (y - fitted) -- shaded region is AR warmup (NaN)", fontsize=10)
ax.set_ylabel("Residual")
ax.legend(fontsize=8)
ax.xaxis.set_major_formatter(mdates.DateFormatter("%Y"))
plt.tight_layout()
plt.savefig(OUT_DIR / "neuralprophet_residuals.png", dpi=150)
display(fig)

No description has been provided for this image

4d. Full pipeline run¶

MBB bootstrap runs with n_sim=100 for speed; production use should set this to 1000+. Expect this cell to take longer than ARIMA or Prophet models due to neural network training.

In [6]:

Copied!





from its2s import run_single_its

result = run_single_its(
    df=df,
    intervention_date=INTERVENTION,
    model_name="neuralprophet",
    config_overrides={
        "bootstrap": {"n_sim": 100},
        "periods":   {"split_method": "days", "test_days": TEST_DAYS, "holdout_days": HOLDOUT_DAYS},
    },
    output_dir=OUT_DIR,
    seed=42,
)

print("PipelineResult fields:")
print(f"  model_name       : {result.model_name}")
print(f"  fit_result       : FitResult with {len(result.fit_result.fitted_values)} fitted values")
print(f"  bootstrap_result : BootstrapCIResult  pred_matrix shape = {result.bootstrap_result.pred_matrix.shape}")
print(f"  n_successful sims: {result.bootstrap_result.n_successful}")
print(f"  metrics_train    : {result.metrics_train}")
print(f"  metrics_test     : {result.metrics_test}")
from its2s import run_single_its

result = run_single_its(
    df=df,
    intervention_date=INTERVENTION,
    model_name="neuralprophet",
    config_overrides={
        "bootstrap": {"n_sim": 100},
        "periods":   {"split_method": "days", "test_days": TEST_DAYS, "holdout_days": HOLDOUT_DAYS},
    },
    output_dir=OUT_DIR,
    seed=42,
)

print("PipelineResult fields:")
print(f"  model_name       : {result.model_name}")
print(f"  fit_result       : FitResult with {len(result.fit_result.fitted_values)} fitted values")
print(f"  bootstrap_result : BootstrapCIResult  pred_matrix shape = {result.bootstrap_result.pred_matrix.shape}")
print(f"  n_successful sims: {result.bootstrap_result.n_successful}")
print(f"  metrics_train    : {result.metrics_train}")
print(f"  metrics_test     : {result.metrics_test}")

Training: 0it [00:00, ?it/s]

Training:   0%|          | 0/100 [00:00<?, ?it/s]

Epoch 1:   0%|          | 0/100 [00:00<?, ?it/s]

Epoch 1:   1%|          | 1/100 [00:00<00:00, 2809.31it/s]

Epoch 1:   1%|          | 1/100 [00:00<00:02, 38.96it/s, loss=1.58, v_num=17, MAE=55.30, RMSE=67.10, Loss=1.580, RegLoss=0.000]

Epoch 1:   0%|          | 0/100 [00:00<?, ?it/s, loss=1.58, v_num=17, MAE=55.30, RMSE=67.10, Loss=1.580, RegLoss=0.000]

Epoch 2:   0%|          | 0/100 [00:00<?, ?it/s, loss=1.58, v_num=17, MAE=55.30, RMSE=67.10, Loss=1.580, RegLoss=0.000]

Epoch 2:   2%|▏         | 2/100 [00:00<00:00, 4851.71it/s, loss=1.58, v_num=17, MAE=55.30, RMSE=67.10, Loss=1.580, RegLoss=0.000]

Epoch 2:   2%|▏         | 2/100 [00:00<00:01, 83.01it/s, loss=1.22, v_num=17, MAE=43.80, RMSE=54.90, Loss=1.200, RegLoss=0.000]

Epoch 2:   0%|          | 0/100 [00:00<?, ?it/s, loss=1.22, v_num=17, MAE=43.80, RMSE=54.90, Loss=1.200, RegLoss=0.000]

Epoch 3:   0%|          | 0/100 [00:00<?, ?it/s, loss=1.22, v_num=17, MAE=43.80, RMSE=54.90, Loss=1.200, RegLoss=0.000]

Epoch 3:   3%|▎         | 3/100 [00:00<00:00, 8905.10it/s, loss=1.22, v_num=17, MAE=43.80, RMSE=54.90, Loss=1.200, RegLoss=0.000]

Epoch 3:   3%|▎         | 3/100 [00:00<00:00, 127.94it/s, loss=0.873, v_num=17, MAE=32.30, RMSE=40.60, Loss=0.860, RegLoss=0.000]

Epoch 3:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.873, v_num=17, MAE=32.30, RMSE=40.60, Loss=0.860, RegLoss=0.000]

Epoch 4:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.873, v_num=17, MAE=32.30, RMSE=40.60, Loss=0.860, RegLoss=0.000]

Epoch 4:   4%|▍         | 4/100 [00:00<00:00, 10845.00it/s, loss=0.873, v_num=17, MAE=32.30, RMSE=40.60, Loss=0.860, RegLoss=0.000]

Epoch 4:   4%|▍         | 4/100 [00:00<00:00, 165.76it/s, loss=0.651, v_num=17, MAE=26.60, RMSE=33.30, Loss=0.670, RegLoss=0.000]

Epoch 4:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.651, v_num=17, MAE=26.60, RMSE=33.30, Loss=0.670, RegLoss=0.000]

Epoch 5:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.651, v_num=17, MAE=26.60, RMSE=33.30, Loss=0.670, RegLoss=0.000]

Epoch 5:   5%|▌         | 5/100 [00:00<00:00, 15098.29it/s, loss=0.651, v_num=17, MAE=26.60, RMSE=33.30, Loss=0.670, RegLoss=0.000]

Epoch 5:   5%|▌         | 5/100 [00:00<00:00, 206.95it/s, loss=0.56, v_num=17, MAE=21.90, RMSE=27.40, Loss=0.540, RegLoss=0.000]

Epoch 5:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.56, v_num=17, MAE=21.90, RMSE=27.40, Loss=0.540, RegLoss=0.000]

Epoch 6:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.56, v_num=17, MAE=21.90, RMSE=27.40, Loss=0.540, RegLoss=0.000]

Epoch 6:   6%|▌         | 6/100 [00:00<00:00, 17848.10it/s, loss=0.56, v_num=17, MAE=21.90, RMSE=27.40, Loss=0.540, RegLoss=0.000]

Epoch 6:   6%|▌         | 6/100 [00:00<00:00, 258.59it/s, loss=0.48, v_num=17, MAE=20.10, RMSE=25.20, Loss=0.488, RegLoss=0.000]

Epoch 6:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.48, v_num=17, MAE=20.10, RMSE=25.20, Loss=0.488, RegLoss=0.000]

Epoch 7:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.48, v_num=17, MAE=20.10, RMSE=25.20, Loss=0.488, RegLoss=0.000]

Epoch 7:   7%|▋         | 7/100 [00:00<00:00, 19612.64it/s, loss=0.48, v_num=17, MAE=20.10, RMSE=25.20, Loss=0.488, RegLoss=0.000]

Epoch 7:   7%|▋         | 7/100 [00:00<00:00, 305.40it/s, loss=0.423, v_num=17, MAE=17.90, RMSE=22.70, Loss=0.429, RegLoss=0.000]

Epoch 7:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.423, v_num=17, MAE=17.90, RMSE=22.70, Loss=0.429, RegLoss=0.000]

Epoch 8:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.423, v_num=17, MAE=17.90, RMSE=22.70, Loss=0.429, RegLoss=0.000]

Epoch 8:   8%|▊         | 8/100 [00:00<00:00, 23045.63it/s, loss=0.423, v_num=17, MAE=17.90, RMSE=22.70, Loss=0.429, RegLoss=0.000]

Epoch 8:   8%|▊         | 8/100 [00:00<00:00, 333.74it/s, loss=0.36, v_num=17, MAE=15.70, RMSE=20.10, Loss=0.360, RegLoss=0.000]

Epoch 8:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.36, v_num=17, MAE=15.70, RMSE=20.10, Loss=0.360, RegLoss=0.000]

Epoch 9:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.36, v_num=17, MAE=15.70, RMSE=20.10, Loss=0.360, RegLoss=0.000]

Epoch 9:   9%|▉         | 9/100 [00:00<00:00, 17967.03it/s, loss=0.36, v_num=17, MAE=15.70, RMSE=20.10, Loss=0.360, RegLoss=0.000]

Epoch 9:   9%|▉         | 9/100 [00:00<00:00, 388.50it/s, loss=0.31, v_num=17, MAE=14.30, RMSE=18.40, Loss=0.322, RegLoss=0.000]

Epoch 9:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.31, v_num=17, MAE=14.30, RMSE=18.40, Loss=0.322, RegLoss=0.000]

Epoch 10:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.31, v_num=17, MAE=14.30, RMSE=18.40, Loss=0.322, RegLoss=0.000]

Epoch 10:  10%|█         | 10/100 [00:00<00:00, 21890.94it/s, loss=0.31, v_num=17, MAE=14.30, RMSE=18.40, Loss=0.322, RegLoss=0.000]

Epoch 10:  10%|█         | 10/100 [00:00<00:00, 438.16it/s, loss=0.253, v_num=17, MAE=11.40, RMSE=14.80, Loss=0.239, RegLoss=0.000]

Epoch 10:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.253, v_num=17, MAE=11.40, RMSE=14.80, Loss=0.239, RegLoss=0.000]

Epoch 11:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.253, v_num=17, MAE=11.40, RMSE=14.80, Loss=0.239, RegLoss=0.000]

Epoch 11:  11%|█         | 11/100 [00:00<00:00, 33750.80it/s, loss=0.253, v_num=17, MAE=11.40, RMSE=14.80, Loss=0.239, RegLoss=0.000]

Epoch 11:  11%|█         | 11/100 [00:00<00:00, 474.00it/s, loss=0.179, v_num=17, MAE=9.320, RMSE=12.10, Loss=0.184, RegLoss=0.000]

Epoch 11:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.179, v_num=17, MAE=9.320, RMSE=12.10, Loss=0.184, RegLoss=0.000]

Epoch 12:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.179, v_num=17, MAE=9.320, RMSE=12.10, Loss=0.184, RegLoss=0.000]

Epoch 12:  12%|█▏        | 12/100 [00:00<00:00, 34100.03it/s, loss=0.179, v_num=17, MAE=9.320, RMSE=12.10, Loss=0.184, RegLoss=0.000]

Epoch 12:  12%|█▏        | 12/100 [00:00<00:00, 509.73it/s, loss=0.135, v_num=17, MAE=7.600, RMSE=10.10, Loss=0.138, RegLoss=0.000]

Epoch 12:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.135, v_num=17, MAE=7.600, RMSE=10.10, Loss=0.138, RegLoss=0.000]

