55 - aws s3 cp ../../../datasets/ s3://mango-pe/pg2-benchmark-training-data/datasets
66 --recursive --exclude "*" --include "*.zip"
77 - aws s3 cp ../../../models/ s3://mango-pe/pg2-benchmark-training-data/models
8- --recursive --exclude "*" --include "*.toml "
8+ --recursive --exclude "*" --include "README.md "
99 - echo "Upload completed at $(date)" > logs/s3_upload_complete.txt
1010 deps:
1111 - path: ../../../datasets/
@@ -15,25 +15,25 @@ stages:
1515 nfiles: 7
1616 - path: ../../../models/
1717 hash: md5
18- md5: d3ad141f0c54799e8f96bd640c686b79 .dir
19- size: 1381199847
20- nfiles: 48393
18+ md5: 8a2a310671abbb469d6d45d56c1ec859 .dir
19+ size: 1381200239
20+ nfiles: 48391
2121 - path: logs/setup.txt
2222 hash: md5
2323 md5: c861b0c201d5e28492f5dff8fab634f5
2424 size: 26
2525 outs:
2626 - path: logs/s3_upload_complete.txt
2727 hash: md5
28- md5: b841533c85ad082aeed20da08556f9cc
28+ md5: 2de978e81f106363692c3c775ae44ecb
2929 size: 50
3030 deploy_to_ecr@model0:
3131 cmd:
3232 - aws ecr describe-repositories --repository-names pls --region us-east-1 >/dev/null
3333 2>&1 || aws ecr create-repository --repository-name pls --region us-east-1 >/dev/null
3434 - aws ecr get-login-password --region us-east-1 | docker login --username AWS
3535 --password-stdin ${AWS_ACCOUNT_ID}.dkr.ecr.us-east-1.amazonaws.com
36- - docker buildx build --build-arg GIT_CACHE_BUST=1 --platform linux/amd64,linux/arm64
36+ - docker buildx build --build-arg GIT_CACHE_BUST=2 --platform linux/amd64,linux/arm64
3737 --secret id=git_auth,src=../git-auth.txt -f ../../../models/pls/Dockerfile -t
3838 ${AWS_ACCOUNT_ID}.dkr.ecr.us-east-1.amazonaws.com/pls:latest ../../.. --push
3939 - echo "ECR push completed at $(date)" > logs/ecr_push_complete.txt
@@ -50,104 +50,104 @@ stages:
5050 outs:
5151 - path: logs/ecr_push_complete.txt
5252 hash: md5
53- md5: e59aeee19986bf50446a104090b14741
53+ md5: 8225867a9f663498ab8e7812c97e0d9a
5454 size: 52
5555 - path: logs/image_uri.txt
5656 hash: md5
5757 md5: 764fdb00245735eff50661b2da0d51a1
5858 size: 56
5959 create_training_job@dataset0-model0:
6060 cmd: "uv run pg2-benchmark sagemaker create-training-job --model-name pls --region-name
61- us-east-1 --sagemaker-role-name rnd-dev-sagemakerexecutionrole --ecr-repository-uri
61+ us-east-1 --sagemaker-role-name rnd-dev-sagemakerexecutionrole --ecr-repository-uri
6262 ${AWS_ACCOUNT_ID}.dkr.ecr.us-east-1.amazonaws.com/pls --s3-training-data-prefix
6363 mango-pe/pg2-benchmark-training-data --s3-output-prefix mango-pe/pg2-benchmark-output
6464 --instance-type ml.m5.xlarge --volume-size 5 --dataset-prefix dummy --model-prefix
6565 pls > logs/create_job_charge_ladder_pls.txt\n"
6666 deps:
6767 - path: ../../../src/pg2_benchmark/cli/sagemaker.py
6868 hash: md5
69- md5: 51bdf7c32397e96258c0b0a5eb2e9759
70- size: 5176
69+ md5: 47ce56161a649f4829dd64c6bde4b4bf
70+ size: 5178
7171 - path: logs/ecr_push_complete.txt
7272 hash: md5
73- md5: e59aeee19986bf50446a104090b14741
73+ md5: 8225867a9f663498ab8e7812c97e0d9a
7474 size: 52
7575 - path: logs/image_uri.txt
7676 hash: md5
7777 md5: 764fdb00245735eff50661b2da0d51a1
7878 size: 56
7979 - path: logs/s3_upload_complete.txt
8080 hash: md5
81- md5: b841533c85ad082aeed20da08556f9cc
81+ md5: 2de978e81f106363692c3c775ae44ecb
8282 size: 50
8383 outs:
8484 - path: logs/create_job_charge_ladder_pls.txt
8585 hash: md5
86- md5: a746456fd6bf63aa1dc9998e9bc74ed0
86+ md5: 9840699441bb4a04e370720b50541308
8787 size: 20
8888 create_training_job@dataset1-model0:
8989 cmd: "uv run pg2-benchmark sagemaker create-training-job --model-name pls --region-name
90- us-east-1 --sagemaker-role-name rnd-dev-sagemakerexecutionrole --ecr-repository-uri
90+ us-east-1 --sagemaker-role-name rnd-dev-sagemakerexecutionrole --ecr-repository-uri
