-
Notifications
You must be signed in to change notification settings - Fork 21.4k
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
[FSDP][4/N] Remove _get_fully_sharded_module_to_states
#104409
Conversation
[ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/104409
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 2660f4c: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
ghstack-source-id: d543bca70c50162e2e0566e973aceee805cbed60 Pull Request resolved: #104409
ghstack-source-id: d543bca70c50162e2e0566e973aceee805cbed60 Pull Request resolved: pytorch#104409
`_get_fully_sharded_module_to_states()` was used to emulate auto wrapping without actually calling `fully_shard`. Since we committed to unifying (see previous PR), we can remove this function and its helpers/tests. [ghstack-poisoned]
ghstack-source-id: 2cba28a2d666236440a33cf9902c8242dd395619 Pull Request resolved: #104409
ghstack-source-id: 2cba28a2d666236440a33cf9902c8242dd395619 Pull Request resolved: pytorch#104409
`_get_fully_sharded_module_to_states()` was used to emulate auto wrapping without actually calling `fully_shard`. Since we committed to unifying (see previous PR), we can remove this function and its helpers/tests. [ghstack-poisoned]
`_get_fully_sharded_module_to_states()` was used to emulate auto wrapping without actually calling `fully_shard`. Since we committed to unifying (see previous PR), we can remove this function and its helpers/tests. [ghstack-poisoned]
ghstack-source-id: c004d4449be8138b816d316dfb2d609e344c9dde Pull Request resolved: pytorch#104409
`_get_fully_sharded_module_to_states()` was used to emulate auto wrapping without actually calling `fully_shard`. Since we committed to unifying (see previous PR), we can remove this function and its helpers/tests. [ghstack-poisoned]
ghstack-source-id: 8396d44dcfe1e3d5c61f934e9802ce44e7f96a3f Pull Request resolved: pytorch#104409
`_get_fully_sharded_module_to_states()` was used to emulate auto wrapping without actually calling `fully_shard`. Since we committed to unifying (see previous PR), we can remove this function and its helpers/tests. [ghstack-poisoned]
ghstack-source-id: b81b73bc70ea7e1145db055a81bca7dbabcd5cd8 Pull Request resolved: pytorch#104409
`_get_fully_sharded_module_to_states()` was used to emulate auto wrapping without actually calling `fully_shard`. Since we committed to unifying (see previous PR), we can remove this function and its helpers/tests. [ghstack-poisoned]
`_get_fully_sharded_module_to_states()` was used to emulate auto wrapping without actually calling `fully_shard`. Since we committed to unifying (see previous PR), we can remove this function and its helpers/tests. [ghstack-poisoned]
ghstack-source-id: 5cccf8de845be6f8c6559c988d9d61b02fc7c287 Pull Request resolved: pytorch#104409
`_get_fully_sharded_module_to_states()` was used to emulate auto wrapping without actually calling `fully_shard`. Since we committed to unifying (see previous PR), we can remove this function and its helpers/tests. [ghstack-poisoned]
ghstack-source-id: 76d394ad6c81c8b1ae4456b74875880b9916edf8 Pull Request resolved: pytorch#104409
ghstack-source-id: 76d394ad6c81c8b1ae4456b74875880b9916edf8 Pull Request resolved: pytorch#104409
`_get_fully_sharded_module_to_states()` was used to emulate auto wrapping without actually calling `fully_shard`. Since we committed to unifying (see previous PR), we can remove this function and its helpers/tests. [ghstack-poisoned]
`_get_fully_sharded_module_to_states()` was used to emulate auto wrapping without actually calling `fully_shard`. Since we committed to unifying (see previous PR), we can remove this function and its helpers/tests. [ghstack-poisoned]
`_get_fully_sharded_module_to_states()` was used to emulate auto wrapping without actually calling `fully_shard`. Since we committed to unifying (see previous PR), we can remove this function and its helpers/tests. [ghstack-poisoned]
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Nice, I got confused by _get_fully_sharded_module_to_states
when it was introduced to the code.
`_get_fully_sharded_module_to_states()` was used to emulate auto wrapping without actually calling `fully_shard`. Since we committed to unifying (see previous PR), we can remove this function and its helpers/tests. [ghstack-poisoned]
`_get_fully_sharded_module_to_states()` was used to emulate auto wrapping without actually calling `fully_shard`. Since we committed to unifying (see previous PR), we can remove this function and its helpers/tests. [ghstack-poisoned]
`_get_fully_sharded_module_to_states()` was used to emulate auto wrapping without actually calling `fully_shard`. Since we committed to unifying (see previous PR), we can remove this function and its helpers/tests. [ghstack-poisoned]
Stack from ghstack (oldest at bottom):
ModuleWrapPolicy
#104427ignored_states
+ auto wrap (for now) #104418_get_fully_sharded_module_to_states
#104409fully_shard
auto wrap #104408_auto_wrap
forfully_shard
#104407ModuleWrapPolicy
to new path #104346_get_fully_sharded_module_to_states()
was used to emulate auto wrapping without actually callingfully_shard
. Since we committed to unifying (see previous PR), we can remove this function and its helpers/tests.