Epoch 13:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.135, v_num=17, MAE=7.600, RMSE=10.10, Loss=0.138, RegLoss=0.000]

Epoch 13:  13%|█▎        | 13/100 [00:00<00:00, 34752.04it/s, loss=0.135, v_num=17, MAE=7.600, RMSE=10.10, Loss=0.138, RegLoss=0.000]

Epoch 13:  13%|█▎        | 13/100 [00:00<00:00, 559.43it/s, loss=0.0965, v_num=17, MAE=5.970, RMSE=7.760, Loss=0.0953, RegLoss=0.000]

Epoch 13:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0965, v_num=17, MAE=5.970, RMSE=7.760, Loss=0.0953, RegLoss=0.000]

Epoch 14:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0965, v_num=17, MAE=5.970, RMSE=7.760, Loss=0.0953, RegLoss=0.000]

Epoch 14:  14%|█▍        | 14/100 [00:00<00:00, 39542.26it/s, loss=0.0965, v_num=17, MAE=5.970, RMSE=7.760, Loss=0.0953, RegLoss=0.000]

Epoch 14:  14%|█▍        | 14/100 [00:00<00:00, 590.00it/s, loss=0.0597, v_num=17, MAE=4.660, RMSE=6.030, Loss=0.062, RegLoss=0.000]

Epoch 14:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0597, v_num=17, MAE=4.660, RMSE=6.030, Loss=0.062, RegLoss=0.000]

Epoch 15:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0597, v_num=17, MAE=4.660, RMSE=6.030, Loss=0.062, RegLoss=0.000]

Epoch 15:  15%|█▌        | 15/100 [00:00<00:00, 42857.33it/s, loss=0.0597, v_num=17, MAE=4.660, RMSE=6.030, Loss=0.062, RegLoss=0.000]

Epoch 15:  15%|█▌        | 15/100 [00:00<00:00, 638.92it/s, loss=0.0411, v_num=17, MAE=3.730, RMSE=4.720, Loss=0.0411, RegLoss=0.000]

Epoch 15:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0411, v_num=17, MAE=3.730, RMSE=4.720, Loss=0.0411, RegLoss=0.000]

Epoch 16:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0411, v_num=17, MAE=3.730, RMSE=4.720, Loss=0.0411, RegLoss=0.000]

Epoch 16:  16%|█▌        | 16/100 [00:00<00:00, 47764.32it/s, loss=0.0411, v_num=17, MAE=3.730, RMSE=4.720, Loss=0.0411, RegLoss=0.000]

Epoch 16:  16%|█▌        | 16/100 [00:00<00:00, 697.10it/s, loss=0.033, v_num=17, MAE=3.450, RMSE=4.250, Loss=0.0334, RegLoss=0.000]

Epoch 16:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.033, v_num=17, MAE=3.450, RMSE=4.250, Loss=0.0334, RegLoss=0.000]

Epoch 17:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.033, v_num=17, MAE=3.450, RMSE=4.250, Loss=0.0334, RegLoss=0.000]

Epoch 17:  17%|█▋        | 17/100 [00:00<00:00, 50605.51it/s, loss=0.033, v_num=17, MAE=3.450, RMSE=4.250, Loss=0.0334, RegLoss=0.000]

Epoch 17:  17%|█▋        | 17/100 [00:00<00:00, 735.17it/s, loss=0.0276, v_num=17, MAE=3.010, RMSE=3.740, Loss=0.0261, RegLoss=0.000]

Epoch 17:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0276, v_num=17, MAE=3.010, RMSE=3.740, Loss=0.0261, RegLoss=0.000]

Epoch 18:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0276, v_num=17, MAE=3.010, RMSE=3.740, Loss=0.0261, RegLoss=0.000]

Epoch 18:  18%|█▊        | 18/100 [00:00<00:00, 51852.66it/s, loss=0.0276, v_num=17, MAE=3.010, RMSE=3.740, Loss=0.0261, RegLoss=0.000]

Epoch 18:  18%|█▊        | 18/100 [00:00<00:00, 766.41it/s, loss=0.0252, v_num=17, MAE=3.030, RMSE=3.740, Loss=0.0258, RegLoss=0.000]

Epoch 18:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0252, v_num=17, MAE=3.030, RMSE=3.740, Loss=0.0258, RegLoss=0.000]

Epoch 19:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0252, v_num=17, MAE=3.030, RMSE=3.740, Loss=0.0258, RegLoss=0.000]

Epoch 19:  19%|█▉        | 19/100 [00:00<00:00, 57580.76it/s, loss=0.0252, v_num=17, MAE=3.030, RMSE=3.740, Loss=0.0258, RegLoss=0.000]

Epoch 19:  19%|█▉        | 19/100 [00:00<00:00, 807.20it/s, loss=0.0249, v_num=17, MAE=2.920, RMSE=3.600, Loss=0.0244, RegLoss=0.000]

Epoch 19:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0249, v_num=17, MAE=2.920, RMSE=3.600, Loss=0.0244, RegLoss=0.000]

Epoch 20:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0249, v_num=17, MAE=2.920, RMSE=3.600, Loss=0.0244, RegLoss=0.000]

Epoch 20:  20%|██        | 20/100 [00:00<00:00, 56641.51it/s, loss=0.0249, v_num=17, MAE=2.920, RMSE=3.600, Loss=0.0244, RegLoss=0.000]

Epoch 20:  20%|██        | 20/100 [00:00<00:00, 847.17it/s, loss=0.0237, v_num=17, MAE=2.930, RMSE=3.660, Loss=0.0246, RegLoss=0.000]

Epoch 20:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0237, v_num=17, MAE=2.930, RMSE=3.660, Loss=0.0246, RegLoss=0.000]

Epoch 21:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0237, v_num=17, MAE=2.930, RMSE=3.660, Loss=0.0246, RegLoss=0.000]

Epoch 21:  21%|██        | 21/100 [00:00<00:00, 54270.11it/s, loss=0.0237, v_num=17, MAE=2.930, RMSE=3.660, Loss=0.0246, RegLoss=0.000]

Epoch 21:  21%|██        | 21/100 [00:00<00:00, 921.09it/s, loss=0.0231, v_num=17, MAE=2.820, RMSE=3.490, Loss=0.0228, RegLoss=0.000]

Epoch 21:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0231, v_num=17, MAE=2.820, RMSE=3.490, Loss=0.0228, RegLoss=0.000]

Epoch 22:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0231, v_num=17, MAE=2.820, RMSE=3.490, Loss=0.0228, RegLoss=0.000]

Epoch 22:  22%|██▏       | 22/100 [00:00<00:00, 59802.13it/s, loss=0.0231, v_num=17, MAE=2.820, RMSE=3.490, Loss=0.0228, RegLoss=0.000]

Epoch 22:  22%|██▏       | 22/100 [00:00<00:00, 931.07it/s, loss=0.0216, v_num=17, MAE=2.800, RMSE=3.460, Loss=0.0224, RegLoss=0.000]

Epoch 22:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0216, v_num=17, MAE=2.800, RMSE=3.460, Loss=0.0224, RegLoss=0.000]

Epoch 23:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0216, v_num=17, MAE=2.800, RMSE=3.460, Loss=0.0224, RegLoss=0.000]

Epoch 23:  23%|██▎       | 23/100 [00:00<00:00, 53004.94it/s, loss=0.0216, v_num=17, MAE=2.800, RMSE=3.460, Loss=0.0224, RegLoss=0.000]

Epoch 23:  23%|██▎       | 23/100 [00:00<00:00, 987.58it/s, loss=0.023, v_num=17, MAE=2.910, RMSE=3.570, Loss=0.0237, RegLoss=0.000]

Epoch 23:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.023, v_num=17, MAE=2.910, RMSE=3.570, Loss=0.0237, RegLoss=0.000]

Epoch 24:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.023, v_num=17, MAE=2.910, RMSE=3.570, Loss=0.0237, RegLoss=0.000]

Epoch 24:  24%|██▍       | 24/100 [00:00<00:00, 68385.39it/s, loss=0.023, v_num=17, MAE=2.910, RMSE=3.570, Loss=0.0237, RegLoss=0.000]

Epoch 24:  24%|██▍       | 24/100 [00:00<00:00, 1055.41it/s, loss=0.0264, v_num=17, MAE=3.040, RMSE=3.700, Loss=0.0257, RegLoss=0.000]

Epoch 24:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0264, v_num=17, MAE=3.040, RMSE=3.700, Loss=0.0257, RegLoss=0.000]

Epoch 25:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0264, v_num=17, MAE=3.040, RMSE=3.700, Loss=0.0257, RegLoss=0.000]

Epoch 25:  25%|██▌       | 25/100 [00:00<00:00, 68849.38it/s, loss=0.0264, v_num=17, MAE=3.040, RMSE=3.700, Loss=0.0257, RegLoss=0.000]

Epoch 25:  25%|██▌       | 25/100 [00:00<00:00, 1096.63it/s, loss=0.0276, v_num=17, MAE=3.080, RMSE=3.790, Loss=0.0265, RegLoss=0.000]

Epoch 25:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0276, v_num=17, MAE=3.080, RMSE=3.790, Loss=0.0265, RegLoss=0.000]

Epoch 26:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0276, v_num=17, MAE=3.080, RMSE=3.790, Loss=0.0265, RegLoss=0.000]

Epoch 26:  26%|██▌       | 26/100 [00:00<00:00, 85330.13it/s, loss=0.0276, v_num=17, MAE=3.080, RMSE=3.790, Loss=0.0265, RegLoss=0.000]

Epoch 26:  26%|██▌       | 26/100 [00:00<00:00, 1108.56it/s, loss=0.0251, v_num=17, MAE=2.900, RMSE=3.550, Loss=0.0237, RegLoss=0.000]

Epoch 26:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0251, v_num=17, MAE=2.900, RMSE=3.550, Loss=0.0237, RegLoss=0.000]

Epoch 27:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0251, v_num=17, MAE=2.900, RMSE=3.550, Loss=0.0237, RegLoss=0.000]

Epoch 27:  27%|██▋       | 27/100 [00:00<00:00, 76260.07it/s, loss=0.0251, v_num=17, MAE=2.900, RMSE=3.550, Loss=0.0237, RegLoss=0.000]

Epoch 27:  27%|██▋       | 27/100 [00:00<00:00, 1179.14it/s, loss=0.0241, v_num=17, MAE=2.840, RMSE=3.560, Loss=0.0235, RegLoss=0.000]

Epoch 27:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0241, v_num=17, MAE=2.840, RMSE=3.560, Loss=0.0235, RegLoss=0.000]

Epoch 28:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0241, v_num=17, MAE=2.840, RMSE=3.560, Loss=0.0235, RegLoss=0.000]

Epoch 28:  28%|██▊       | 28/100 [00:00<00:00, 81442.80it/s, loss=0.0241, v_num=17, MAE=2.840, RMSE=3.560, Loss=0.0235, RegLoss=0.000]

