Skip to content
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.

Sign up for GitHub

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

Add max_seq_len to pipeline component input #3183

Merged
merged 4 commits into from
Jul 22, 2024
Merged

Add max_seq_len to pipeline component input #3183

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
5fb5e69
Add max_seq_len to pipeline component input
aggarwal-k Jul 22, 2024
d393a6c
Update default max_seq_length
aggarwal-k Jul 22, 2024
c5ce5c4
Update assets/training/finetune_acft_hf_nlp/components_v2/pipeline_co…
srsaggam Jul 22, 2024
70d9be0
Update component versions
aggarwal-k Jul 22, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion assets/training/finetune_acft_hf_nlp/components_v2/data_import/common/spec.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
$schema: https://azuremlschemas.azureedge.net/latest/commandComponent.schema.json
name: oss_text_generation_data_import
version: 0.0.19
version: 0.0.20
type: command

is_deterministic: True
Expand Down
2 changes: 1 addition & 1 deletion assets/training/finetune_acft_hf_nlp/components_v2/finetune/chat_completion/spec.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
$schema: https://azuremlschemas.azureedge.net/latest/commandComponent.schema.json
name: oss_chat_completion_finetune
version: 0.0.19
version: 0.0.20
type: command

is_deterministic: True
Expand Down
2 changes: 1 addition & 1 deletion assets/training/finetune_acft_hf_nlp/components_v2/finetune/text_generation/spec.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
$schema: https://azuremlschemas.azureedge.net/latest/commandComponent.schema.json
name: oss_text_generation_finetune
version: 0.0.19
version: 0.0.20
type: command

is_deterministic: True
Expand Down
10 changes: 8 additions & 2 deletions ...training/finetune_acft_hf_nlp/components_v2/pipeline_components/chat_completion/spec.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
$schema: https://azuremlschemas.azureedge.net/latest/pipelineComponent.schema.json
name: oss_chat_completion_pipeline
version: 0.0.19
version: 0.0.20
type: pipeline
display_name: OSS Chat Completion Pipeline
description: FTaaS Pipeline component for chat completion
Expand Down Expand Up @@ -54,6 +54,12 @@ inputs:

# Finetuning parameters
# Training parameters
max_seq_length:
type: integer
optional: true
default: 8192
description: Default is -1 which means the padding is done up to the model's max length. Else will be padded to `max_seq_length`.
srsaggam marked this conversation as resolved.
Show resolved Hide resolved

num_train_epochs:
type: integer
default: 1
Expand Down Expand Up @@ -153,7 +159,7 @@ jobs:
dataset_input: '${{parent.jobs.oss_text_generation_data_import.outputs.output_dataset}}'
batch_size: 1000
pad_to_max_length: "false"
max_seq_length: 8192
max_seq_length: '${{parent.inputs.max_seq_length}}'
number_of_gpu_to_use_finetuning: '${{parent.inputs.number_of_gpu_to_use_finetuning}}'
apply_lora: "true"
lora_alpha: 128
Expand Down
10 changes: 8 additions & 2 deletions ...training/finetune_acft_hf_nlp/components_v2/pipeline_components/text_generation/spec.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
$schema: https://azuremlschemas.azureedge.net/latest/pipelineComponent.schema.json
name: oss_text_generation_pipeline
version: 0.0.19
version: 0.0.20
type: pipeline
display_name: OSS Text Generation Pipeline
description: FTaaS Pipeline component for text generation
Expand Down Expand Up @@ -69,6 +69,12 @@ inputs:

# Finetuning parameters
# Training parameters
max_seq_length:
type: integer
optional: true
default: 8192
description: Default is -1 which means the padding is done up to the model's max length. Else will be padded to `max_seq_length`.

num_train_epochs:
type: integer
default: 1
Expand Down Expand Up @@ -170,7 +176,7 @@ jobs:
ground_truth_key: '${{parent.inputs.ground_truth_key}}'
batch_size: 1000
pad_to_max_length: "false"
max_seq_length: 4096
max_seq_length: '${{parent.inputs.max_seq_length}}'
number_of_gpu_to_use_finetuning: '${{parent.inputs.number_of_gpu_to_use_finetuning}}'
apply_lora: "true"
lora_alpha: 128
Expand Down
Loading