[QEff. Finetune] Adding the support to resume the fine tuning using pre computed #233
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
… prev run whoch would have stopped in between. There's no necessity to pass tokenizer_name if a model_name is passed. It will take the same name as model_name by default. If a different tokenizer_name is required than the model_name, then it can be passed separately as an argument. Signed-off-by: Swati Allabadi <quic_sallabad@quicinc.com>
quic-mamta
requested changes
Jan 23, 2025
…ers into finetune Signed-off-by: Swati Allabadi <quic_sallabad@quicinc.com>
quic-swatia
force-pushed
the
finetune
branch
2 times, most recently
from
0dfaa2f to
da18e86
Compare
Signed-off-by: Swati Allabadi <quic_sallabad@quicinc.com>
quic-swatia
force-pushed
the
finetune
branch
from
0d4affa to
6e729fe
Compare
…ts and check for loss convergence. Signed-off-by: Swati Allabadi <quic_sallabad@quicinc.com>
Contributor
|
If we don't change output_dir then after resuming FT, will new tensorboard data be appended to previous tensorboard log files? |
Contributor
Author
|
Irrespective of the value of the output_dir, the tensorboard files get saved inside directory named runs. For each fine tuning job, a new directory is created inside runs. So, if we run the following command : "tensorboard --logdir runs --bind_all", tensorboard data from both the jobs will show up together in a single plot. |
quic-mamta
approved these changes
Feb 25, 2025
vbaddi
approved these changes
Mar 18, 2025
quic-swatia
pushed a commit
that referenced
this pull request
Mar 20, 2025
…re computed (#233) 1) Adding the support to resume the fine tuning using checkpoints from a prev run which would have stopped in between. 2) Checkpoints, both intermediate and for complete epoch, will get saved for each epoch through these changes. 3) There's no necessity to pass tokenizer_name if a model_name is passed. It will take the same name as model_name by default. If a different tokenizer_name is required than the model_name, then it can be passed separately as an argument in the command. --------- Signed-off-by: Swati Allabadi <quic_swatia@quicinc.com> Co-authored-by: Swati Allabadi <quic-swatia@quicinc.com>
quic-rishinr
pushed a commit
to quic-rishinr/efficient-transformers
that referenced
this pull request
Mar 21, 2025
…re computed (quic#233) 1) Adding the support to resume the fine tuning using checkpoints from a prev run which would have stopped in between. 2) Checkpoints, both intermediate and for complete epoch, will get saved for each epoch through these changes. 3) There's no necessity to pass tokenizer_name if a model_name is passed. It will take the same name as model_name by default. If a different tokenizer_name is required than the model_name, then it can be passed separately as an argument in the command. --------- Signed-off-by: Swati Allabadi <quic_sallabad@quicinc.com> Co-authored-by: Swati Allabadi <quic-swatia@quicinc.com> Signed-off-by: Rishin Raj <quic_rishinr@quicinc.com>
qcdipankar
pushed a commit
to qcdipankar/efficient-transformers
that referenced
this pull request
Apr 1, 2025
…re computed (quic#233) 1) Adding the support to resume the fine tuning using checkpoints from a prev run which would have stopped in between. 2) Checkpoints, both intermediate and for complete epoch, will get saved for each epoch through these changes. 3) There's no necessity to pass tokenizer_name if a model_name is passed. It will take the same name as model_name by default. If a different tokenizer_name is required than the model_name, then it can be passed separately as an argument in the command. --------- Signed-off-by: Swati Allabadi <quic_sallabad@quicinc.com> Co-authored-by: Swati Allabadi <quic-swatia@quicinc.com> Signed-off-by: Dipankar Sarkar <quic_dipankar@quicinc.com>
eplatero97
pushed a commit
to eplatero97/efficient-transformers
that referenced
this pull request
Apr 29, 2025
…re computed (quic#233) 1) Adding the support to resume the fine tuning using checkpoints from a prev run which would have stopped in between. 2) Checkpoints, both intermediate and for complete epoch, will get saved for each epoch through these changes. 3) There's no necessity to pass tokenizer_name if a model_name is passed. It will take the same name as model_name by default. If a different tokenizer_name is required than the model_name, then it can be passed separately as an argument in the command. --------- Signed-off-by: Swati Allabadi <quic_swatia@quicinc.com> Co-authored-by: Swati Allabadi <quic-swatia@quicinc.com>