Epoch 28:  28%|██▊       | 28/100 [00:00<00:00, 1171.94it/s, loss=0.0208, v_num=17, MAE=2.740, RMSE=3.400, Loss=0.0214, RegLoss=0.000]

Epoch 28:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0208, v_num=17, MAE=2.740, RMSE=3.400, Loss=0.0214, RegLoss=0.000]

Epoch 29:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0208, v_num=17, MAE=2.740, RMSE=3.400, Loss=0.0214, RegLoss=0.000]

Epoch 29:  29%|██▉       | 29/100 [00:00<00:00, 79865.28it/s, loss=0.0208, v_num=17, MAE=2.740, RMSE=3.400, Loss=0.0214, RegLoss=0.000]

Epoch 29:  29%|██▉       | 29/100 [00:00<00:00, 1217.46it/s, loss=0.0228, v_num=17, MAE=2.840, RMSE=3.520, Loss=0.0236, RegLoss=0.000]

Epoch 29:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0228, v_num=17, MAE=2.840, RMSE=3.520, Loss=0.0236, RegLoss=0.000]

Epoch 30:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0228, v_num=17, MAE=2.840, RMSE=3.520, Loss=0.0236, RegLoss=0.000]

Epoch 30:  30%|███       | 30/100 [00:00<00:00, 86184.33it/s, loss=0.0228, v_num=17, MAE=2.840, RMSE=3.520, Loss=0.0236, RegLoss=0.000]

Epoch 30:  30%|███       | 30/100 [00:00<00:00, 1280.89it/s, loss=0.0268, v_num=17, MAE=2.930, RMSE=3.690, Loss=0.0259, RegLoss=0.000]

Epoch 30:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0268, v_num=17, MAE=2.930, RMSE=3.690, Loss=0.0259, RegLoss=0.000]

Epoch 31:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0268, v_num=17, MAE=2.930, RMSE=3.690, Loss=0.0259, RegLoss=0.000]

Epoch 31:  31%|███       | 31/100 [00:00<00:00, 107992.88it/s, loss=0.0268, v_num=17, MAE=2.930, RMSE=3.690, Loss=0.0259, RegLoss=0.000]

Epoch 31:  31%|███       | 31/100 [00:00<00:00, 1336.15it/s, loss=0.0235, v_num=17, MAE=2.880, RMSE=3.560, Loss=0.0243, RegLoss=0.000]

Epoch 31:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0235, v_num=17, MAE=2.880, RMSE=3.560, Loss=0.0243, RegLoss=0.000]

Epoch 32:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0235, v_num=17, MAE=2.880, RMSE=3.560, Loss=0.0243, RegLoss=0.000]

Epoch 32:  32%|███▏      | 32/100 [00:00<00:00, 87896.35it/s, loss=0.0235, v_num=17, MAE=2.880, RMSE=3.560, Loss=0.0243, RegLoss=0.000]

Epoch 32:  32%|███▏      | 32/100 [00:00<00:00, 1402.70it/s, loss=0.028, v_num=17, MAE=3.060, RMSE=3.770, Loss=0.0265, RegLoss=0.000]

Epoch 32:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.028, v_num=17, MAE=3.060, RMSE=3.770, Loss=0.0265, RegLoss=0.000]

Epoch 33:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.028, v_num=17, MAE=3.060, RMSE=3.770, Loss=0.0265, RegLoss=0.000]

Epoch 33:  33%|███▎      | 33/100 [00:00<00:00, 95063.21it/s, loss=0.028, v_num=17, MAE=3.060, RMSE=3.770, Loss=0.0265, RegLoss=0.000]

Epoch 33:  33%|███▎      | 33/100 [00:00<00:00, 1406.58it/s, loss=0.023, v_num=17, MAE=2.790, RMSE=3.470, Loss=0.0228, RegLoss=0.000]

Epoch 33:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.023, v_num=17, MAE=2.790, RMSE=3.470, Loss=0.0228, RegLoss=0.000]

Epoch 34:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.023, v_num=17, MAE=2.790, RMSE=3.470, Loss=0.0228, RegLoss=0.000]

Epoch 34:  34%|███▍      | 34/100 [00:00<00:00, 98826.29it/s, loss=0.023, v_num=17, MAE=2.790, RMSE=3.470, Loss=0.0228, RegLoss=0.000]

Epoch 34:  34%|███▍      | 34/100 [00:00<00:00, 1475.83it/s, loss=0.0216, v_num=17, MAE=2.740, RMSE=3.400, Loss=0.0216, RegLoss=0.000]

Epoch 34:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0216, v_num=17, MAE=2.740, RMSE=3.400, Loss=0.0216, RegLoss=0.000]

Epoch 35:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0216, v_num=17, MAE=2.740, RMSE=3.400, Loss=0.0216, RegLoss=0.000]

Epoch 35:  35%|███▌      | 35/100 [00:00<00:00, 103892.88it/s, loss=0.0216, v_num=17, MAE=2.740, RMSE=3.400, Loss=0.0216, RegLoss=0.000]

Epoch 35:  35%|███▌      | 35/100 [00:00<00:00, 1483.36it/s, loss=0.0216, v_num=17, MAE=2.780, RMSE=3.460, Loss=0.0219, RegLoss=0.000]

Epoch 35:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0216, v_num=17, MAE=2.780, RMSE=3.460, Loss=0.0219, RegLoss=0.000]

Epoch 36:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0216, v_num=17, MAE=2.780, RMSE=3.460, Loss=0.0219, RegLoss=0.000]

Epoch 36:  36%|███▌      | 36/100 [00:00<00:00, 99732.46it/s, loss=0.0216, v_num=17, MAE=2.780, RMSE=3.460, Loss=0.0219, RegLoss=0.000]

Epoch 36:  36%|███▌      | 36/100 [00:00<00:00, 1538.41it/s, loss=0.0208, v_num=17, MAE=2.680, RMSE=3.300, Loss=0.0202, RegLoss=0.000]

Epoch 36:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0208, v_num=17, MAE=2.680, RMSE=3.300, Loss=0.0202, RegLoss=0.000]

Epoch 37:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0208, v_num=17, MAE=2.680, RMSE=3.300, Loss=0.0202, RegLoss=0.000]

Epoch 37:  37%|███▋      | 37/100 [00:00<00:00, 109365.22it/s, loss=0.0208, v_num=17, MAE=2.680, RMSE=3.300, Loss=0.0202, RegLoss=0.000]

Epoch 37:  37%|███▋      | 37/100 [00:00<00:00, 1568.31it/s, loss=0.0195, v_num=17, MAE=2.650, RMSE=3.260, Loss=0.0199, RegLoss=0.000]

Epoch 37:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0195, v_num=17, MAE=2.650, RMSE=3.260, Loss=0.0199, RegLoss=0.000]

Epoch 38:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0195, v_num=17, MAE=2.650, RMSE=3.260, Loss=0.0199, RegLoss=0.000]

Epoch 38:  38%|███▊      | 38/100 [00:00<00:00, 128846.85it/s, loss=0.0195, v_num=17, MAE=2.650, RMSE=3.260, Loss=0.0199, RegLoss=0.000]

Epoch 38:  38%|███▊      | 38/100 [00:00<00:00, 1618.14it/s, loss=0.0204, v_num=17, MAE=2.690, RMSE=3.330, Loss=0.0209, RegLoss=0.000]

Epoch 38:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0204, v_num=17, MAE=2.690, RMSE=3.330, Loss=0.0209, RegLoss=0.000]

Epoch 39:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0204, v_num=17, MAE=2.690, RMSE=3.330, Loss=0.0209, RegLoss=0.000]

Epoch 39:  39%|███▉      | 39/100 [00:00<00:00, 110825.11it/s, loss=0.0204, v_num=17, MAE=2.690, RMSE=3.330, Loss=0.0209, RegLoss=0.000]

Epoch 39:  39%|███▉      | 39/100 [00:00<00:00, 1687.43it/s, loss=0.0185, v_num=17, MAE=2.570, RMSE=3.190, Loss=0.019, RegLoss=0.000]

Epoch 39:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0185, v_num=17, MAE=2.570, RMSE=3.190, Loss=0.019, RegLoss=0.000]

Epoch 40:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0185, v_num=17, MAE=2.570, RMSE=3.190, Loss=0.019, RegLoss=0.000]

Epoch 40:  40%|████      | 40/100 [00:00<00:00, 109583.38it/s, loss=0.0185, v_num=17, MAE=2.570, RMSE=3.190, Loss=0.019, RegLoss=0.000]

Epoch 40:  40%|████      | 40/100 [00:00<00:00, 1681.02it/s, loss=0.0185, v_num=17, MAE=2.580, RMSE=3.190, Loss=0.0192, RegLoss=0.000]

Epoch 40:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0185, v_num=17, MAE=2.580, RMSE=3.190, Loss=0.0192, RegLoss=0.000]

Epoch 41:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0185, v_num=17, MAE=2.580, RMSE=3.190, Loss=0.0192, RegLoss=0.000]

Epoch 41:  41%|████      | 41/100 [00:00<00:00, 117785.25it/s, loss=0.0185, v_num=17, MAE=2.580, RMSE=3.190, Loss=0.0192, RegLoss=0.000]

Epoch 41:  41%|████      | 41/100 [00:00<00:00, 1752.88it/s, loss=0.0199, v_num=17, MAE=2.640, RMSE=3.260, Loss=0.0204, RegLoss=0.000]

Epoch 41:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0199, v_num=17, MAE=2.640, RMSE=3.260, Loss=0.0204, RegLoss=0.000]

Epoch 42:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0199, v_num=17, MAE=2.640, RMSE=3.260, Loss=0.0204, RegLoss=0.000]

Epoch 42:  42%|████▏     | 42/100 [00:00<00:00, 118626.78it/s, loss=0.0199, v_num=17, MAE=2.640, RMSE=3.260, Loss=0.0204, RegLoss=0.000]

Epoch 42:  42%|████▏     | 42/100 [00:00<00:00, 1771.40it/s, loss=0.0206, v_num=17, MAE=2.600, RMSE=3.250, Loss=0.0197, RegLoss=0.000]

Epoch 42:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0206, v_num=17, MAE=2.600, RMSE=3.250, Loss=0.0197, RegLoss=0.000]

Epoch 43:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0206, v_num=17, MAE=2.600, RMSE=3.250, Loss=0.0197, RegLoss=0.000]

Epoch 43:  43%|████▎     | 43/100 [00:00<00:00, 128733.10it/s, loss=0.0206, v_num=17, MAE=2.600, RMSE=3.250, Loss=0.0197, RegLoss=0.000]

Epoch 43:  43%|████▎     | 43/100 [00:00<00:00, 1875.10it/s, loss=0.0183, v_num=17, MAE=2.590, RMSE=3.200, Loss=0.019, RegLoss=0.000]

