Fix needed: AttributeError: 'StaticCache' object has no attribute 'max_batch_size'. Did you mean: 'batch_size'?

[textbc1]

Hi,
Error occured in 2nd iteraton of generate() (2nd iteration appears due to need for accessing Cache is more from 2nd)

Exact error: AttributeError: 'StaticCache' object has no attribute 'max_batch_size'. Did you mean: 'batch_size'?

Error shown in the terminal:
generation = model.generate(input_ids=input_ids, prompt_input_ids=prompt_input_ids, attention_mask=attention_mask, prompt_attention_mask=prompt_attention_mask,
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:_Develop_dev\px3.pixi\envs\default\Lib\site-packages\torch\utils_contextlib.py", line 116, in decorate_context
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "D:_Develop_dev\px3.pixi\envs\default\Lib\site-packages\parler_tts\modeling_parler_tts.py", line 3494, in generate
model_kwargs["past_key_values"] = self._get_cache(
^^^^^^^^^^^^^^^^
File "D:_Develop_dev\px3.pixi\envs\default\Lib\site-packages\parler_tts\modeling_parler_tts.py", line 3278, in _get_cache
or cache_to_check.max_batch_size != max_batch_size
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:_Develop_dev\px3.pixi\envs\default\Lib\site-packages\torch\nn\modules\module.py", line 1928, in getattr
raise AttributeError(
AttributeError: 'StaticCache' object has no attribute 'max_batch_size'. Did you mean: 'batch_size'?

Flag set in the code before any call to generate() was made:
model.generation_config.cache_implementation = "static"

Could you please help by giving a fix, if it only parler-tts code, even source code fix is ok if exact changes are provided will apply locally and build

Thank you in advance

[textbc1]

Locally in the file \site-packages\parler_tts\modeling_parler_tts.py
I checked with logging and could find batch_size (but not max_batch_size) attribute.

I made the change (near line 3278) and error disappeared , hope this is right solution:

    need_new_cache = (
        not hasattr(self, "_cache")
        or (not isinstance(cache_to_check, cache_cls))
        or cache_to_check.batch_size != max_batch_size   <-- changed line (renamed cache_to_check.max_batch_size lhs of comparison to cache_to_check.batch.size)
        or cache_to_check.max_cache_len < max_cache_len
    )