Fix TimesFm doc issue (#37552)

* fix doc

* code block

src/transformers/models/timesfm/modeling_timesfm.py

@@ -33,7 +33,6 @@ from ...modeling_outputs import BaseModelOutput
 from ...modeling_utils import ALL_ATTENTION_FUNCTIONS, PreTrainedModel
 from ...processing_utils import Unpack
 from ...utils import (
-    add_code_sample_docstrings,
     add_start_docstrings,
     add_start_docstrings_to_model_forward,
     can_return_tuple,
@@ -44,8 +43,6 @@ from .configuration_timesfm import TimesFmConfig
 
 
 logger = logging.get_logger(__name__)
-
-_CHECKPOINT_FOR_DOC = "google/timesfm-2.0-500m-pytorch"
 _CONFIG_FOR_DOC = "TimesFmConfig"
 
 
@@ -734,11 +731,6 @@ class TimesFmModelForPrediction(TimesFmPreTrainedModel):
     @can_return_tuple
     @add_start_docstrings_to_model_forward(TIMESFM_INPUTS_DOCSTRING)
     @replace_return_docstrings(output_type=TimesFmOutputForPrediction, config_class=_CONFIG_FOR_DOC)
-    @add_code_sample_docstrings(
-        checkpoint=_CHECKPOINT_FOR_DOC,
-        output_type=TimesFmOutputForPrediction,
-        config_class=_CONFIG_FOR_DOC,
-    )
     def forward(
         self,
         past_values: Sequence[torch.Tensor],
@@ -752,28 +744,40 @@ class TimesFmModelForPrediction(TimesFmPreTrainedModel):
         output_hidden_states: Optional[bool] = None,
     ) -> TimesFmOutputForPrediction:
         r"""
-        window_size (`int`, *optional*):
+            window_size (`int`, *optional*):
-            Window size of trend + residual decomposition. If None then we do not do decomposition.
+                Window size of trend + residual decomposition. If None then we do not do decomposition.
-        future_values (`torch.Tensor`, *optional*):
+            future_values (`torch.Tensor`, *optional*):
-            Optional future time series values to be used for loss computation.
+                Optional future time series values to be used for loss computation.
-        forecast_context_len (`int`, *optional*):
+            forecast_context_len (`int`, *optional*):
-            Optional max context length.
+                Optional max context length.
-        return_forecast_on_context (`bool`, *optional*):
+            return_forecast_on_context (`bool`, *optional*):
-            True to return the forecast on the context when available, i.e. after the first input patch.
+                True to return the forecast on the context when available, i.e. after the first input patch.
-        truncate_negative (`bool`, *optional*):
+            truncate_negative (`bool`, *optional*):
-            Truncate to only non-negative values if any of the contexts have non-negative values,
+                Truncate to only non-negative values if any of the contexts have non-negative values,
-            otherwise do nothing.
+                otherwise do nothing.
-        output_attentions (`bool`, *optional*):
+            output_attentions (`bool`, *optional*):
-            Whether to output the attentions.
+                Whether to output the attentions.
-        output_hidden_states (`bool`, *optional*):
+            output_hidden_states (`bool`, *optional*):
-            Whether to output the hidden states.
+                Whether to output the hidden states.
 
         Returns:
-            A TimesFmOutputForPrediction object or a tuple containing:
+
-                - the mean forecast of size (# past_values, # forecast horizon),
+        Example:
-                - the full forecast (mean + quantiles) of size
+
-                    (# past_values,  # forecast horizon, 1 + # quantiles).
+        ```python
-                - loss: the mean squared error loss + quantile loss if `future_values` is provided.
+        >>> from transformers import TimesFmModelForPrediction
+
+        >>> model = TimesFmModelForPrediction.from_pretrained("google/timesfm-2.0-500m-pytorch")
+
+        >>> forecast_input = [torch.linspace(0, 20, 100).sin(), torch.linspace(0, 20, 200).sin(), torch.linspace(0, 20, 400).sin()]
+        >>> frequency_input = torch.tensor([0, 1, 2], dtype=torch.long)
+
+        >>> # Generate
+        >>> with torch.no_grad():
+        >>>     outputs = model(past_values=forecast_input, freq=frequency_input, return_dict=True)
+        >>>     point_forecast_conv = outputs.mean_predictions
+        >>>     quantile_forecast_conv = outputs.full_predictions
+        ```
         """
         if forecast_context_len is None:
             fcontext_len = self.context_len