Epoch 43:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0183, v_num=17, MAE=2.590, RMSE=3.200, Loss=0.019, RegLoss=0.000]

Epoch 44:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0183, v_num=17, MAE=2.590, RMSE=3.200, Loss=0.019, RegLoss=0.000]

Epoch 44:  44%|████▍     | 44/100 [00:00<00:00, 119604.26it/s, loss=0.0183, v_num=17, MAE=2.590, RMSE=3.200, Loss=0.019, RegLoss=0.000]

Epoch 44:  44%|████▍     | 44/100 [00:00<00:00, 1881.87it/s, loss=0.0207, v_num=17, MAE=2.720, RMSE=3.370, Loss=0.021, RegLoss=0.000]

Epoch 44:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0207, v_num=17, MAE=2.720, RMSE=3.370, Loss=0.021, RegLoss=0.000]

Epoch 45:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0207, v_num=17, MAE=2.720, RMSE=3.370, Loss=0.021, RegLoss=0.000]

Epoch 45:  45%|████▌     | 45/100 [00:00<00:00, 128659.63it/s, loss=0.0207, v_num=17, MAE=2.720, RMSE=3.370, Loss=0.021, RegLoss=0.000]

Epoch 45:  45%|████▌     | 45/100 [00:00<00:00, 1897.86it/s, loss=0.0182, v_num=17, MAE=2.500, RMSE=3.120, Loss=0.0181, RegLoss=0.000]

Epoch 45:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0182, v_num=17, MAE=2.500, RMSE=3.120, Loss=0.0181, RegLoss=0.000]

Epoch 46:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0182, v_num=17, MAE=2.500, RMSE=3.120, Loss=0.0181, RegLoss=0.000]

Epoch 46:  46%|████▌     | 46/100 [00:00<00:00, 130451.65it/s, loss=0.0182, v_num=17, MAE=2.500, RMSE=3.120, Loss=0.0181, RegLoss=0.000]

Epoch 46:  46%|████▌     | 46/100 [00:00<00:00, 1960.37it/s, loss=0.0197, v_num=17, MAE=2.650, RMSE=3.290, Loss=0.0205, RegLoss=0.000]

Epoch 46:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0197, v_num=17, MAE=2.650, RMSE=3.290, Loss=0.0205, RegLoss=0.000]

Epoch 47:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0197, v_num=17, MAE=2.650, RMSE=3.290, Loss=0.0205, RegLoss=0.000]

Epoch 47:  47%|████▋     | 47/100 [00:00<00:00, 137758.41it/s, loss=0.0197, v_num=17, MAE=2.650, RMSE=3.290, Loss=0.0205, RegLoss=0.000]

Epoch 47:  47%|████▋     | 47/100 [00:00<00:00, 2036.74it/s, loss=0.0198, v_num=17, MAE=2.480, RMSE=3.070, Loss=0.0177, RegLoss=0.000]

Epoch 47:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0198, v_num=17, MAE=2.480, RMSE=3.070, Loss=0.0177, RegLoss=0.000]

Epoch 48:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0198, v_num=17, MAE=2.480, RMSE=3.070, Loss=0.0177, RegLoss=0.000]

Epoch 48:  48%|████▊     | 48/100 [00:00<00:00, 139907.29it/s, loss=0.0198, v_num=17, MAE=2.480, RMSE=3.070, Loss=0.0177, RegLoss=0.000]

Epoch 48:  48%|████▊     | 48/100 [00:00<00:00, 2075.77it/s, loss=0.0183, v_num=17, MAE=2.570, RMSE=3.200, Loss=0.0191, RegLoss=0.000]

Epoch 48:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0183, v_num=17, MAE=2.570, RMSE=3.200, Loss=0.0191, RegLoss=0.000]

Epoch 49:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0183, v_num=17, MAE=2.570, RMSE=3.200, Loss=0.0191, RegLoss=0.000]

Epoch 49:  49%|████▉     | 49/100 [00:00<00:00, 152576.76it/s, loss=0.0183, v_num=17, MAE=2.570, RMSE=3.200, Loss=0.0191, RegLoss=0.000]

Epoch 49:  49%|████▉     | 49/100 [00:00<00:00, 2135.24it/s, loss=0.0185, v_num=17, MAE=2.480, RMSE=3.070, Loss=0.0175, RegLoss=0.000]

Epoch 49:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0185, v_num=17, MAE=2.480, RMSE=3.070, Loss=0.0175, RegLoss=0.000]

Epoch 50:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0185, v_num=17, MAE=2.480, RMSE=3.070, Loss=0.0175, RegLoss=0.000]

Epoch 50:  50%|█████     | 50/100 [00:00<00:00, 143542.23it/s, loss=0.0185, v_num=17, MAE=2.480, RMSE=3.070, Loss=0.0175, RegLoss=0.000]

Epoch 50:  50%|█████     | 50/100 [00:00<00:00, 2182.71it/s, loss=0.0179, v_num=17, MAE=2.570, RMSE=3.170, Loss=0.0187, RegLoss=0.000]

Epoch 50:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0179, v_num=17, MAE=2.570, RMSE=3.170, Loss=0.0187, RegLoss=0.000]

Epoch 51:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0179, v_num=17, MAE=2.570, RMSE=3.170, Loss=0.0187, RegLoss=0.000]

Epoch 51:  51%|█████     | 51/100 [00:00<00:00, 143659.84it/s, loss=0.0179, v_num=17, MAE=2.570, RMSE=3.170, Loss=0.0187, RegLoss=0.000]

Epoch 51:  51%|█████     | 51/100 [00:00<00:00, 2179.30it/s, loss=0.0185, v_num=17, MAE=2.530, RMSE=3.110, Loss=0.018, RegLoss=0.000]

Epoch 51:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0185, v_num=17, MAE=2.530, RMSE=3.110, Loss=0.018, RegLoss=0.000]

Epoch 52:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0185, v_num=17, MAE=2.530, RMSE=3.110, Loss=0.018, RegLoss=0.000]

Epoch 52:  52%|█████▏    | 52/100 [00:00<00:00, 154355.14it/s, loss=0.0185, v_num=17, MAE=2.530, RMSE=3.110, Loss=0.018, RegLoss=0.000]

Epoch 52:  52%|█████▏    | 52/100 [00:00<00:00, 2190.13it/s, loss=0.0183, v_num=17, MAE=2.570, RMSE=3.140, Loss=0.0184, RegLoss=0.000]

Epoch 52:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0183, v_num=17, MAE=2.570, RMSE=3.140, Loss=0.0184, RegLoss=0.000]

Epoch 53:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0183, v_num=17, MAE=2.570, RMSE=3.140, Loss=0.0184, RegLoss=0.000]

Epoch 53:  53%|█████▎    | 53/100 [00:00<00:00, 147314.85it/s, loss=0.0183, v_num=17, MAE=2.570, RMSE=3.140, Loss=0.0184, RegLoss=0.000]

Epoch 53:  53%|█████▎    | 53/100 [00:00<00:00, 2272.92it/s, loss=0.0181, v_num=17, MAE=2.500, RMSE=3.090, Loss=0.0179, RegLoss=0.000]

Epoch 53:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0181, v_num=17, MAE=2.500, RMSE=3.090, Loss=0.0179, RegLoss=0.000]

Epoch 54:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0181, v_num=17, MAE=2.500, RMSE=3.090, Loss=0.0179, RegLoss=0.000]

Epoch 54:  54%|█████▍    | 54/100 [00:00<00:00, 161780.30it/s, loss=0.0181, v_num=17, MAE=2.500, RMSE=3.090, Loss=0.0179, RegLoss=0.000]

Epoch 54:  54%|█████▍    | 54/100 [00:00<00:00, 2352.04it/s, loss=0.0176, v_num=17, MAE=2.530, RMSE=3.150, Loss=0.0181, RegLoss=0.000]

Epoch 54:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0176, v_num=17, MAE=2.530, RMSE=3.150, Loss=0.0181, RegLoss=0.000]

Epoch 55:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0176, v_num=17, MAE=2.530, RMSE=3.150, Loss=0.0181, RegLoss=0.000]

Epoch 55:  55%|█████▌    | 55/100 [00:00<00:00, 168261.65it/s, loss=0.0176, v_num=17, MAE=2.530, RMSE=3.150, Loss=0.0181, RegLoss=0.000]

Epoch 55:  55%|█████▌    | 55/100 [00:00<00:00, 2355.36it/s, loss=0.0174, v_num=17, MAE=2.520, RMSE=3.090, Loss=0.0175, RegLoss=0.000]

Epoch 55:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0174, v_num=17, MAE=2.520, RMSE=3.090, Loss=0.0175, RegLoss=0.000]

Epoch 56:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0174, v_num=17, MAE=2.520, RMSE=3.090, Loss=0.0175, RegLoss=0.000]

Epoch 56:  56%|█████▌    | 56/100 [00:00<00:00, 161875.27it/s, loss=0.0174, v_num=17, MAE=2.520, RMSE=3.090, Loss=0.0175, RegLoss=0.000]

Epoch 56:  56%|█████▌    | 56/100 [00:00<00:00, 2397.77it/s, loss=0.0169, v_num=17, MAE=2.450, RMSE=3.020, Loss=0.0171, RegLoss=0.000]

Epoch 56:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0169, v_num=17, MAE=2.450, RMSE=3.020, Loss=0.0171, RegLoss=0.000]

Epoch 57:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0169, v_num=17, MAE=2.450, RMSE=3.020, Loss=0.0171, RegLoss=0.000]

Epoch 57:  57%|█████▋    | 57/100 [00:00<00:00, 188843.07it/s, loss=0.0169, v_num=17, MAE=2.450, RMSE=3.020, Loss=0.0171, RegLoss=0.000]

Epoch 57:  57%|█████▋    | 57/100 [00:00<00:00, 2462.64it/s, loss=0.0165, v_num=17, MAE=2.430, RMSE=2.990, Loss=0.0168, RegLoss=0.000]

Epoch 57:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0165, v_num=17, MAE=2.430, RMSE=2.990, Loss=0.0168, RegLoss=0.000]

Epoch 58:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0165, v_num=17, MAE=2.430, RMSE=2.990, Loss=0.0168, RegLoss=0.000]

Epoch 58:  58%|█████▊    | 58/100 [00:00<00:00, 154261.02it/s, loss=0.0165, v_num=17, MAE=2.430, RMSE=2.990, Loss=0.0168, RegLoss=0.000]

Epoch 58:  58%|█████▊    | 58/100 [00:00<00:00, 2478.75it/s, loss=0.0166, v_num=17, MAE=2.490, RMSE=3.060, Loss=0.0172, RegLoss=0.000]

Epoch 58:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0166, v_num=17, MAE=2.490, RMSE=3.060, Loss=0.0172, RegLoss=0.000]

