-
Notifications
You must be signed in to change notification settings - Fork 592
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Training notebook - File not found erroe #721
Comments
update, looks like I was saving the file with BOM. Removed that but still getting: Traceback (most recent call last): and doesnt APPEAR to have any progress being made. |
OKAY - so not enough samples it seems. However when I try to start training I am getting the below: ImportError: cannot import name '_compare_version' from 'torchmetrics.utilities.imports' (/usr/local/lib/python3.11/dist-packages/torchmetrics/utilities/imports.py) |
From what I can tell, this stems from the install section. There are several packages that are no longer available: ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts. Updating Numpy version fixes some of these. As well as torch. But doing so breaks....something in the training step: Updated Libraries code: clone:!git clone -q https://github.com/rmcpantoja/piper fixing recent compativility isswes:!pip install -q torchaudio==2.6.0 torchmetrics==0.11.4 faster_whisper Useful vars:use_whisper = True Output from updated step 2: |
Noting that it was updated 2 Years ago. Hope this can be addressed :) |
I tried with many different versions of torch, torchtext, torchaudio and a few other libraries, but nothing worked. Let me knowif you foud any fix. |
Running the training notebook, extracting files goes fine. I see them in the folders. However when I run the preprocessor, I see its not picking up the files. Checking my CSV file:
wavs/1.wav|Transcript 1.
wavs/2.wav|Transscript 2.
wavs/3.wav|Transscriopt 3.
Error:
/content/piper/src/python
WARNING:preprocess:Missing wavs/1.wav
INFO:preprocess:Single speaker dataset
INFO:preprocess:Wrote dataset config
INFO:preprocess:Processing 2 utterance(s) with 2 worker(s)
Traceback (most recent call last):
File "", line 198, in _run_module_as_main
File "", line 88, in _run_code
File "/content/piper/src/python/piper_train/preprocess.py", line 502, in
main()
File "/content/piper/src/python/piper_train/preprocess.py", line 225, in main
for utt_batch in batched(
File "/content/piper/src/python/piper_train/preprocess.py", line 491, in batched
raise ValueError("n must be at least one")
ValueError: n must be at least one
It looks like its still running:
Executing (5m 38s)
<cell line: 0>
navigate_next
system()
navigate_next
_system_compat()
navigate_next
_run_command()
navigate_next
_monitor_process()
navigate_next
_poll_process()
Not sure if its really doing anything because I do not see the GPU being used.
Any help would be appreciated
The text was updated successfully, but these errors were encountered: