hf-transformers-bot commited on
Commit
b378100
1 Parent(s): 717424a

Upload folder using huggingface_hub

Browse files
Files changed (32) hide show
  1. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=False/.hydra/config.yaml +91 -0
  2. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=False/.hydra/hydra.yaml +141 -0
  3. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=False/.hydra/overrides.yaml +3 -0
  4. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=False/benchmark.json +443 -0
  5. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=False/benchmark.log +0 -0
  6. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=False/benchmark_config.json +102 -0
  7. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=False/benchmark_report.json +339 -0
  8. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=False/summary.json +13 -0
  9. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=True/.hydra/config.yaml +91 -0
  10. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=True/.hydra/hydra.yaml +141 -0
  11. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=True/.hydra/overrides.yaml +3 -0
  12. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=True/benchmark.log +1 -0
  13. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=True/benchmark_config.json +102 -0
  14. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=False/.hydra/config.yaml +91 -0
  15. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=False/.hydra/hydra.yaml +141 -0
  16. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=False/.hydra/overrides.yaml +3 -0
  17. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=False/benchmark.json +443 -0
  18. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=False/benchmark.log +0 -0
  19. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=False/benchmark_config.json +102 -0
  20. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=False/benchmark_report.json +339 -0
  21. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=False/summary.json +13 -0
  22. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=True/.hydra/config.yaml +91 -0
  23. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=True/.hydra/hydra.yaml +141 -0
  24. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=True/.hydra/overrides.yaml +3 -0
  25. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=True/benchmark.json +443 -0
  26. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=True/benchmark.log +0 -0
  27. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=True/benchmark_config.json +102 -0
  28. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=True/benchmark_report.json +339 -0
  29. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=True/summary.json +13 -0
  30. 2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/multirun.yaml +232 -0
  31. 2024-05-31/summaries.json +41 -0
  32. 2024-05-31/summary.json +40 -0
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=False/.hydra/config.yaml ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ name: pytorch_generate
2
+ backend:
3
+ name: pytorch
4
+ version: 2.3.0+cu121
5
+ _target_: optimum_benchmark.backends.pytorch.backend.PyTorchBackend
6
+ task: null
7
+ library: null
8
+ model: google/gemma-2b
9
+ processor: null
10
+ device: cuda
11
+ device_ids: '0'
12
+ seed: 42
13
+ inter_op_num_threads: null
14
+ intra_op_num_threads: null
15
+ model_kwargs: {}
16
+ processor_kwargs: {}
17
+ hub_kwargs: {}
18
+ no_weights: true
19
+ device_map: null
20
+ torch_dtype: float16
21
+ eval_mode: true
22
+ to_bettertransformer: false