Epoch 59:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0166, v_num=17, MAE=2.490, RMSE=3.060, Loss=0.0172, RegLoss=0.000]

Epoch 59:  59%|█████▉    | 59/100 [00:00<00:00, 175009.86it/s, loss=0.0166, v_num=17, MAE=2.490, RMSE=3.060, Loss=0.0172, RegLoss=0.000]

Epoch 59:  59%|█████▉    | 59/100 [00:00<00:00, 2531.11it/s, loss=0.0178, v_num=17, MAE=2.520, RMSE=3.100, Loss=0.018, RegLoss=0.000]

Epoch 59:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0178, v_num=17, MAE=2.520, RMSE=3.100, Loss=0.018, RegLoss=0.000]

Epoch 60:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0178, v_num=17, MAE=2.520, RMSE=3.100, Loss=0.018, RegLoss=0.000]

Epoch 60:  60%|██████    | 60/100 [00:00<00:00, 163096.72it/s, loss=0.0178, v_num=17, MAE=2.520, RMSE=3.100, Loss=0.018, RegLoss=0.000]

Epoch 60:  60%|██████    | 60/100 [00:00<00:00, 2540.31it/s, loss=0.0188, v_num=17, MAE=2.470, RMSE=3.090, Loss=0.0181, RegLoss=0.000]

Epoch 60:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0188, v_num=17, MAE=2.470, RMSE=3.090, Loss=0.0181, RegLoss=0.000]

Epoch 61:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0188, v_num=17, MAE=2.470, RMSE=3.090, Loss=0.0181, RegLoss=0.000]

Epoch 61:  61%|██████    | 61/100 [00:00<00:00, 164008.04it/s, loss=0.0188, v_num=17, MAE=2.470, RMSE=3.090, Loss=0.0181, RegLoss=0.000]

Epoch 61:  61%|██████    | 61/100 [00:00<00:00, 2644.69it/s, loss=0.0175, v_num=17, MAE=2.440, RMSE=3.000, Loss=0.0168, RegLoss=0.000]

Epoch 61:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0175, v_num=17, MAE=2.440, RMSE=3.000, Loss=0.0168, RegLoss=0.000]

Epoch 62:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0175, v_num=17, MAE=2.440, RMSE=3.000, Loss=0.0168, RegLoss=0.000]

Epoch 62:  62%|██████▏   | 62/100 [00:00<00:00, 146505.27it/s, loss=0.0175, v_num=17, MAE=2.440, RMSE=3.000, Loss=0.0168, RegLoss=0.000]

Epoch 62:  62%|██████▏   | 62/100 [00:00<00:00, 2679.43it/s, loss=0.0169, v_num=17, MAE=2.450, RMSE=3.050, Loss=0.0175, RegLoss=0.000]

Epoch 62:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0169, v_num=17, MAE=2.450, RMSE=3.050, Loss=0.0175, RegLoss=0.000]

Epoch 63:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0169, v_num=17, MAE=2.450, RMSE=3.050, Loss=0.0175, RegLoss=0.000]

Epoch 63:  63%|██████▎   | 63/100 [00:00<00:00, 187538.08it/s, loss=0.0169, v_num=17, MAE=2.450, RMSE=3.050, Loss=0.0175, RegLoss=0.000]

Epoch 63:  63%|██████▎   | 63/100 [00:00<00:00, 2683.50it/s, loss=0.0174, v_num=17, MAE=2.450, RMSE=3.020, Loss=0.0172, RegLoss=0.000]

Epoch 63:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0174, v_num=17, MAE=2.450, RMSE=3.020, Loss=0.0172, RegLoss=0.000]

Epoch 64:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0174, v_num=17, MAE=2.450, RMSE=3.020, Loss=0.0172, RegLoss=0.000]

Epoch 64:  64%|██████▍   | 64/100 [00:00<00:00, 186154.96it/s, loss=0.0174, v_num=17, MAE=2.450, RMSE=3.020, Loss=0.0172, RegLoss=0.000]

Epoch 64:  64%|██████▍   | 64/100 [00:00<00:00, 2701.92it/s, loss=0.0188, v_num=17, MAE=2.580, RMSE=3.200, Loss=0.0189, RegLoss=0.000]

Epoch 64:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0188, v_num=17, MAE=2.580, RMSE=3.200, Loss=0.0189, RegLoss=0.000]

Epoch 65:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0188, v_num=17, MAE=2.580, RMSE=3.200, Loss=0.0189, RegLoss=0.000]

Epoch 65:  65%|██████▌   | 65/100 [00:00<00:00, 150457.92it/s, loss=0.0188, v_num=17, MAE=2.580, RMSE=3.200, Loss=0.0189, RegLoss=0.000]

Epoch 65:  65%|██████▌   | 65/100 [00:00<00:00, 2815.81it/s, loss=0.0196, v_num=17, MAE=2.500, RMSE=3.100, Loss=0.0181, RegLoss=0.000]

Epoch 65:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0196, v_num=17, MAE=2.500, RMSE=3.100, Loss=0.0181, RegLoss=0.000]

Epoch 66:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0196, v_num=17, MAE=2.500, RMSE=3.100, Loss=0.0181, RegLoss=0.000]

Epoch 66:  66%|██████▌   | 66/100 [00:00<00:00, 191839.27it/s, loss=0.0196, v_num=17, MAE=2.500, RMSE=3.100, Loss=0.0181, RegLoss=0.000]

Epoch 66:  66%|██████▌   | 66/100 [00:00<00:00, 2800.30it/s, loss=0.0179, v_num=17, MAE=2.470, RMSE=3.040, Loss=0.0171, RegLoss=0.000]

Epoch 66:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0179, v_num=17, MAE=2.470, RMSE=3.040, Loss=0.0171, RegLoss=0.000]

Epoch 67:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0179, v_num=17, MAE=2.470, RMSE=3.040, Loss=0.0171, RegLoss=0.000]

Epoch 67:  67%|██████▋   | 67/100 [00:00<00:00, 185613.19it/s, loss=0.0179, v_num=17, MAE=2.470, RMSE=3.040, Loss=0.0171, RegLoss=0.000]

Epoch 67:  67%|██████▋   | 67/100 [00:00<00:00, 2863.97it/s, loss=0.0162, v_num=17, MAE=2.370, RMSE=2.930, Loss=0.0164, RegLoss=0.000]

Epoch 67:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0162, v_num=17, MAE=2.370, RMSE=2.930, Loss=0.0164, RegLoss=0.000]

Epoch 68:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0162, v_num=17, MAE=2.370, RMSE=2.930, Loss=0.0164, RegLoss=0.000]

Epoch 68:  68%|██████▊   | 68/100 [00:00<00:00, 190395.64it/s, loss=0.0162, v_num=17, MAE=2.370, RMSE=2.930, Loss=0.0164, RegLoss=0.000]

Epoch 68:  68%|██████▊   | 68/100 [00:00<00:00, 2925.89it/s, loss=0.017, v_num=17, MAE=2.500, RMSE=3.100, Loss=0.0178, RegLoss=0.000]

Epoch 68:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.017, v_num=17, MAE=2.500, RMSE=3.100, Loss=0.0178, RegLoss=0.000]

Epoch 69:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.017, v_num=17, MAE=2.500, RMSE=3.100, Loss=0.0178, RegLoss=0.000]

Epoch 69:  69%|██████▉   | 69/100 [00:00<00:00, 187926.61it/s, loss=0.017, v_num=17, MAE=2.500, RMSE=3.100, Loss=0.0178, RegLoss=0.000]

Epoch 69:  69%|██████▉   | 69/100 [00:00<00:00, 3010.61it/s, loss=0.0181, v_num=17, MAE=2.460, RMSE=3.050, Loss=0.0175, RegLoss=0.000]

Epoch 69:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0181, v_num=17, MAE=2.460, RMSE=3.050, Loss=0.0175, RegLoss=0.000]

Epoch 70:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0181, v_num=17, MAE=2.460, RMSE=3.050, Loss=0.0175, RegLoss=0.000]

Epoch 70:  70%|███████   | 70/100 [00:00<00:00, 203043.76it/s, loss=0.0181, v_num=17, MAE=2.460, RMSE=3.050, Loss=0.0175, RegLoss=0.000]

Epoch 70:  70%|███████   | 70/100 [00:00<00:00, 2993.88it/s, loss=0.0176, v_num=17, MAE=2.500, RMSE=3.070, Loss=0.0173, RegLoss=0.000]

Epoch 70:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0176, v_num=17, MAE=2.500, RMSE=3.070, Loss=0.0173, RegLoss=0.000]

Epoch 71:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0176, v_num=17, MAE=2.500, RMSE=3.070, Loss=0.0173, RegLoss=0.000]

Epoch 71:  71%|███████   | 71/100 [00:00<00:00, 229780.54it/s, loss=0.0176, v_num=17, MAE=2.500, RMSE=3.070, Loss=0.0173, RegLoss=0.000]

Epoch 71:  71%|███████   | 71/100 [00:00<00:00, 3090.83it/s, loss=0.0169, v_num=17, MAE=2.420, RMSE=2.990, Loss=0.0167, RegLoss=0.000]

Epoch 71:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0169, v_num=17, MAE=2.420, RMSE=2.990, Loss=0.0167, RegLoss=0.000]

Epoch 72:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0169, v_num=17, MAE=2.420, RMSE=2.990, Loss=0.0167, RegLoss=0.000]

Epoch 72:  72%|███████▏  | 72/100 [00:00<00:00, 202813.89it/s, loss=0.0169, v_num=17, MAE=2.420, RMSE=2.990, Loss=0.0167, RegLoss=0.000]

Epoch 72:  72%|███████▏  | 72/100 [00:00<00:00, 3142.45it/s, loss=0.0164, v_num=17, MAE=2.410, RMSE=2.980, Loss=0.0167, RegLoss=0.000]

Epoch 72:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0164, v_num=17, MAE=2.410, RMSE=2.980, Loss=0.0167, RegLoss=0.000]

Epoch 73:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0164, v_num=17, MAE=2.410, RMSE=2.980, Loss=0.0167, RegLoss=0.000]

Epoch 73:  73%|███████▎  | 73/100 [00:00<00:00, 196271.92it/s, loss=0.0164, v_num=17, MAE=2.410, RMSE=2.980, Loss=0.0167, RegLoss=0.000]

Epoch 73:  73%|███████▎  | 73/100 [00:00<00:00, 3120.22it/s, loss=0.0163, v_num=17, MAE=2.420, RMSE=2.990, Loss=0.0166, RegLoss=0.000]

Epoch 73:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0163, v_num=17, MAE=2.420, RMSE=2.990, Loss=0.0166, RegLoss=0.000]

Epoch 74:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0163, v_num=17, MAE=2.420, RMSE=2.990, Loss=0.0166, RegLoss=0.000]

