Performance on Seasonal Intermittent Demand Data

[Arsa-Nik]

I am using TimesFM on monthly demand data (specifically seasonal intermittent), and I wonder whether my current parameter setup is correct and if you have further insights on how to improve its performance—I'm in the process of fine-tuning it next.
P.S. I'm using v1.2.1, and it seems _median_index is not initialized in this version so I manually initialized it below.

data_1 = {
    'unique_id': ['ID_1'] * 52,  
    'ds': pd.date_range(start='2018-12-01', periods=52, freq='MS').tolist(),
    'values': [1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 2.0, 4.0, 46.0, 395.0,
               0.0, 0.0, 5.0, 2.0, 2.0, 0.0, 0.0, 3.0, 7.0, 23.0, 84.0,
               494.0, 177.0, 387.0, 8.0, 2.0, 5.0, 1.0, 0.0, 5.0, 10.0,
               33.0, 86.0, 2213.0, 1331.0, 141.0, 19.0, 5.0, 1.0, 5.0,
               7.0, 6.0, 7.0, 17.0, 202.0, 3240.0, 2178.0, 8.0, 2.0, 1.0]
}
train_data = pd.DataFrame(data_1)
tfm = timesfm.TimesFm(
      hparams=timesfm.TimesFmHparams(
          backend="gpu",
          per_core_batch_size=32,
          horizon_len=18,
      ),
      checkpoint=timesfm.TimesFmCheckpoint(
          huggingface_repo_id="google/timesfm-1.0-200m"),
  )
tfm._median_index = -1 
forecast = tfm.forecast_on_df(
        inputs=train_data,
        freq='MS',
        num_jobs=-1, 
    )
    
plt.figure(figsize=(14, 8))
plt.plot(train_data['ds'], train_data['values'], label='Actuals', marker='o', linestyle='-', color='blue')
plt.plot(forecast['ds'], forecast['timesfm'], label='Forecast', marker='o', linestyle='--', color='orange')
plt.fill_between(forecast['ds'], forecast['timesfm-q-0.1'], forecast['timesfm-q-0.9'], color='orange', alpha=0.2, label='90% Confidence Interval')
plt.title('Actuals and Forecasts with Prediction Intervals', fontsize=16)
plt.xlabel('Date', fontsize=14)
plt.ylabel('Values', fontsize=14)
plt.legend(fontsize=12)
plt.grid(True)
plt.tight_layout()
plt.show()