9191 ${AWS_ACCOUNT_ID}.dkr.ecr.us-east-1.amazonaws.com/pls --s3-training-data-prefix
9292 mango-pe/pg2-benchmark-training-data --s3-output-prefix mango-pe/pg2-benchmark-output
9393 --instance-type ml.m5.xlarge --volume-size 5 --dataset-prefix neime --model-prefix
9494 pls > logs/create_job_neime_pls.txt\n"
9595 deps:
9696 - path: ../../../src/pg2_benchmark/cli/sagemaker.py
9797 hash: md5
98- md5: 51bdf7c32397e96258c0b0a5eb2e9759
99- size: 5176
98+ md5: 47ce56161a649f4829dd64c6bde4b4bf
99+ size: 5178
100100 - path: logs/ecr_push_complete.txt
101101 hash: md5
102- md5: e59aeee19986bf50446a104090b14741
102+ md5: 8225867a9f663498ab8e7812c97e0d9a
103103 size: 52
104104 - path: logs/image_uri.txt
105105 hash: md5
106106 md5: 764fdb00245735eff50661b2da0d51a1
107107 size: 56
108108 - path: logs/s3_upload_complete.txt
109109 hash: md5
110- md5: b841533c85ad082aeed20da08556f9cc
110+ md5: 2de978e81f106363692c3c775ae44ecb
111111 size: 50
112112 outs:
113113 - path: logs/create_job_neime_pls.txt
114114 hash: md5
115- md5: bc8eb0dd0c04cc015516c0237401a538
115+ md5: b2760cf024c3f96ce24798aa7566f8c0
116116 size: 20
117117 monitor_training_job@dataset0-model0:
118118 cmd: "uv run pg2-benchmark sagemaker monitor-training-job --region-name us-east-1
119119 --job-name $(cat logs/create_job_charge_ladder_pls.txt) > logs/monitor_job_charge_ladder_pls.txt\n"
120120 deps:
121121 - path: ../../../src/pg2_benchmark/cli/sagemaker.py
122122 hash: md5
123- md5: 51bdf7c32397e96258c0b0a5eb2e9759
124- size: 5176
123+ md5: 47ce56161a649f4829dd64c6bde4b4bf
124+ size: 5178
125125 - path: logs/create_job_charge_ladder_pls.txt
126126 hash: md5
127- md5: a746456fd6bf63aa1dc9998e9bc74ed0
127+ md5: 9840699441bb4a04e370720b50541308
128128 size: 20
129129 outs:
130130 - path: logs/monitor_job_charge_ladder_pls.txt
131131 hash: md5
132- md5: ebce6202916ab76ffc188be21d7222ae
133- size: 158
132+ md5: b83ea2ee7bc48007cd5f240683cdc043
133+ size: 245
134134 monitor_training_job@dataset1-model0:
135135 cmd: "uv run pg2-benchmark sagemaker monitor-training-job --region-name us-east-1
136136 --job-name $(cat logs/create_job_neime_pls.txt) > logs/monitor_job_neime_pls.txt\n"
137137 deps:
138138 - path: ../../../src/pg2_benchmark/cli/sagemaker.py
139139 hash: md5
140- md5: 51bdf7c32397e96258c0b0a5eb2e9759
141- size: 5176
140+ md5: 47ce56161a649f4829dd64c6bde4b4bf
141+ size: 5178
142142 - path: logs/create_job_neime_pls.txt
143143 hash: md5
144- md5: bc8eb0dd0c04cc015516c0237401a538
144+ md5: b2760cf024c3f96ce24798aa7566f8c0
145145 size: 20
146146 outs:
147147 - path: logs/monitor_job_neime_pls.txt
148148 hash: md5
149- md5: 1b1592af89d391864d56058b621d2d4c
150- size: 245
149+ md5: 930f15c1dc41b2d72f0d24c430dd4609
150+ size: 216
151151 calculate_metric@dataset0-model0:
152152 cmd:
153153 - aws s3 cp s3://mango-pe/pg2-benchmark-output/$(cat logs/create_job_charge_ladder_pls.txt)/output/model.tar.gz
@@ -159,12 +159,12 @@ stages:
159159 deps:
160160 - path: logs/monitor_job_charge_ladder_pls.txt
161161 hash: md5
162- md5: ebce6202916ab76ffc188be21d7222ae
163- size: 158
162+ md5: b83ea2ee7bc48007cd5f240683cdc043
163+ size: 245
164164 outs:
165165 - path: metric/charge_ladder_pls.csv
166166 hash: md5
167- md5: a60602f67fbd13b9da7ba2535995e97e
167+ md5: a94b29b80493faa875ea39623058344b
168168 size: 1457
169169 calculate_metric@dataset1-model0:
170170 cmd:
@@ -177,13 +177,13 @@ stages:
177177 deps:
178178 - path: logs/monitor_job_neime_pls.txt
179179 hash: md5
180- md5: 1b1592af89d391864d56058b621d2d4c
181- size: 245
180+ md5: 930f15c1dc41b2d72f0d24c430dd4609
181+ size: 216
182182 outs:
183183 - path: metric/neime_pls.csv
184184 hash: md5
185- md5: b3475d5df5e71acd2bcdd5854f5640d3
186- size: 1626
185+ md5: 43a6a51e0ff20d2a322412eac6e524bb
186+ size: 1616
187187 setup:
188188 cmd:
189189 - mkdir -p logs output metric
0 commit comments