Epoch 74:  74%|███████▍  | 74/100 [00:00<00:00, 225565.77it/s, loss=0.0163, v_num=17, MAE=2.420, RMSE=2.990, Loss=0.0166, RegLoss=0.000]

Epoch 74:  74%|███████▍  | 74/100 [00:00<00:00, 3093.27it/s, loss=0.0163, v_num=17, MAE=2.410, RMSE=2.990, Loss=0.0166, RegLoss=0.000]

Epoch 74:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0163, v_num=17, MAE=2.410, RMSE=2.990, Loss=0.0166, RegLoss=0.000]

Epoch 75:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0163, v_num=17, MAE=2.410, RMSE=2.990, Loss=0.0166, RegLoss=0.000]

Epoch 75:  75%|███████▌  | 75/100 [00:00<00:00, 218605.14it/s, loss=0.0163, v_num=17, MAE=2.410, RMSE=2.990, Loss=0.0166, RegLoss=0.000]

Epoch 75:  75%|███████▌  | 75/100 [00:00<00:00, 3203.36it/s, loss=0.0161, v_num=17, MAE=2.380, RMSE=2.960, Loss=0.0164, RegLoss=0.000]

Epoch 75:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0161, v_num=17, MAE=2.380, RMSE=2.960, Loss=0.0164, RegLoss=0.000]

Epoch 76:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0161, v_num=17, MAE=2.380, RMSE=2.960, Loss=0.0164, RegLoss=0.000]

Epoch 76:  76%|███████▌  | 76/100 [00:00<00:00, 236649.67it/s, loss=0.0161, v_num=17, MAE=2.380, RMSE=2.960, Loss=0.0164, RegLoss=0.000]

Epoch 76:  76%|███████▌  | 76/100 [00:00<00:00, 3341.23it/s, loss=0.016, v_num=17, MAE=2.390, RMSE=2.960, Loss=0.0165, RegLoss=0.000]

Epoch 76:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.016, v_num=17, MAE=2.390, RMSE=2.960, Loss=0.0165, RegLoss=0.000]

Epoch 77:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.016, v_num=17, MAE=2.390, RMSE=2.960, Loss=0.0165, RegLoss=0.000]

Epoch 77:  77%|███████▋  | 77/100 [00:00<00:00, 255912.37it/s, loss=0.016, v_num=17, MAE=2.390, RMSE=2.960, Loss=0.0165, RegLoss=0.000]

Epoch 77:  77%|███████▋  | 77/100 [00:00<00:00, 3381.09it/s, loss=0.016, v_num=17, MAE=2.380, RMSE=2.960, Loss=0.0164, RegLoss=0.000]

Epoch 77:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.016, v_num=17, MAE=2.380, RMSE=2.960, Loss=0.0164, RegLoss=0.000]

Epoch 78:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.016, v_num=17, MAE=2.380, RMSE=2.960, Loss=0.0164, RegLoss=0.000]

Epoch 78:  78%|███████▊  | 78/100 [00:00<00:00, 259030.65it/s, loss=0.016, v_num=17, MAE=2.380, RMSE=2.960, Loss=0.0164, RegLoss=0.000]

Epoch 78:  78%|███████▊  | 78/100 [00:00<00:00, 3401.35it/s, loss=0.016, v_num=17, MAE=2.430, RMSE=2.980, Loss=0.0166, RegLoss=0.000]

Epoch 78:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.016, v_num=17, MAE=2.430, RMSE=2.980, Loss=0.0166, RegLoss=0.000]

Epoch 79:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.016, v_num=17, MAE=2.430, RMSE=2.980, Loss=0.0166, RegLoss=0.000]

Epoch 79:  79%|███████▉  | 79/100 [00:00<00:00, 224339.89it/s, loss=0.016, v_num=17, MAE=2.430, RMSE=2.980, Loss=0.0166, RegLoss=0.000]

Epoch 79:  79%|███████▉  | 79/100 [00:00<00:00, 3398.85it/s, loss=0.016, v_num=17, MAE=2.390, RMSE=2.970, Loss=0.0164, RegLoss=0.000]

Epoch 79:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.016, v_num=17, MAE=2.390, RMSE=2.970, Loss=0.0164, RegLoss=0.000]

Epoch 80:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.016, v_num=17, MAE=2.390, RMSE=2.970, Loss=0.0164, RegLoss=0.000]

Epoch 80:  80%|████████  | 80/100 [00:00<00:00, 232532.45it/s, loss=0.016, v_num=17, MAE=2.390, RMSE=2.970, Loss=0.0164, RegLoss=0.000]

Epoch 80:  80%|████████  | 80/100 [00:00<00:00, 3435.53it/s, loss=0.0156, v_num=17, MAE=2.340, RMSE=2.910, Loss=0.0161, RegLoss=0.000]

Epoch 80:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0156, v_num=17, MAE=2.340, RMSE=2.910, Loss=0.0161, RegLoss=0.000]

Epoch 81:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0156, v_num=17, MAE=2.340, RMSE=2.910, Loss=0.0161, RegLoss=0.000]

Epoch 81:  81%|████████  | 81/100 [00:00<00:00, 227554.34it/s, loss=0.0156, v_num=17, MAE=2.340, RMSE=2.910, Loss=0.0161, RegLoss=0.000]

Epoch 81:  81%|████████  | 81/100 [00:00<00:00, 3401.64it/s, loss=0.0157, v_num=17, MAE=2.400, RMSE=2.970, Loss=0.0165, RegLoss=0.000]

Epoch 81:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0157, v_num=17, MAE=2.400, RMSE=2.970, Loss=0.0165, RegLoss=0.000]

Epoch 82:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0157, v_num=17, MAE=2.400, RMSE=2.970, Loss=0.0165, RegLoss=0.000]

Epoch 82:  82%|████████▏ | 82/100 [00:00<00:00, 224645.94it/s, loss=0.0157, v_num=17, MAE=2.400, RMSE=2.970, Loss=0.0165, RegLoss=0.000]

Epoch 82:  82%|████████▏ | 82/100 [00:00<00:00, 3317.25it/s, loss=0.016, v_num=17, MAE=2.410, RMSE=2.970, Loss=0.0164, RegLoss=0.000]

Epoch 82:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.016, v_num=17, MAE=2.410, RMSE=2.970, Loss=0.0164, RegLoss=0.000]

Epoch 83:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.016, v_num=17, MAE=2.410, RMSE=2.970, Loss=0.0164, RegLoss=0.000]

Epoch 83:  83%|████████▎ | 83/100 [00:00<00:00, 217851.83it/s, loss=0.016, v_num=17, MAE=2.410, RMSE=2.970, Loss=0.0164, RegLoss=0.000]

Epoch 83:  83%|████████▎ | 83/100 [00:00<00:00, 3462.23it/s, loss=0.0162, v_num=17, MAE=2.420, RMSE=2.980, Loss=0.0167, RegLoss=0.000]

Epoch 83:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0162, v_num=17, MAE=2.420, RMSE=2.980, Loss=0.0167, RegLoss=0.000]

Epoch 84:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0162, v_num=17, MAE=2.420, RMSE=2.980, Loss=0.0167, RegLoss=0.000]

Epoch 84:  84%|████████▍ | 84/100 [00:00<00:00, 223980.63it/s, loss=0.0162, v_num=17, MAE=2.420, RMSE=2.980, Loss=0.0167, RegLoss=0.000]

Epoch 84:  84%|████████▍ | 84/100 [00:00<00:00, 3557.51it/s, loss=0.0162, v_num=17, MAE=2.360, RMSE=2.940, Loss=0.0163, RegLoss=0.000]

Epoch 84:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0162, v_num=17, MAE=2.360, RMSE=2.940, Loss=0.0163, RegLoss=0.000]

Epoch 85:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0162, v_num=17, MAE=2.360, RMSE=2.940, Loss=0.0163, RegLoss=0.000]

Epoch 85:  85%|████████▌ | 85/100 [00:00<00:00, 163539.38it/s, loss=0.0162, v_num=17, MAE=2.360, RMSE=2.940, Loss=0.0163, RegLoss=0.000]

Epoch 85:  85%|████████▌ | 85/100 [00:00<00:00, 3584.69it/s, loss=0.0164, v_num=17, MAE=2.470, RMSE=3.040, Loss=0.0169, RegLoss=0.000]

Epoch 85:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0164, v_num=17, MAE=2.470, RMSE=3.040, Loss=0.0169, RegLoss=0.000]

Epoch 86:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0164, v_num=17, MAE=2.470, RMSE=3.040, Loss=0.0169, RegLoss=0.000]

Epoch 86:  86%|████████▌ | 86/100 [00:00<00:00, 242902.45it/s, loss=0.0164, v_num=17, MAE=2.470, RMSE=3.040, Loss=0.0169, RegLoss=0.000]

Epoch 86:  86%|████████▌ | 86/100 [00:00<00:00, 3632.68it/s, loss=0.0164, v_num=17, MAE=2.390, RMSE=2.950, Loss=0.0163, RegLoss=0.000]

Epoch 86:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0164, v_num=17, MAE=2.390, RMSE=2.950, Loss=0.0163, RegLoss=0.000]

Epoch 87:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0164, v_num=17, MAE=2.390, RMSE=2.950, Loss=0.0163, RegLoss=0.000]

Epoch 87:  87%|████████▋ | 87/100 [00:00<00:00, 250105.86it/s, loss=0.0164, v_num=17, MAE=2.390, RMSE=2.950, Loss=0.0163, RegLoss=0.000]

Epoch 87:  87%|████████▋ | 87/100 [00:00<00:00, 3721.46it/s, loss=0.0162, v_num=17, MAE=2.430, RMSE=3.000, Loss=0.0168, RegLoss=0.000]

Epoch 87:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0162, v_num=17, MAE=2.430, RMSE=3.000, Loss=0.0168, RegLoss=0.000]

Epoch 88:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0162, v_num=17, MAE=2.430, RMSE=3.000, Loss=0.0168, RegLoss=0.000]

Epoch 88:  88%|████████▊ | 88/100 [00:00<00:00, 235244.58it/s, loss=0.0162, v_num=17, MAE=2.430, RMSE=3.000, Loss=0.0168, RegLoss=0.000]

Epoch 88:  88%|████████▊ | 88/100 [00:00<00:00, 3678.00it/s, loss=0.0162, v_num=17, MAE=2.370, RMSE=2.940, Loss=0.0161, RegLoss=0.000]

Epoch 88:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0162, v_num=17, MAE=2.370, RMSE=2.940, Loss=0.0161, RegLoss=0.000]

Epoch 89:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0162, v_num=17, MAE=2.370, RMSE=2.940, Loss=0.0161, RegLoss=0.000]

Epoch 89:  89%|████████▉ | 89/100 [00:00<00:00, 258155.64it/s, loss=0.0162, v_num=17, MAE=2.370, RMSE=2.940, Loss=0.0161, RegLoss=0.000]

