Skip to content

Commit

Permalink
reorder and fix path
Browse files Browse the repository at this point in the history
Signed-off-by: Chen Cui <chcui@nvidia.com>
  • Loading branch information
cuichenx committed Sep 28, 2024
1 parent 60eb0f7 commit 40ae724
Showing 1 changed file with 47 additions and 47 deletions.
94 changes: 47 additions & 47 deletions .github/workflows/cicd-main.yml
Original file line number Diff line number Diff line change
Expand Up @@ -5188,7 +5188,18 @@ jobs:
AFTER_SCRIPT: |
rm -rf tests/collections/llm/gpt_pretrain_results
rm -rf tests/collections/llm/gpt_index_mappings
L2_NeMo_2_HF_MODEL_IMPORT:
needs: [cicd-test-container-setup]
uses: ./.github/workflows/_test_template.yml
with:
RUNNER: self-hosted-azure
SCRIPT: |
python tests/collections/llm/gpt/model/test_model_import.py
AFTER_SCRIPT: |
rm -rf ~/.cache/nemo/models
L2_NeMo_2_SSM_Pretraining:
needs: [cicd-test-container-setup]
uses: ./.github/workflows/_test_template.yml
Expand Down Expand Up @@ -5223,6 +5234,24 @@ jobs:
AFTER_SCRIPT: |
rm -rf /home/TestData/nlp/megatron_mamba/nemo-ux-mamba/cicd_test_sft/${{ github.run_id }}
L2_NeMo_2_T5_Pretraining:
needs: [cicd-test-container-setup]
uses: ./.github/workflows/_test_template.yml
if: contains(fromJSON(needs.cicd-test-container-setup.outputs.test_to_run), 'L2_NeMo_2_T5_Pretraining') || needs.cicd-test-container-setup.outputs.all == 'true'
with:
RUNNER: self-hosted-azure
SCRIPT: |
NVTE_FUSED_ATTN=0 NVTE_FLASH_ATTN=0 python tests/collections/llm/megatron_t5_pretraining.py \
--devices=2 \
--max-steps=3 \
--experiment-dir=tests/collections/llm/t5_pretrain_results \
--data-path=/home/TestData/nlp/megatron_t5/data/pile_val_small_bert_tokenizer_text_document \
--index-mapping-dir=tests/collections/llm/t5_index_mappings
AFTER_SCRIPT: |
rm -rf tests/collections/llm/t5_pretrain_results
rm -rf tests/collections/llm/t5_index_mappings
L2_NeMo_2_GPT_SFT_TP1PP1_MBS1:
needs: [cicd-test-container-setup]
uses: ./.github/workflows/_test_template.yml
Expand All @@ -5232,7 +5261,7 @@ jobs:
SCRIPT: |
python tests/collections/llm/gpt_finetuning.py \
--restore_path /TestData/nemo2_ckpt/llama_68M \
--restore_path /home/TestData/nemo2_ckpt/llama_68M \
--devices 2 \
--max_steps 3 \
--experiment_dir tests/collections/llm/${{ github.run_id }} \
Expand All @@ -5242,7 +5271,7 @@ jobs:
--mbs 1
python tests/collections/llm/gpt_finetuning.py \
--restore_path /TestData/nemo2_ckpt/llama_68M \
--restore_path /home/TestData/nemo2_ckpt/llama_68M \
--devices 2 \
--max_steps 6 \
--experiment_dir tests/collections/llm/${{ github.run_id }} \
Expand All @@ -5263,7 +5292,7 @@ jobs:
SCRIPT: |
python tests/collections/llm/gpt_finetuning.py \
--restore_path /TestData/nemo2_ckpt/llama_68M \
--restore_path /home/TestData/nemo2_ckpt/llama_68M \
