State-of-the-Art Deep Learning scripts organized by models - easy to train and deploy with reproducible accuracy and performance on enterprise-grade infrastructure.
Describe the bug
When I run train.py on PyTorch/SpeechSynthesis/Tacotron2/, following error occured:
Traceback (most recent call last):
File "train.py", line 44, in <module>
import data_functions
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/data_functions.py", line 29, in <module>
from tacotron2.data_function import TextMelCollate
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/data_function.py", line 33, in <module>
from tacotron2.text import text_to_sequence
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/__init__.py", line 3, in <module>
from tacotron2.text import cleaners
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/cleaners.py", line 16, in <module>
from .numbers import normalize_numbers
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/numbers.py", line 3, in <module>
import inflect
File "/opt/conda/lib/python3.8/site-packages/inflect/__init__.py", line 77, in <module>
from pydantic.typing import Annotated
ImportError: cannot import name 'Annotated' from 'pydantic.typing' (/opt/conda/lib/python3.8/site-packages/pydantic/typing.py)
Traceback (most recent call last):
File "train.py", line 44, in <module>
import data_functions
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/data_functions.py", line 29, in <module>
from tacotron2.data_function import TextMelCollate
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/data_function.py", line 33, in <module>
from tacotron2.text import text_to_sequence
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/__init__.py", line 3, in <module>
from tacotron2.text import cleaners
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/cleaners.py", line 16, in <module>
from .numbers import normalize_numbers
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/numbers.py", line 3, in <module>
import inflect
File "/opt/conda/lib/python3.8/site-packages/inflect/__init__.py", line 77, in <module>
from pydantic.typing import Annotated
ImportError: cannot import name 'Annotated' from 'pydantic.typing' (/opt/conda/lib/python3.8/site-packages/pydantic/typing.py)
Traceback (most recent call last):
File "train.py", line 44, in <module>
import data_functions
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/data_functions.py", line 29, in <module>
from tacotron2.data_function import TextMelCollate
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/data_function.py", line 33, in <module>
from tacotron2.text import text_to_sequence
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/__init__.py", line 3, in <module>
from tacotron2.text import cleaners
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/cleaners.py", line 16, in <module>
from .numbers import normalize_numbers
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/numbers.py", line 3, in <module>
import inflect
File "/opt/conda/lib/python3.8/site-packages/inflect/__init__.py", line 77, in <module>
from pydantic.typing import Annotated
ImportError: cannot import name 'Annotated' from 'pydantic.typing' (/opt/conda/lib/python3.8/site-packages/pydantic/typing.py)
Traceback (most recent call last):
File "train.py", line 44, in <module>
import data_functions
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/data_functions.py", line 29, in <module>
from tacotron2.data_function import TextMelCollate
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/data_function.py", line 33, in <module>
from tacotron2.text import text_to_sequence
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/__init__.py", line 3, in <module>
from tacotron2.text import cleaners
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/cleaners.py", line 16, in <module>
from .numbers import normalize_numbers
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/numbers.py", line 3, in <module>
import inflect
File "/opt/conda/lib/python3.8/site-packages/inflect/__init__.py", line 77, in <module>
from pydantic.typing import Annotated
ImportError: cannot import name 'Annotated' from 'pydantic.typing' (/opt/conda/lib/python3.8/site-packages/pydantic/typing.py)
Traceback (most recent call last):
File "train.py", line 44, in <module>
import data_functions
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/data_functions.py", line 29, in <module>
from tacotron2.data_function import TextMelCollate
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/data_function.py", line 33, in <module>
from tacotron2.text import text_to_sequence
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/__init__.py", line 3, in <module>
from tacotron2.text import cleaners
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/cleaners.py", line 16, in <module>
from .numbers import normalize_numbers
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/numbers.py", line 3, in <module>
import inflect
File "/opt/conda/lib/python3.8/site-packages/inflect/__init__.py", line 77, in <module>
from pydantic.typing import Annotated