Epoch 89:  89%|████████▉ | 89/100 [00:00<00:00, 3780.03it/s, loss=0.016, v_num=17, MAE=2.430, RMSE=2.980, Loss=0.0167, RegLoss=0.000]

Epoch 89:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.016, v_num=17, MAE=2.430, RMSE=2.980, Loss=0.0167, RegLoss=0.000]

Epoch 90:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.016, v_num=17, MAE=2.430, RMSE=2.980, Loss=0.0167, RegLoss=0.000]

Epoch 90:  90%|█████████ | 90/100 [00:00<00:00, 224427.68it/s, loss=0.016, v_num=17, MAE=2.430, RMSE=2.980, Loss=0.0167, RegLoss=0.000]

Epoch 90:  90%|█████████ | 90/100 [00:00<00:00, 3934.46it/s, loss=0.0161, v_num=17, MAE=2.370, RMSE=2.950, Loss=0.0161, RegLoss=0.000]

Epoch 90:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0161, v_num=17, MAE=2.370, RMSE=2.950, Loss=0.0161, RegLoss=0.000]

Epoch 91:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0161, v_num=17, MAE=2.370, RMSE=2.950, Loss=0.0161, RegLoss=0.000]

Epoch 91:  91%|█████████ | 91/100 [00:00<00:00, 258591.91it/s, loss=0.0161, v_num=17, MAE=2.370, RMSE=2.950, Loss=0.0161, RegLoss=0.000]

Epoch 91:  91%|█████████ | 91/100 [00:00<00:00, 3857.79it/s, loss=0.0187, v_num=17, MAE=2.530, RMSE=3.140, Loss=0.0195, RegLoss=0.000]

Epoch 91:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0187, v_num=17, MAE=2.530, RMSE=3.140, Loss=0.0195, RegLoss=0.000]

Epoch 92:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0187, v_num=17, MAE=2.530, RMSE=3.140, Loss=0.0195, RegLoss=0.000]

Epoch 92:  92%|█████████▏| 92/100 [00:00<00:00, 204491.77it/s, loss=0.0187, v_num=17, MAE=2.530, RMSE=3.140, Loss=0.0195, RegLoss=0.000]

Epoch 92:  92%|█████████▏| 92/100 [00:00<00:00, 3858.61it/s, loss=0.0198, v_num=17, MAE=2.500, RMSE=3.060, Loss=0.0171, RegLoss=0.000]

Epoch 92:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0198, v_num=17, MAE=2.500, RMSE=3.060, Loss=0.0171, RegLoss=0.000]

Epoch 93:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0198, v_num=17, MAE=2.500, RMSE=3.060, Loss=0.0171, RegLoss=0.000]

Epoch 93:  93%|█████████▎| 93/100 [00:00<00:00, 269385.55it/s, loss=0.0198, v_num=17, MAE=2.500, RMSE=3.060, Loss=0.0171, RegLoss=0.000]

Epoch 93:  93%|█████████▎| 93/100 [00:00<00:00, 3978.04it/s, loss=0.0164, v_num=17, MAE=2.350, RMSE=2.910, Loss=0.016, RegLoss=0.000]

Epoch 93:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0164, v_num=17, MAE=2.350, RMSE=2.910, Loss=0.016, RegLoss=0.000]

Epoch 94:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0164, v_num=17, MAE=2.350, RMSE=2.910, Loss=0.016, RegLoss=0.000]

Epoch 94:  94%|█████████▍| 94/100 [00:00<00:00, 266936.07it/s, loss=0.0164, v_num=17, MAE=2.350, RMSE=2.910, Loss=0.016, RegLoss=0.000]

Epoch 94:  94%|█████████▍| 94/100 [00:00<00:00, 4014.83it/s, loss=0.0161, v_num=17, MAE=2.450, RMSE=3.020, Loss=0.0169, RegLoss=0.000]

Epoch 94:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0161, v_num=17, MAE=2.450, RMSE=3.020, Loss=0.0169, RegLoss=0.000]

Epoch 95:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0161, v_num=17, MAE=2.450, RMSE=3.020, Loss=0.0169, RegLoss=0.000]

Epoch 95:  95%|█████████▌| 95/100 [00:00<00:00, 295812.09it/s, loss=0.0161, v_num=17, MAE=2.450, RMSE=3.020, Loss=0.0169, RegLoss=0.000]

Epoch 95:  95%|█████████▌| 95/100 [00:00<00:00, 4110.58it/s, loss=0.017, v_num=17, MAE=2.410, RMSE=3.000, Loss=0.0169, RegLoss=0.000]

Epoch 95:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.017, v_num=17, MAE=2.410, RMSE=3.000, Loss=0.0169, RegLoss=0.000]

Epoch 96:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.017, v_num=17, MAE=2.410, RMSE=3.000, Loss=0.0169, RegLoss=0.000]

Epoch 96:  96%|█████████▌| 96/100 [00:00<00:00, 300936.61it/s, loss=0.017, v_num=17, MAE=2.410, RMSE=3.000, Loss=0.0169, RegLoss=0.000]

Epoch 96:  96%|█████████▌| 96/100 [00:00<00:00, 4054.39it/s, loss=0.0167, v_num=17, MAE=2.420, RMSE=2.980, Loss=0.0165, RegLoss=0.000]

Epoch 96:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0167, v_num=17, MAE=2.420, RMSE=2.980, Loss=0.0165, RegLoss=0.000]

Epoch 97:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0167, v_num=17, MAE=2.420, RMSE=2.980, Loss=0.0165, RegLoss=0.000]

Epoch 97:  97%|█████████▋| 97/100 [00:00<00:00, 282141.12it/s, loss=0.0167, v_num=17, MAE=2.420, RMSE=2.980, Loss=0.0165, RegLoss=0.000]

Epoch 97:  97%|█████████▋| 97/100 [00:00<00:00, 4205.32it/s, loss=0.0165, v_num=17, MAE=2.420, RMSE=2.980, Loss=0.0166, RegLoss=0.000]

Epoch 97:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0165, v_num=17, MAE=2.420, RMSE=2.980, Loss=0.0166, RegLoss=0.000]

Epoch 98:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0165, v_num=17, MAE=2.420, RMSE=2.980, Loss=0.0166, RegLoss=0.000]

Epoch 98:  98%|█████████▊| 98/100 [00:00<00:00, 267083.69it/s, loss=0.0165, v_num=17, MAE=2.420, RMSE=2.980, Loss=0.0166, RegLoss=0.000]

Epoch 98:  98%|█████████▊| 98/100 [00:00<00:00, 4317.71it/s, loss=0.0161, v_num=17, MAE=2.380, RMSE=2.940, Loss=0.0162, RegLoss=0.000]

Epoch 98:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0161, v_num=17, MAE=2.380, RMSE=2.940, Loss=0.0162, RegLoss=0.000]

Epoch 99:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0161, v_num=17, MAE=2.380, RMSE=2.940, Loss=0.0162, RegLoss=0.000]

Epoch 99:  99%|█████████▉| 99/100 [00:00<00:00, 297234.14it/s, loss=0.0161, v_num=17, MAE=2.380, RMSE=2.940, Loss=0.0162, RegLoss=0.000]

Epoch 99:  99%|█████████▉| 99/100 [00:00<00:00, 4174.40it/s, loss=0.0163, v_num=17, MAE=2.460, RMSE=3.020, Loss=0.0169, RegLoss=0.000]

Epoch 99:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0163, v_num=17, MAE=2.460, RMSE=3.020, Loss=0.0169, RegLoss=0.000]

Epoch 100:   0%|          | 0/100 [00:00<?, ?it/s, loss=0.0163, v_num=17, MAE=2.460, RMSE=3.020, Loss=0.0169, RegLoss=0.000]

Epoch 100: 100%|██████████| 100/100 [00:00<00:00, 292489.82it/s, loss=0.0163, v_num=17, MAE=2.460, RMSE=3.020, Loss=0.0169, RegLoss=0.000]

Epoch 100: 100%|██████████| 100/100 [00:00<00:00, 4383.08it/s, loss=0.0174, v_num=17, MAE=2.470, RMSE=3.030, Loss=0.0173, RegLoss=0.000]

Epoch 100: 100%|██████████| 100/100 [00:00<00:00, 4190.95it/s, loss=0.0174, v_num=17, MAE=2.470, RMSE=3.030, Loss=0.0173, RegLoss=0.000]

Predicting: 19it [00:00, ?it/s]

Predicting:   0%|          | 0/2 [00:00<-00:00, -1695569.70it/s]

Predicting DataLoader 0:   0%|          | 0/2 [00:00<?, ?it/s]

Predicting DataLoader 0: 100%|██████████| 2/2 [00:00<00:00, 770.37it/s]

Predicting DataLoader 0: 100%|██████████| 2/2 [00:00<00:00, 674.05it/s]

11:11:15  WARNING  its2s.diagnostics - Ljung-Box test (p=0.0000) suggests significant residual autocorrelation for model 'neuralprophet'. Bootstrap CIs may undercover.

Predicting: 19it [00:00, ?it/s]

Predicting:   0%|          | 0/1 [00:00<-00:00, -2414902.30it/s]

Predicting DataLoader 0:   0%|          | 0/1 [00:00<?, ?it/s]

Predicting DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 941.69it/s]

Predicting DataLoader 0: 100%|██████████| 1/1 [00:00<00:00, 726.79it/s]

Predicting: 19it [00:00, ?it/s]

Predicting:   0%|          | 0/2 [00:00<-00:00, -1048576.00it/s]

Predicting DataLoader 0:   0%|          | 0/2 [00:00<?, ?it/s]

Predicting DataLoader 0: 100%|██████████| 2/2 [00:00<00:00, 752.68it/s]

Predicting DataLoader 0: 100%|██████████| 2/2 [00:00<00:00, 664.18it/s]

PipelineResult fields:
  model_name       : neuralprophet
  fit_result       : FitResult with 1169 fitted values
  bootstrap_result : BootstrapCIResult  pred_matrix shape = (407, 100)
  n_successful sims: 0
  metrics_train    : MetricsResult(rmse=3.008841134522328, mae=2.415157090262152, mape=4.715230099525625, mase=None, mase_m=7, mase_denominator=None)
  metrics_test     : MetricsResult(rmse=3.189931623510256, mae=2.5421783625209797, mape=4.803648382383328, mase=1.105892374419183, mase_m=7, mase_denominator=2.298757475252632)

4e. Metrics and excess table¶

MASE is reported for the held-out test window only (the Train cell is NaN by design): it is the ratio of the model's MAE to the in-sample MAE of the seasonal-naive forecast at the resolved period m (mase_m and mase_denominator on the result). A value below 1 means the model beats the seasonal-naive benchmark. The test window serves as a single-use adequacy check of the fitted model, not a retuning target.