src/transformers/models/timesfm/modular_timesfm.py

@@ -27,7 +27,6 @@ from ...modeling_outputs import BaseModelOutput
 from ...modeling_utils import ALL_ATTENTION_FUNCTIONS, PreTrainedModel
 from ...processing_utils import Unpack
 from ...utils import (
-    add_code_sample_docstrings,
     add_start_docstrings,
     add_start_docstrings_to_model_forward,
     can_return_tuple,
@@ -690,11 +689,6 @@ class TimesFmModelForPrediction(TimesFmPreTrainedModel):
     @can_return_tuple
     @add_start_docstrings_to_model_forward(TIMESFM_INPUTS_DOCSTRING)
     @replace_return_docstrings(output_type=TimesFmOutputForPrediction, config_class=_CONFIG_FOR_DOC)
-    @add_code_sample_docstrings(
-        checkpoint=_CHECKPOINT_FOR_DOC,
-        output_type=TimesFmOutputForPrediction,
-        config_class=_CONFIG_FOR_DOC,
-    )
     def forward(
         self,
         past_values: Sequence[torch.Tensor],
@@ -708,28 +702,40 @@ class TimesFmModelForPrediction(TimesFmPreTrainedModel):
         output_hidden_states: Optional[bool] = None,
     ) -> TimesFmOutputForPrediction:
         r"""
-        window_size (`int`, *optional*):
+            window_size (`int`, *optional*):
-            Window size of trend + residual decomposition. If None then we do not do decomposition.
+                Window size of trend + residual decomposition. If None then we do not do decomposition.
-        future_values (`torch.Tensor`, *optional*):
+            future_values (`torch.Tensor`, *optional*):
-            Optional future time series values to be used for loss computation.
+                Optional future time series values to be used for loss computation.
-        forecast_context_len (`int`, *optional*):
+            forecast_context_len (`int`, *optional*):
-            Optional max context length.
+                Optional max context length.
-        return_forecast_on_context (`bool`, *optional*):
+            return_forecast_on_context (`bool`, *optional*):
-            True to return the forecast on the context when available, i.e. after the first input patch.
+                True to return the forecast on the context when available, i.e. after the first input patch.
-        truncate_negative (`bool`, *optional*):
+            truncate_negative (`bool`, *optional*):
-            Truncate to only non-negative values if any of the contexts have non-negative values,
+                Truncate to only non-negative values if any of the contexts have non-negative values,
-            otherwise do nothing.
+                otherwise do nothing.
-        output_attentions (`bool`, *optional*):
+            output_attentions (`bool`, *optional*):
-            Whether to output the attentions.
+                Whether to output the attentions.
-        output_hidden_states (`bool`, *optional*):
+            output_hidden_states (`bool`, *optional*):
-            Whether to output the hidden states.
+                Whether to output the hidden states.
 
         Returns:
-            A TimesFmOutputForPrediction object or a tuple containing:
+
-                - the mean forecast of size (# past_values, # forecast horizon),
+        Example:
-                - the full forecast (mean + quantiles) of size
+
-                    (# past_values,  # forecast horizon, 1 + # quantiles).
+        ```python
-                - loss: the mean squared error loss + quantile loss if `future_values` is provided.
+        >>> from transformers import TimesFmModelForPrediction
+
+        >>> model = TimesFmModelForPrediction.from_pretrained("google/timesfm-2.0-500m-pytorch")
+
+        >>> forecast_input = [torch.linspace(0, 20, 100).sin(), torch.linspace(0, 20, 200).sin(), torch.linspace(0, 20, 400).sin()]
+        >>> frequency_input = torch.tensor([0, 1, 2], dtype=torch.long)
+
+        >>> # Generate
+        >>> with torch.no_grad():
+        >>>     outputs = model(past_values=forecast_input, freq=frequency_input, return_dict=True)
+        >>>     point_forecast_conv = outputs.mean_predictions
+        >>>     quantile_forecast_conv = outputs.full_predictions
+        ```
         """
         if forecast_context_len is None:
             fcontext_len = self.context_len