--devices 2 \
--max_steps 3 \
--experiment_dir tests/collections/llm/${{ github.run_id }} \
Expand All @@ -5273,7 +5302,7 @@ jobs:
--mbs 2
python tests/collections/llm/gpt_finetuning.py \
--restore_path /TestData/nemo2_ckpt/llama_68M \
--restore_path /home/TestData/nemo2_ckpt/llama_68M \
--devices 2 \
--max_steps 6 \
--experiment_dir tests/collections/llm/${{ github.run_id }} \
Expand All @@ -5294,7 +5323,7 @@ jobs:
SCRIPT: |
python tests/collections/llm/gpt_finetuning.py \
--restore_path /TestData/nemo2_ckpt/llama_68M \
--restore_path /home/TestData/nemo2_ckpt/llama_68M \
--devices 2 \
--max_steps 3 \
--experiment_dir tests/collections/llm/${{ github.run_id }} \
Expand All @@ -5304,7 +5333,7 @@ jobs:
--mbs 2
python tests/collections/llm/gpt_finetuning.py \
--restore_path /TestData/nemo2_ckpt/llama_68M \
--restore_path /home/TestData/nemo2_ckpt/llama_68M \
--devices 2 \
--max_steps 6 \
--experiment_dir tests/collections/llm/${{ github.run_id }} \
Expand All @@ -5325,7 +5354,7 @@ jobs:
SCRIPT: |
python tests/collections/llm/gpt_finetuning.py \
--restore_path /TestData/nemo2_ckpt/llama_68M \
--restore_path /home/TestData/nemo2_ckpt/llama_68M \
--devices 2 \
--max_steps 3 \
--experiment_dir tests/collections/llm/${{ github.run_id }} \
Expand All @@ -5335,7 +5364,7 @@ jobs:
--mbs 2
python tests/collections/llm/gpt_finetuning.py \
--restore_path /TestData/nemo2_ckpt/llama_68M \
--restore_path /home/TestData/nemo2_ckpt/llama_68M \
--devices 2 \
--max_steps 6 \
--experiment_dir tests/collections/llm/${{ github.run_id }} \
Expand All @@ -5356,7 +5385,7 @@ jobs:
SCRIPT: |
python tests/collections/llm/gpt_finetuning.py \
--restore_path /TestData/nemo2_ckpt/llama_68M \
--restore_path /home/TestData/nemo2_ckpt/llama_68M \
--devices 2 \
--max_steps 3 \
--experiment_dir tests/collections/llm/${{ github.run_id }} \
Expand All @@ -5366,7 +5395,7 @@ jobs:
--mbs 1
python tests/collections/llm/gpt_finetuning.py \
--restore_path /TestData/nemo2_ckpt/llama_68M \
--restore_path /home/TestData/nemo2_ckpt/llama_68M \
--devices 2 \
--max_steps 6 \
--experiment_dir tests/collections/llm/${{ github.run_id }} \
Expand All @@ -5387,7 +5416,7 @@ jobs:
SCRIPT: |
python tests/collections/llm/gpt_finetuning.py \
--restore_path /TestData/nemo2_ckpt/llama_68M \
--restore_path /home/TestData/nemo2_ckpt/llama_68M \
--devices 2 \
--max_steps 3 \
--experiment_dir tests/collections/llm/${{ github.run_id }} \
Expand All @@ -5397,7 +5426,7 @@ jobs:
--mbs 2
python tests/collections/llm/gpt_finetuning.py \
--restore_path /TestData/nemo2_ckpt/llama_68M \
--restore_path /home/TestData/nemo2_ckpt/llama_68M \
--devices 2 \
--max_steps 6 \
--experiment_dir tests/collections/llm/${{ github.run_id }} \
Expand All @@ -5418,7 +5447,7 @@ jobs:
SCRIPT: |
python tests/collections/llm/gpt_finetuning.py \
