Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
Speech To Speech
Manage
Activity
Members
Labels
Plan
Issues
Issue boards
Milestones
Iterations
Wiki
Code
Merge requests
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Snippets
Locked files
Build
Pipelines
Jobs
Pipeline schedules
Artifacts
Deploy
Releases
Package registry
Container Registry
Model registry
Operate
Environments
Terraform modules
Monitor
Incidents
Service Desk
Analyze
Value stream analytics
Contributor analytics
CI/CD analytics
Repository analytics
Code review analytics
Issue analytics
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
Community forum
Contribute to GitLab
Provide feedback
Terms and privacy
Keyboard shortcuts
?
Snippets
Groups
Projects
Show more breadcrumbs
mirrored_repos
MachineLearning
huggingface
Speech To Speech
Commits
653b7ca5
Commit
653b7ca5
authored
7 months ago
by
Andres Marafioti
Browse files
Options
Downloads
Patches
Plain Diff
Assigning min new tokens to a compiled whisper graph on a thread breaks it
parent
fc9f9602
No related branches found
No related tags found
No related merge requests found
Changes
2
Hide whitespace changes
Inline
Side-by-side
Showing
2 changed files
STT/whisper_stt_handler.py
+1
-1
1 addition, 1 deletion
STT/whisper_stt_handler.py
arguments_classes/whisper_stt_arguments.py
+0
-6
0 additions, 6 deletions
arguments_classes/whisper_stt_arguments.py
with
1 addition
and
7 deletions
STT/whisper_stt_handler.py
+
1
−
1
View file @
653b7ca5
...
@@ -67,7 +67,7 @@ class WhisperSTTHandler(BaseHandler):
...
@@ -67,7 +67,7 @@ class WhisperSTTHandler(BaseHandler):
# generating more tokens than previously will trigger CUDA graphs capture
# generating more tokens than previously will trigger CUDA graphs capture
# one should warmup with a number of generated tokens above max tokens targeted for subsequent generation
# one should warmup with a number of generated tokens above max tokens targeted for subsequent generation
warmup_gen_kwargs
=
{
warmup_gen_kwargs
=
{
"
min_new_tokens
"
:
self
.
gen_kwargs
[
"
m
in
_new_tokens
"
],
"
min_new_tokens
"
:
self
.
gen_kwargs
[
"
m
ax
_new_tokens
"
],
# Yes, assign max_new_tokens to min_new_tokens
"
max_new_tokens
"
:
self
.
gen_kwargs
[
"
max_new_tokens
"
],
"
max_new_tokens
"
:
self
.
gen_kwargs
[
"
max_new_tokens
"
],
**
self
.
gen_kwargs
,
**
self
.
gen_kwargs
,
}
}
...
...
This diff is collapsed.
Click to expand it.
arguments_classes/whisper_stt_arguments.py
+
0
−
6
View file @
653b7ca5
...
@@ -33,12 +33,6 @@ class WhisperSTTHandlerArguments:
...
@@ -33,12 +33,6 @@ class WhisperSTTHandlerArguments:
"
help
"
:
"
The maximum number of new tokens to generate. Default is 128.
"
"
help
"
:
"
The maximum number of new tokens to generate. Default is 128.
"
},
},
)
)
stt_gen_min_new_tokens
:
int
=
field
(
default
=
0
,
metadata
=
{
"
help
"
:
"
The minimum number of new tokens to generate. Default is 0.
"
},
)
stt_gen_num_beams
:
int
=
field
(
stt_gen_num_beams
:
int
=
field
(
default
=
1
,
default
=
1
,
metadata
=
{
metadata
=
{
...
...
This diff is collapsed.
Click to expand it.
Preview
0%
Loading
Try again
or
attach a new file
.
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Save comment
Cancel
Please
register
or
sign in
to comment