ImportError: cannot import name 'Annotated' from 'pydantic.typing' (/opt/conda/lib/python3.8/site-packages/pydantic/typing.py)
Traceback (most recent call last):
File "train.py", line 44, in <module>
import data_functions
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/data_functions.py", line 29, in <module>
from tacotron2.data_function import TextMelCollate
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/data_function.py", line 33, in <module>
from tacotron2.text import text_to_sequence
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/__init__.py", line 3, in <module>
from tacotron2.text import cleaners
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/cleaners.py", line 16, in <module>
from .numbers import normalize_numbers
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/numbers.py", line 3, in <module>
import inflect
File "/opt/conda/lib/python3.8/site-packages/inflect/__init__.py", line 77, in <module>
from pydantic.typing import Annotated
ImportError: cannot import name 'Annotated' from 'pydantic.typing' (/opt/conda/lib/python3.8/site-packages/pydantic/typing.py)
Traceback (most recent call last):
File "train.py", line 44, in <module>
import data_functions
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/data_functions.py", line 29, in <module>
from tacotron2.data_function import TextMelCollate
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/data_function.py", line 33, in <module>
from tacotron2.text import text_to_sequence
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/__init__.py", line 3, in <module>
from tacotron2.text import cleaners
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/cleaners.py", line 16, in <module>
from .numbers import normalize_numbers
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/numbers.py", line 3, in <module>
import inflect
File "/opt/conda/lib/python3.8/site-packages/inflect/__init__.py", line 77, in <module>
from pydantic.typing import Annotated
ImportError: cannot import name 'Annotated' from 'pydantic.typing' (/opt/conda/lib/python3.8/site-packages/pydantic/typing.py)
Traceback (most recent call last):
File "train.py", line 44, in <module>
import data_functions
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/data_functions.py", line 29, in <module>
from tacotron2.data_function import TextMelCollate
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/data_function.py", line 33, in <module>
from tacotron2.text import text_to_sequence
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/__init__.py", line 3, in <module>
from tacotron2.text import cleaners
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/cleaners.py", line 16, in <module>
from .numbers import normalize_numbers
File "/workspace/benchmark/SpeechSynthesis/Tacotron2/tacotron2/text/numbers.py", line 3, in <module>
import inflect
File "/opt/conda/lib/python3.8/site-packages/inflect/__init__.py", line 77, in <module>
from pydantic.typing import Annotated
ImportError: cannot import name 'Annotated' from 'pydantic.typing' (/opt/conda/lib/python3.8/site-packages/pydantic/typing.py)
To Reproduce
When I attempted to execute the following command in conjunction with these instructions, the aforementioned error occurred.
bash scripts/train_tacotron2.sh
Expected behaviortrain.py works without ImportError.
Environment
Container version (e.g. pytorch:19.05-py3): nvcr.io/nvidia/pytorch:20.06-py3
GPUs in the system: (e.g. 8x Tesla V100-SXM2-16GB): 8xA100 80GB SXM4
CUDA driver version (e.g. 418.67): 525.105.17
Based on my investigation, it seems that this error has emerged due to a lack of backward compatibility between Pydantic 2.0 and Inflect. The issue was highlighted in this GitHub discussion and the proposed solution was to use a specific version of pydantic (1.10.10). As a result, I was able to fix the error by adding pydantic==1.10.10 to the requirements.txt file, immediately after the inflect entry, like this:
However, to manage the version dependency more effectively, I suggest using a dependency management tool like Poetry, as it can be challenging to handle this solely through requirements.txt.
Related to Tacotron2/Pytorch
Describe the bug When I run
train.pyonPyTorch/SpeechSynthesis/Tacotron2/, following error occured:To Reproduce
When I attempted to execute the following command in conjunction with these instructions, the aforementioned error occurred.
Expected behavior
train.pyworks without ImportError.Environment
Based on my investigation, it seems that this error has emerged due to a lack of backward compatibility between Pydantic 2.0 and Inflect. The issue was highlighted in this GitHub discussion and the proposed solution was to use a specific version of pydantic (1.10.10). As a result, I was able to fix the error by adding pydantic==1.10.10 to the requirements.txt file, immediately after the inflect entry, like this:
However, to manage the version dependency more effectively, I suggest using a dependency management tool like Poetry, as it can be challenging to handle this solely through requirements.txt.