--restore_path /TestData/nemo2_ckpt/llama_68M \
--restore_path /home/TestData/nemo2_ckpt/llama_68M \
--devices 2 \
--max_steps 3 \
--experiment_dir tests/collections/llm/${{ github.run_id }} \
Expand All @@ -5428,7 +5457,7 @@ jobs:
--mbs 2
python tests/collections/llm/gpt_finetuning.py \
--restore_path /TestData/nemo2_ckpt/llama_68M \
--restore_path /home/TestData/nemo2_ckpt/llama_68M \
--devices 2 \
--max_steps 6 \
--experiment_dir tests/collections/llm/${{ github.run_id }} \
Expand All @@ -5449,7 +5478,7 @@ jobs:
SCRIPT: |
python tests/collections/llm/gpt_finetuning.py \
--restore_path /TestData/nemo2_ckpt/llama_68M \
--restore_path /home/TestData/nemo2_ckpt/llama_68M \
--devices 2 \
--max_steps 3 \
--experiment_dir tests/collections/llm/${{ github.run_id }} \
Expand All @@ -5459,7 +5488,7 @@ jobs:
--mbs 2
python tests/collections/llm/gpt_finetuning.py \
--restore_path /TestData/nemo2_ckpt/llama_68M \
--restore_path /home/TestData/nemo2_ckpt/llama_68M \
--devices 2 \
--max_steps 6 \
--experiment_dir tests/collections/llm/${{ github.run_id }} \
Expand All @@ -5471,35 +5500,6 @@ jobs:
AFTER_SCRIPT: |
rm -rf tests/collections/llm/${{ github.run_id }}
L2_NeMo_2_HF_MODEL_IMPORT:
needs: [cicd-test-container-setup]
uses: ./.github/workflows/_test_template.yml
with:
RUNNER: self-hosted-azure
SCRIPT: |
python tests/collections/llm/gpt/model/test_model_import.py
AFTER_SCRIPT: |
rm -rf ~/.cache/nemo/models
L2_NeMo_2_T5_Pretraining:
needs: [cicd-test-container-setup]
uses: ./.github/workflows/_test_template.yml
if: contains(fromJSON(needs.cicd-test-container-setup.outputs.test_to_run), 'L2_NeMo_2_T5_Pretraining') || needs.cicd-test-container-setup.outputs.all == 'true'
with:
RUNNER: self-hosted-azure
SCRIPT: |
NVTE_FUSED_ATTN=0 NVTE_FLASH_ATTN=0 python tests/collections/llm/megatron_t5_pretraining.py \
--devices=2 \
--max-steps=3 \
--experiment-dir=tests/collections/llm/t5_pretrain_results \
--data-path=/home/TestData/nlp/megatron_t5/data/pile_val_small_bert_tokenizer_text_document \
--index-mapping-dir=tests/collections/llm/t5_index_mappings
AFTER_SCRIPT: |
rm -rf tests/collections/llm/t5_pretrain_results
rm -rf tests/collections/llm/t5_index_mappings
Nemo_CICD_Test:
needs:
- pre-flight
Expand Down Expand Up @@ -5637,6 +5637,7 @@ jobs:
- L2_NeMo_2_HF_MODEL_IMPORT
- L2_NeMo_2_SSM_Pretraining
- L2_NeMo_2_SSM_Finetuning
- L2_NeMo_2_T5_Pretraining
- L2_NeMo_2_GPT_SFT_TP1PP1_MBS1
- L2_NeMo_2_GPT_SFT_TP1PP1_MBS2
- L2_NeMo_2_GPT_SFT_TP1PP2_MBS2
Expand All @@ -5645,7 +5646,6 @@ jobs:
- L2_NeMo_2_GPT_LoRA_TP1PP1_MBS2
- L2_NeMo_2_GPT_LoRA_TP1PP2_MBS2
- L2_NeMo_2_GPT_LoRA_TP2PP1_MBS2
- L2_NeMo_2_T5_Pretraining
if: always()
runs-on: ubuntu-latest
steps:
Expand Down

0 comments on commit 40ae724

Please sign in to comment.