23
+ low_cpu_mem_usage: null
24
+ attn_implementation: null
25
+ cache_implementation: null
26
+ autocast_enabled: false
27
+ autocast_dtype: null
28
+ torch_compile: false
29
+ torch_compile_target: forward
30
+ torch_compile_config:
31
+ backend: inductor
32
+ mode: reduce-overhead
33
+ fullgraph: true
34
+ quantization_scheme: null
35
+ quantization_config: {}
36
+ deepspeed_inference: false
37
+ deepspeed_inference_config: {}
38
+ peft_type: null
39
+ peft_config: {}
40
+ scenario:
41
+ name: inference
42
+ _target_: optimum_benchmark.scenarios.inference.scenario.InferenceScenario
43
+ iterations: 2
44
+ duration: 0
45
+ warmup_runs: 10
46
+ input_shapes:
47
+ batch_size: 1
48
+ sequence_length: 7
49
+ new_tokens: null
50
+ latency: true
51
+ memory: true
52
+ energy: false
53
+ forward_kwargs: {}
54
+ generate_kwargs:
55
+ max_new_tokens: 128
56
+ min_new_tokens: 128
57
+ do_sample: false
58
+ call_kwargs: {}
59
+ launcher:
60
+ name: process
61
+ _target_: optimum_benchmark.launchers.process.launcher.ProcessLauncher
62
+ device_isolation: true
63
+ device_isolation_action: warn
64
+ start_method: spawn
65
+ environment:
66
+ cpu: ' AMD EPYC 7R32'
67
+ cpu_count: 16
68
+ cpu_ram_mb: 66697.29792
69
+ system: Linux
70
+ machine: x86_64
71
+ platform: Linux-5.10.216-204.855.amzn2.x86_64-x86_64-with-glibc2.29
72
+ processor: x86_64
73
+ python_version: 3.8.10
74
+ gpu:
75
+ - NVIDIA A10G
76
+ gpu_count: 1
77
+ gpu_vram_mb: 24146608128
78
+ optimum_benchmark_version: 0.2.1
79
+ optimum_benchmark_commit: null
80
+ transformers_version: 4.42.0.dev0
81
+ transformers_commit: a51a3fbc4827db26e896292b6409246fe8152cd4
82
+ accelerate_version: 0.31.0.dev0
83
+ accelerate_commit: null
84
+ diffusers_version: null
85
+ diffusers_commit: null
86
+ optimum_version: 1.21.0.dev0
87
+ optimum_commit: null
88
+ timm_version: 0.9.16
89
+ timm_commit: null
90
+ peft_version: 0.11.2.dev0
91
+ peft_commit: null
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=False/.hydra/hydra.yaml ADDED
@@ -0,0 +1,141 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ hydra:
2
+ run:
3
+ dir: _benchmark/tmpyph4ye2g/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/${hydra.job.override_dirname}
4
+ sweep:
5
+ dir: _benchmark/tmpyph4ye2g/commit=a51a3fbc4827db26e896292b6409246fe8152cd4
6
+ subdir: ${hydra.job.override_dirname}
7
+ launcher:
8
+ _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
9
+ sweeper:
10
+ _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
11
+ max_batch_size: null
12
+ params: null
13
+ help:
14
+ app_name: ${hydra.job.name}
15
+ header: '${hydra.help.app_name} is powered by Hydra.
16
+
17
+ '
18
+ footer: 'Powered by Hydra (https://hydra.cc)
19
+
20
+ Use --hydra-help to view Hydra specific help
21
+
22
+ '
23
+ template: '${hydra.help.header}
24
+
25
+ == Configuration groups ==
26
+
27
+ Compose your configuration from those groups (group=option)
28
+
29
+
30
+ $APP_CONFIG_GROUPS
31
+
32
+
33
+ == Config ==
34
+
35
+ Override anything in the config (foo.bar=value)
36
+
37
+
38
+ $CONFIG
39
+
40
+
41
+ ${hydra.help.footer}
42
+
43
+ '
44
+ hydra_help:
45
+ template: 'Hydra (${hydra.runtime.version})
46
+
47
+ See https://hydra.cc for more info.
48
+
49
+
50
+ == Flags ==
51
+
52
+ $FLAGS_HELP
53
+
54
+
55
+ == Configuration groups ==
56
+
57
+ Compose your configuration from those groups (For example, append hydra/job_logging=disabled
58
+ to command line)
59
+
60
+
61
+ $HYDRA_CONFIG_GROUPS
62
+
63
+
64
+ Use ''--cfg hydra'' to Show the Hydra config.
65
+
66
+ '
67
+ hydra_help: ???
68
+ hydra_logging:
69
+ version: 1
70
+ root:
71
+ level: ERROR
72
+ disable_existing_loggers: true
73
+ job_logging:
74
+ version: 1
75
+ root:
76
+ level: ERROR
77
+ disable_existing_loggers: true
78
+ env: {}
79
+ mode: MULTIRUN
80
+ searchpath: []
81
+ callbacks: {}
82
+ output_subdir: .hydra
83
+ overrides:
84
+ hydra:
85
+ - hydra/job_logging=disabled
86
+ - hydra/hydra_logging=disabled
87
+ - hydra.sweep.dir=_benchmark/tmpyph4ye2g/commit\=a51a3fbc4827db26e896292b6409246fe8152cd4
88
+ - hydra.run.dir=_benchmark/tmpyph4ye2g/commit\=a51a3fbc4827db26e896292b6409246fe8152cd4/${hydra.job.override_dirname}
89
+ - hydra.mode=MULTIRUN
90
+ task:
91
+ - backend.model=google/gemma-2b
92
+ - backend.cache_implementation=null
93
+ - backend.torch_compile=False
94
+ job:
95
+ name: cli
96
+ chdir: true
97
+ override_dirname: backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=False
98
+ id: '0'
99
+ num: 0
100
+ config_name: generation
101
+ env_set:
102
+ OVERRIDE_BENCHMARKS: '1'
103
+ LOG_LEVEL: WARN
104
+ env_copy: []
105
+ config:
106
+ override_dirname:
107
+ kv_sep: '='
108
+ item_sep: ','
109
+ exclude_keys: []
110
+ runtime:
111
+ version: 1.3.2
112
+ version_base: '1.3'
113
+ cwd: /transformers
114
+ config_sources:
115
+ - path: hydra.conf
116
+ schema: pkg
117
+ provider: hydra
118
+ - path: optimum_benchmark
119
+ schema: pkg
120
+ provider: main
121
+ - path: /transformers/benchmark/config
122
+ schema: file
123
+ provider: command-line
124
+ - path: ''
125
+ schema: structured
126
+ provider: schema
127
+ output_dir: /transformers/_benchmark/tmpyph4ye2g/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=False
128
+ choices:
129
+ backend: pytorch
130
+ launcher: process
131
+ scenario: inference
132
+ hydra/env: default
133
+ hydra/callbacks: null
134
+ hydra/job_logging: disabled
135
+ hydra/hydra_logging: disabled
136
+ hydra/hydra_help: default
137
+ hydra/help: default
138
+ hydra/sweeper: basic
139
+ hydra/launcher: basic
140
+ hydra/output: default
141
+ verbose: false
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=False/.hydra/overrides.yaml ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ - backend.model=google/gemma-2b
2
+ - backend.cache_implementation=null
3
+ - backend.torch_compile=False
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=False/benchmark.json ADDED
@@ -0,0 +1,443 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "config": {
3
+ "name": "pytorch_generate",
4
+ "backend": {
5
+ "name": "pytorch",
6
+ "version": "2.3.0+cu121",
7
+ "_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
8
+ "task": "text-generation",
9
+ "library": "transformers",
10
+ "model": "google/gemma-2b",
11
+ "processor": "google/gemma-2b",
12
+ "device": "cuda",
13
+ "device_ids": "0",
14
+ "seed": 42,
15
+ "inter_op_num_threads": null,
16
+ "intra_op_num_threads": null,
17
+ "model_kwargs": {},
18
+ "processor_kwargs": {},
19
+ "hub_kwargs": {},
20
+ "no_weights": true,
21
+ "device_map": null,
22
+ "torch_dtype": "float16",
23
+ "eval_mode": true,
24
+ "to_bettertransformer": false,
25
+ "low_cpu_mem_usage": null,
26
+ "attn_implementation": null,
27
+ "cache_implementation": null,
28
+ "autocast_enabled": false,
29
+ "autocast_dtype": null,
30
+ "torch_compile": false,
