Validate that an input matches the PreparedInput schema.

Computes quantiles from a distribution specified by `samples` and their corresponding probability mass `sample_weights`. `samples` are first

Parameters ---------- inner_model : PreTrainedModel A hugging-face transformers PreTrainedModel, e.g., T5ForCondi

Construct a layernorm module in the T5 style. No bias and no subtraction of mean.

Generate batches of data for the Chronos-2 model. In training mode, this iterator is infinite. Yields ------ dict

See, https://github.com/faif/python-patterns.

Evaluate Chronos models. Parameters ---------- config_path : Path Path to the evaluation config. See ./configs/. metrics_path

Evaluate Chronos-2 models. Parameters ---------- config_path : Path Path to the evaluation config. See ./configs/. metrics_pa

Evaluate Chronos-Bolt models. Parameters ---------- config_path : Path Path to the evaluation config. See ./configs/. metrics

Fields that maybe modified during the fine-tuning stage.

Get encoder embeddings for the given time series. Parameters ---------- context Input series. This is ei

Get encoder embeddings for the given time series. Parameters ---------- inputs The time series to get em

Predict future sample tokens for the given token sequences. Arguments ``prediction_length``, ``num_samples``, ``temperature``,

Multi-head attention forward pass. Args: hidden_states : Input tensor of shape [batch_size, seq_len, d_model] mask :

Forward pass of the Chronos2 model. Parameters ---------- context Input tensor of shape (batch_size, context_leng

Load the model, either from a local path, S3 prefix, or from the HuggingFace Hub. Supports the same arguments as ``AutoConfig`` and `

Load the model, either from a local path S3 prefix or from the HuggingFace Hub. Supports the same arguments as ``AutoConfig`` and ``A

Load the model, either from a local path, S3 prefix or from the HuggingFace Hub. Supports the same arguments as ``AutoConfig`` and ``

Generate a synthetic time series from KernelSynth. Parameters ---------- max_kernels, optional The maximum number of base kernels

Check if the given entry has enough observations in the ``"target"`` attribute. Parameters ---------- entry The data entry (

Compute the aggregated relative score as reported in the Chronos paper. Results will be saved to {results_dir}/{model_name}-agg-rel-scores.cs

Get forecasts for the given time series. Predictions will be returned in fp32 on the cpu. Parameters ----------

Make predictions for evaluation on a fev.Task. Parameters ---------- task Benchmark task on which the ev

Refer to the base method (``BaseChronosPipeline.predict_quantiles``).

Applies a random binary operator (+ or *) with equal probability on kernels ``a`` and ``b``. Parameters ---------- a A G

Draw a sample from a GP prior. An efficient version that allows specification of the sampling method. The default sampling method used in Gau

(batch, seq_len, inner_dim) -> (batch, n_heads, seq_len, kv_proj_dim)