In [7]:

Copied!





metrics_df = pd.DataFrame({
    "RMSE":  [result.metrics_train.rmse,  result.metrics_test.rmse],
    "MAE":   [result.metrics_train.mae,   result.metrics_test.mae],
    "MAPE":  [result.metrics_train.mape,  result.metrics_test.mape],
    "MASE":  [result.metrics_train.mase,  result.metrics_test.mase],
}, index=["Train", "Test"])
print(metrics_df.round(3).to_string())

mt = result.metrics_test
print(f"\nMASE benchmark: in-sample seasonal-naive MAE = "
      f"{mt.mase_denominator:.3f} at m = {mt.mase_m}")
metrics_df = pd.DataFrame({
    "RMSE":  [result.metrics_train.rmse,  result.metrics_test.rmse],
    "MAE":   [result.metrics_train.mae,   result.metrics_test.mae],
    "MAPE":  [result.metrics_train.mape,  result.metrics_test.mape],
    "MASE":  [result.metrics_train.mase,  result.metrics_test.mase],
}, index=["Train", "Test"])
print(metrics_df.round(3).to_string())

mt = result.metrics_test
print(f"\nMASE benchmark: in-sample seasonal-naive MAE = "
      f"{mt.mase_denominator:.3f} at m = {mt.mase_m}")

        RMSE    MAE   MAPE   MASE
Train  3.009  2.415  4.715    NaN
Test   3.190  2.542  4.804  1.106

MASE benchmark: in-sample seasonal-naive MAE = 2.299 at m = 7

In [8]:

Copied!





print("Period-level excess:")
print(result.excess_table.period_excess.to_string(index=False))
print("\nDaily excess - first 10 holdout days:")
print(result.excess_table.obs_excess.head(10).to_string(index=False))
print("Period-level excess:")
print(result.excess_table.period_excess.to_string(index=False))
print("\nDaily excess - first 10 holdout days:")
print(result.excess_table.obs_excess.head(10).to_string(index=False))

Period-level excess:
      period start_date   end_date  n_obs  total_observed  total_expected  total_excess  excess_ci_lo  excess_ci_hi  excess_pct
Full holdout 2022-03-15 2022-04-25     42     3029.800778     2970.226562     59.574215   3029.800778   3029.800778    2.005713

Daily excess - first 10 holdout days:
      date  observed  expected  expected_ci_lo  expected_ci_hi    excess  excess_ci_lo  excess_ci_hi  excess_pct  excess_pct_ci_lo  excess_pct_ci_hi
2022-03-15 76.449700 64.602005             NaN             NaN 11.847695           NaN           NaN   18.339516               NaN               NaN
2022-03-16 70.565458 65.051552             NaN             NaN  5.513906           NaN           NaN    8.476210               NaN               NaN
2022-03-17 74.032036 68.638535             NaN             NaN  5.393501           NaN           NaN    7.857833               NaN               NaN
2022-03-18 69.205309 64.688263             NaN             NaN  4.517046           NaN           NaN    6.982791               NaN               NaN
2022-03-19 70.763015 71.350449             NaN             NaN -0.587434           NaN           NaN   -0.823308               NaN               NaN
2022-03-20 75.696935 63.502388             NaN             NaN 12.194547           NaN           NaN   19.203289               NaN               NaN
2022-03-21 72.180154 62.357079             NaN             NaN  9.823075           NaN           NaN   15.752944               NaN               NaN
2022-03-22 74.416647 64.414391             NaN             NaN 10.002256           NaN           NaN   15.527984               NaN               NaN
2022-03-23 73.242643 58.376335             NaN             NaN 14.866308           NaN           NaN   25.466327               NaN               NaN
2022-03-24 72.587821 68.863937             NaN             NaN  3.723884           NaN           NaN    5.407597               NaN               NaN

Note — NaN confidence intervals (known limitation). All CI columns (expected_ci_lo, expected_ci_hi, excess_ci_lo, excess_ci_hi) are NaN because all 100 bootstrap simulations failed (n_successful = 0 above). The root cause is NeuralProphet's AR warmup: the first n_lags rows of fit_result.fitted_values are NaN, and Moving Block Bootstrap builds the perturbed training series as fitted_values + resampled_residuals — leaving those rows as NaN. NeuralProphet cannot refit on a training series with NaN target values, so every simulation errors out. This is a known incompatibility between NeuralProphet and the current MBB implementation; a fix is tracked in a separate issue.

In [9]:

Copied!





from its2s.metrics.excess import calc_ate_summary

ate = calc_ate_summary(result.excess_table.obs_excess)
print("Average Treatment Effect (ATE) summary:")
print(ate.to_string(index=False))
print("\n  Total ATE      = sum of daily excess over full holdout")
print("  Mean ATE per obs = average excess per observation")
print(f"  Simulated effect was +8/day for {HOLDOUT_DAYS} days -> expected total excess ~{8 * HOLDOUT_DAYS}")
from its2s.metrics.excess import calc_ate_summary

ate = calc_ate_summary(result.excess_table.obs_excess)
print("Average Treatment Effect (ATE) summary:")
print(ate.to_string(index=False))
print("\n  Total ATE      = sum of daily excess over full holdout")
print("  Mean ATE per obs = average excess per observation")
print(f"  Simulated effect was +8/day for {HOLDOUT_DAYS} days -> expected total excess ~{8 * HOLDOUT_DAYS}")

Average Treatment Effect (ATE) summary:
          metric  estimate  ci_lo  ci_hi  n_obs
       Total ATE 59.574105    0.0    0.0     42
Mean ATE per obs  1.418431    0.0    0.0     42

  Total ATE      = sum of daily excess over full holdout
  Mean ATE per obs = average excess per observation
  Simulated effect was +8/day for 42 days -> expected total excess ~336

4f. Counterfactual plot (annotated)¶

In [10]:

Copied!





br = result.bootstrap_result
pred_dates = pd.to_datetime(br.dates)
intervention_ts = pd.Timestamp(INTERVENTION)

fig, ax = plt.subplots(figsize=(14, 5))

for part in [splits.train_df, splits.test_df, splits.holdout_df]:
    ax.plot(part["ds"], part["y"], color="#333333", linewidth=0.6, alpha=0.7)
ax.plot([], [], color="#333333", linewidth=0.6, alpha=0.7, label="Observed")

ax.plot(pred_dates, br.predicted, color="#B2182B", linewidth=1.4,
        label="Counterfactual (no-intervention)")
ax.fill_between(pred_dates, br.conf_lo, br.conf_hi,
                color="#B2182B", alpha=0.15, label="95% CI (MBB)")

ax.axvspan(intervention_ts, splits.holdout_df["ds"].max(),
           color="#FEE08B", alpha=0.25, label="Holdout (post-intervention)")
ax.axvline(intervention_ts, color="#4DAF4A", linestyle="--", linewidth=1.3,
           label="Intervention date")

last_date = pred_dates[pred_dates >= intervention_ts][-1]
last_obs  = splits.holdout_df.loc[splits.holdout_df["ds"] == last_date, "y"].values
last_pred = br.predicted[pred_dates == last_date]
if len(last_obs) and len(last_pred):
    ax.annotate(
        f"Excess ~ {float(last_obs[0] - last_pred[0]):.1f}",
        xy=(last_date, float(last_pred[0])),
        xytext=(last_date - pd.Timedelta(days=90), float(last_pred[0]) + 6),
        arrowprops=dict(arrowstyle="->", color="black"),
        fontsize=9,
    )

ax.set_xlabel("Date")
ax.set_ylabel("y (daily outcome)")
ax.set_title(
    f"NeuralProphet counterfactual  |  Test RMSE: {result.metrics_test.rmse:.2f}"
    f"  |  Test MAPE: {result.metrics_test.mape:.1f}%",
    fontsize=10,
)
ax.legend(loc="upper left", fontsize=8)
ax.xaxis.set_major_formatter(mdates.DateFormatter("%Y"))
plt.tight_layout()
plt.savefig(OUT_DIR / "neuralprophet_counterfactual.png", dpi=150)
display(fig)
br = result.bootstrap_result
pred_dates = pd.to_datetime(br.dates)
intervention_ts = pd.Timestamp(INTERVENTION)

fig, ax = plt.subplots(figsize=(14, 5))

for part in [splits.train_df, splits.test_df, splits.holdout_df]:
    ax.plot(part["ds"], part["y"], color="#333333", linewidth=0.6, alpha=0.7)
ax.plot([], [], color="#333333", linewidth=0.6, alpha=0.7, label="Observed")

ax.plot(pred_dates, br.predicted, color="#B2182B", linewidth=1.4,
        label="Counterfactual (no-intervention)")
ax.fill_between(pred_dates, br.conf_lo, br.conf_hi,
                color="#B2182B", alpha=0.15, label="95% CI (MBB)")

ax.axvspan(intervention_ts, splits.holdout_df["ds"].max(),
           color="#FEE08B", alpha=0.25, label="Holdout (post-intervention)")
ax.axvline(intervention_ts, color="#4DAF4A", linestyle="--", linewidth=1.3,
           label="Intervention date")

last_date = pred_dates[pred_dates >= intervention_ts][-1]
last_obs  = splits.holdout_df.loc[splits.holdout_df["ds"] == last_date, "y"].values
last_pred = br.predicted[pred_dates == last_date]
if len(last_obs) and len(last_pred):
    ax.annotate(
        f"Excess ~ {float(last_obs[0] - last_pred[0]):.1f}",
        xy=(last_date, float(last_pred[0])),
        xytext=(last_date - pd.Timedelta(days=90), float(last_pred[0]) + 6),
        arrowprops=dict(arrowstyle="->", color="black"),
        fontsize=9,
    )

ax.set_xlabel("Date")
ax.set_ylabel("y (daily outcome)")
ax.set_title(
    f"NeuralProphet counterfactual  |  Test RMSE: {result.metrics_test.rmse:.2f}"
    f"  |  Test MAPE: {result.metrics_test.mape:.1f}%",
    fontsize=10,
)
ax.legend(loc="upper left", fontsize=8)
ax.xaxis.set_major_formatter(mdates.DateFormatter("%Y"))
plt.tight_layout()
plt.savefig(OUT_DIR / "neuralprophet_counterfactual.png", dpi=150)
display(fig)

Key takeaways¶

NeuralProphetModel.fit() trains a neural network with n_lags autoregressive terms and optional lagged regressors. Training is slower than statistical or tree-based models.
The AR warmup period: the first n_lags rows cannot produce a fitted value because there is no preceding history. These rows yield NaN in fit_result.fitted_values and fit_result.residuals.
Moving Block Bootstrap resamples only the non-NaN residuals. The effective residual pool is n_train_rows - n_lags.
run_single_its() orchestrates: load_config -> prepare_splits -> fit -> bootstrap -> metrics -> excess -> save.
Excess = observed - counterfactual_predicted. With a true +8/day effect over 42 days, total excess should land near 336 (noise aside).