31
+ "torch_compile_target": "forward",
32
+ "torch_compile_config": {
33
+ "backend": "inductor",
34
+ "mode": "reduce-overhead",
35
+ "fullgraph": true
36
+ },
37
+ "quantization_scheme": null,
38
+ "quantization_config": {},
39
+ "deepspeed_inference": false,
40
+ "deepspeed_inference_config": {},
41
+ "peft_type": null,
42
+ "peft_config": {}
43
+ },
44
+ "scenario": {
45
+ "name": "inference",
46
+ "_target_": "optimum_benchmark.scenarios.inference.scenario.InferenceScenario",
47
+ "iterations": 2,
48
+ "duration": 0,
49
+ "warmup_runs": 10,
50
+ "input_shapes": {
51
+ "batch_size": 1,
52
+ "num_choices": 2,
53
+ "sequence_length": 7
54
+ },
55
+ "new_tokens": null,
56
+ "latency": true,
57
+ "memory": true,
58
+ "energy": false,
59
+ "forward_kwargs": {},
60
+ "generate_kwargs": {
61
+ "max_new_tokens": 128,
62
+ "min_new_tokens": 128,
63
+ "do_sample": false
64
+ },
65
+ "call_kwargs": {}
66
+ },
67
+ "launcher": {
68
+ "name": "process",
69
+ "_target_": "optimum_benchmark.launchers.process.launcher.ProcessLauncher",
70
+ "device_isolation": true,
71
+ "device_isolation_action": "warn",
72
+ "start_method": "spawn"
73
+ },
74
+ "environment": {
75
+ "cpu": " AMD EPYC 7R32",
76
+ "cpu_count": 16,
77
+ "cpu_ram_mb": 66697.29792,
78
+ "system": "Linux",
79
+ "machine": "x86_64",
80
+ "platform": "Linux-5.10.216-204.855.amzn2.x86_64-x86_64-with-glibc2.29",
81
+ "processor": "x86_64",
82
+ "python_version": "3.8.10",
83
+ "gpu": [
84
+ "NVIDIA A10G"
85
+ ],
86
+ "gpu_count": 1,
87
+ "gpu_vram_mb": 24146608128,
88
+ "optimum_benchmark_version": "0.2.1",
89
+ "optimum_benchmark_commit": null,
90
+ "transformers_version": "4.42.0.dev0",
91
+ "transformers_commit": "a51a3fbc4827db26e896292b6409246fe8152cd4",
92
+ "accelerate_version": "0.31.0.dev0",
93
+ "accelerate_commit": null,
94
+ "diffusers_version": null,
95
+ "diffusers_commit": null,
96
+ "optimum_version": "1.21.0.dev0",
97
+ "optimum_commit": null,
98
+ "timm_version": "0.9.16",
99
+ "timm_commit": null,
100
+ "peft_version": "0.11.2.dev0",
101
+ "peft_commit": null
102
+ }
103
+ },
104
+ "report": {
105
+ "prefill": {
106
+ "memory": {
107
+ "unit": "MB",
108
+ "max_ram": 1739.145216,
109
+ "max_global_vram": 6795.296768,
110
+ "max_process_vram": 0.0,
111
+ "max_reserved": 6148.849664,
112
+ "max_allocated": 5033.552896
113
+ },
114
+ "latency": {
115
+ "unit": "s",
116
+ "count": 2,
117
+ "total": 0.03906150245666504,
118
+ "mean": 0.01953075122833252,
119
+ "stdev": 0.0002228479385375983,
120
+ "p50": 0.01953075122833252,
121
+ "p90": 0.0197090295791626,
122
+ "p95": 0.019731314373016357,
123
+ "p99": 0.019749142208099365,
124
+ "values": [
125
+ 0.01975359916687012,
126
+ 0.019307903289794922
127
+ ]
128
+ },
129
+ "throughput": {
130
+ "unit": "tokens/s",
131
+ "value": 358.4091527337344
132
+ },
133
+ "energy": null,
134
+ "efficiency": null
135
+ },
136
+ "decode": {
137
+ "memory": {
138
+ "unit": "MB",
139
+ "max_ram": 1739.18208,
140
+ "max_global_vram": 6795.296768,
141
+ "max_process_vram": 0.0,
142
+ "max_reserved": 6148.849664,
143
+ "max_allocated": 5033.553408
144
+ },
145
+ "latency": {
146
+ "unit": "s",
147
+ "count": 2,
148
+ "total": 4.7793039550781256,
149
+ "mean": 2.3896519775390628,
150
+ "stdev": 0.00408459472656264,
