Add generation params to eval script (#138)

*Description of changes:* Adds generation params to command line options
for the evaluation script.


By submitting this pull request, I confirm that you can use, modify,
copy, and redistribute this contribution, under the terms of your
choice.

scripts/evaluation/evaluate.py

@@ -1,6 +1,6 @@
 import logging
 from pathlib import Path
-from typing import Iterable
+from typing import Iterable, Optional
 
 import datasets
 import numpy as np
@@ -232,8 +232,9 @@ def generate_sample_forecasts(
     test_data_input: Iterable,
     pipeline: ChronosPipeline,
     prediction_length: int,
-    num_samples: int,
     batch_size: int,
+    num_samples: int,
+    **predict_kwargs,
 ):
     # Generate forecast samples
     forecast_samples = []
@@ -244,6 +245,7 @@ def generate_sample_forecasts(
                 context,
                 prediction_length=prediction_length,
                 num_samples=num_samples,
+                **predict_kwargs,
             ).numpy()
         )
     forecast_samples = np.concatenate(forecast_samples)
@@ -268,6 +270,9 @@ def main(
     torch_dtype: str = "bfloat16",
     batch_size: int = 32,
     num_samples: int = 20,
+    temperature: Optional[float] = None,
+    top_k: Optional[int] = None,
+    top_p: Optional[float] = None,
 ):
     if isinstance(torch_dtype, str):
         torch_dtype = getattr(torch, torch_dtype)
@@ -277,7 +282,7 @@ def main(
     pipeline = ChronosPipeline.from_pretrained(
         chronos_model_id,
         device_map=device,
-        torch_dtype=torch.bfloat16,
+        torch_dtype=torch_dtype,
     )
 
     # Load backtest configs
@@ -300,8 +305,11 @@ def main(
             test_data.input,
             pipeline=pipeline,
             prediction_length=prediction_length,
-            num_samples=num_samples,
             batch_size=batch_size,
+            num_samples=num_samples,
+            temperature=temperature,
+            top_k=top_k,
+            top_p=top_p,
         )
 
         logger.info(f"Evaluating forecasts for {dataset_name}")