TypeError: TimesFmBase.__init__() got an unexpected keyword argument 'context_len'

[thks-for-the-help]

import pandas as pd
import numpy as np
import timesfm as tf

# Carica il file CSV
df = pd.read_csv("file.csv", parse_dates=["Datetime"])
df = df.set_index("Datetime")

# Configurazione del modello
tfm = tf.TimesFm(
    context_len=128,  # Contesto massimo (modificabile)
    horizon_len=5,    # Orizzonte di previsione
    input_patch_len=32,
    output_patch_len=128,
    num_layers=20,
    model_dims=1280,
    backend="gpu",    # Cambia in "cpu" se necessario
)

I receive back:

TypeError: TimesFmBase.init() got an unexpected keyword argument 'context_len'

despite is the standard instruction were officially given . Thanks

[rajatsen91]

WE have moved to using TimesFM Hparams. Please see https://github.com/google-research/timesfm?tab=readme-ov-file#usage

[catsled]

hparams=timesfm.TimesFmHparams(
^^^^^^^^^^^^^^^^^^^^^^^
TypeError: TimesFmHparams.init() got an unexpected keyword argument 'use_positional_embedding'

[l1k3l]

hparams=timesfm.TimesFmHparams( ^^^^^^^^^^^^^^^^^^^^^^^ TypeError: TimesFmHparams.init() got an unexpected keyword argument 'use_positional_embedding'

I have the same issue🙋🏽‍♂️