eplatero97
pushed a commit
to eplatero97/efficient-transformers
that referenced
this pull request
Apr 29, 2025
…re computed (quic#233) 1) Adding the support to resume the fine tuning using checkpoints from a prev run which would have stopped in between. 2) Checkpoints, both intermediate and for complete epoch, will get saved for each epoch through these changes. 3) There's no necessity to pass tokenizer_name if a model_name is passed. It will take the same name as model_name by default. If a different tokenizer_name is required than the model_name, then it can be passed separately as an argument in the command. --------- Signed-off-by: Swati Allabadi <quic_swatia@quicinc.com> Co-authored-by: Swati Allabadi <quic-swatia@quicinc.com>
eplatero97
pushed a commit
to eplatero97/efficient-transformers
that referenced
this pull request
Apr 29, 2025
…re computed (quic#233) 1) Adding the support to resume the fine tuning using checkpoints from a prev run which would have stopped in between. 2) Checkpoints, both intermediate and for complete epoch, will get saved for each epoch through these changes. 3) There's no necessity to pass tokenizer_name if a model_name is passed. It will take the same name as model_name by default. If a different tokenizer_name is required than the model_name, then it can be passed separately as an argument in the command. --------- Signed-off-by: Swati Allabadi <quic_swatia@quicinc.com> Co-authored-by: Swati Allabadi <quic-swatia@quicinc.com> Signed-off-by: eplatero <quic_eplatero@quicinc.com>
eplatero97
pushed a commit
to eplatero97/efficient-transformers
that referenced
this pull request
Apr 29, 2025
…re computed (quic#233) 1) Adding the support to resume the fine tuning using checkpoints from a prev run which would have stopped in between. 2) Checkpoints, both intermediate and for complete epoch, will get saved for each epoch through these changes. 3) There's no necessity to pass tokenizer_name if a model_name is passed. It will take the same name as model_name by default. If a different tokenizer_name is required than the model_name, then it can be passed separately as an argument in the command. --------- Signed-off-by: Swati Allabadi <quic_swatia@quicinc.com> Co-authored-by: Swati Allabadi <quic-swatia@quicinc.com> Signed-off-by: eplatero <quic_eplatero@quicinc.com>
eplatero97
pushed a commit
to eplatero97/efficient-transformers
that referenced
this pull request
Apr 29, 2025
…re computed (quic#233) 1) Adding the support to resume the fine tuning using checkpoints from a prev run which would have stopped in between. 2) Checkpoints, both intermediate and for complete epoch, will get saved for each epoch through these changes. 3) There's no necessity to pass tokenizer_name if a model_name is passed. It will take the same name as model_name by default. If a different tokenizer_name is required than the model_name, then it can be passed separately as an argument in the command. --------- Signed-off-by: Swati Allabadi <quic_swatia@quicinc.com> Co-authored-by: Swati Allabadi <quic-swatia@quicinc.com> Signed-off-by: eplatero <quic_eplatero@quicinc.com>
eplatero97
pushed a commit
to eplatero97/efficient-transformers
that referenced
this pull request
Apr 29, 2025
…re computed (quic#233) 1) Adding the support to resume the fine tuning using checkpoints from a prev run which would have stopped in between. 2) Checkpoints, both intermediate and for complete epoch, will get saved for each epoch through these changes. 3) There's no necessity to pass tokenizer_name if a model_name is passed. It will take the same name as model_name by default. If a different tokenizer_name is required than the model_name, then it can be passed separately as an argument in the command. --------- Signed-off-by: Swati Allabadi <quic_swatia@quicinc.com> Co-authored-by: Swati Allabadi <quic-swatia@quicinc.com> Signed-off-by: eplatero <quic_eplatero@quicinc.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
If a different tokenizer_name is required than the model_name, then it can be passed separately as an argument in the command.