DeepView.Predict/.github/workflows/benchmarking.yaml at main · CentML/DeepView.Predict · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
name: benchmark

on:
  workflow_dispatch:

env:
  DEVICE_PAIRS: L4,A100;L4,T4;T4,A100;L4,V100;V100,A100;T4,V100

jobs:
  create-summary:
    needs: [experiments-t4, experiments-l4, experiments-a100, experiments-v100]
    runs-on: ubuntu-latest
    steps:
      - name: Fetch repository
        uses: actions/checkout@v4

      - name: create directory to gather all csv files
        run: |
          mkdir all_results

      - name: Download All Artifacts
        uses: actions/download-artifact@v4
        with:
          path: all_results
          merge-multiple: true

      - name: check files
        run: |
          ls -R all_results

      - name: generate end-to-end and per operation results
        run: |
          sudo apt install python3-pip -y
          pip3 install pandas dominate
          python3 ./experiments/process_results.py --in-dir all_results --out-e2e out_e2e --out-ops out_ops

      - name: upload artifact
        uses: actions/upload-artifact@v4
        with:
          name: artifacts-combined
          path: |
            ./experiments/out_e2e/
            ./experiments/out_ops/

      - name: Configure AWS credentials
        uses: aws-actions/configure-aws-credentials@v4
        with:
          aws-access-key-id: ${{ secrets.HABITAT_AWS_ACCESS_KEY }}
          aws-region: ${{ secrets.HABITAT_AWS_REGION }}
          aws-secret-access-key: ${{ secrets.HABITAT_AWS_SECRET_ACCESS_KEY }}

      - name: Generate html pages
        run: |
          python3 ./experiments/generate_html_summary.py --e2e ./experiments/out_e2e --ops ./experiments/out_ops
          mkdir benchmark_results
          mv *.html benchmark_results

      - name: Upload to S3
        run: |
          aws s3 cp ./benchmark_results s3://centml-releases/habitat/benchmark --recursive


  experiments-t4:
    runs-on: [self-hosted, dev, t4]
    steps:
      - name: Fetch repository
        uses: actions/checkout@v4

      - name: run experiments
        run: |
          ./experiments/benchmarker_helper_script.sh python3.10
        env:
          LOCAL_DEVICE: T4

      - name: upload artifact
        uses: actions/upload-artifact@v4
        with:
          name: artifacts-t4
          path: ./experiments/results/

  experiments-v100:
    runs-on: [self-hosted, dev, v100]
    steps:
      - name: Fetch repository
        uses: actions/checkout@v4

      - name: run experiments
        run: |
          ./experiments/benchmarker_helper_script.sh python3.10
        env:
          LOCAL_DEVICE: V100

      - name: upload artifact
        uses: actions/upload-artifact@v4
        with:
          name: artifacts-v100
          path: ./experiments/results/

  experiments-l4:
    runs-on: [self-hosted, dev, l4]
    steps:
      - name: Fetch repository
        uses: actions/checkout@v4

      - name: run experiments
        run: |
          ./experiments/benchmarker_helper_script.sh python3.10
        env:
          LOCAL_DEVICE: L4

      - name: upload artifact
        uses: actions/upload-artifact@v4
        with:
          name: artifacts-l4
          path: ./experiments/results/

  experiments-a100:
    runs-on: [self-hosted, dev, a100]
    steps:
      - name: Fetch repository
        uses: actions/checkout@v4

      - name: run experiments
        run: |
          ./experiments/benchmarker_helper_script.sh python3.10
        env:
          LOCAL_DEVICE: A100

      - name: upload artifact
        uses: actions/upload-artifact@v4
        with:
          name: artifacts-a100
          path: ./experiments/results/