151
+ "p50": 2.3896519775390628,
152
+ "p90": 2.3929196533203125,
153
+ "p95": 2.393328112792969,
154
+ "p99": 2.393654880371094,
155
+ "values": [
156
+ 2.3855673828125,
157
+ 2.393736572265625
158
+ ]
159
+ },
160
+ "throughput": {
161
+ "unit": "tokens/s",
162
+ "value": 53.14581419960094
163
+ },
164
+ "energy": null,
165
+ "efficiency": null
166
+ },
167
+ "per_token": {
168
+ "memory": null,
169
+ "latency": {
170
+ "unit": "s",
171
+ "count": 253,
172
+ "total": 4.779476997375489,
173
+ "mean": 0.01889121342836161,
174
+ "stdev": 0.0012441678618480634,
175
+ "p50": 0.018684928894042968,
176
+ "p90": 0.01933967361450195,
177
+ "p95": 0.0194596866607666,
178
+ "p99": 0.019907501678466798,
179
+ "values": [
180
+ 0.01986867141723633,
181
+ 0.01901158332824707,
182
+ 0.018514944076538087,
183
+ 0.01844428825378418,
184
+ 0.018447359085083007,
185
+ 0.01848320007324219,
186
+ 0.018506752014160157,
187
+ 0.01844428825378418,
188
+ 0.018543615341186523,
189
+ 0.018512895584106445,
190
+ 0.018498559951782227,
191
+ 0.018513919830322266,
192
+ 0.01862758445739746,
193
+ 0.018489343643188477,
194
+ 0.018524160385131837,
195
+ 0.01843302345275879,
196
+ 0.018519039154052733,
197
+ 0.018765823364257812,
198
+ 0.01860403251647949,
199
+ 0.018566144943237304,
200
+ 0.01850060844421387,
201
+ 0.018524160385131837,
202
+ 0.018508800506591795,
203
+ 0.018510847091674804,
204
+ 0.018551807403564453,
205
+ 0.0184770565032959,
206
+ 0.018527231216430663,
207
+ 0.019532800674438477,
208
+ 0.01948057556152344,
209
+ 0.01862553596496582,
210
+ 0.019138559341430664,
211
+ 0.019131391525268555,
212
+ 0.018581504821777343,
213
+ 0.018537471771240235,
214
+ 0.018573312759399413,
215
+ 0.018564096450805666,
216
+ 0.01860095977783203,
217
+ 0.018536447525024414,
218
+ 0.018537471771240235,
219
+ 0.018543615341186523,
220
+ 0.01856716728210449,
221
+ 0.018536447525024414,
222
+ 0.018547712326049806,
223
+ 0.018543615341186523,
224
+ 0.018561023712158203,
225
+ 0.018552831649780274,
226
+ 0.018897920608520507,
227
+ 0.019412992477416992,
228
+ 0.019200000762939453,
229
+ 0.01902079963684082,
230
+ 0.019315711975097655,
231
+ 0.019345407485961915,
232
+ 0.018940927505493164,
233
+ 0.018981887817382814,
234
+ 0.018922496795654296,
235
+ 0.019143680572509765,
236
+ 0.01866547203063965,
237
+ 0.018684928894042968,
238
+ 0.018594816207885743,
239
+ 0.01861427116394043,
240
+ 0.018914304733276367,
241
+ 0.01959321594238281,
242
+ 0.019260416030883788,
243
+ 0.018807807922363282,
244
+ 0.018685951232910156,
245
+ 0.018634752273559572,
246
+ 0.018742271423339844,
247
+ 0.018588672637939452,
248
+ 0.018692096710205077,
249
+ 0.0186296329498291,
250
+ 0.018671615600585938,
251
+ 0.01860915184020996,
252
+ 0.0186429443359375,
253
+ 0.018748416900634765,
254
+ 0.018667520523071288,
255
+ 0.018700288772583007,
256
+ 0.018669567108154296,
257
+ 0.018740224838256835,
258
+ 0.018699264526367186,
259
+ 0.018689023971557618,
260
+ 0.018727935791015626,
261
+ 0.01878937530517578,
262
+ 0.018752511978149415,
263
+ 0.018709503173828124,
264
+ 0.018678783416748047,
265
+ 0.018707456588745116,
266
+ 0.018704383850097657,
267
+ 0.018731008529663085,
268
+ 0.018715648651123046,
269
+ 0.018889759063720705,
270
+ 0.018770912170410155,
271
+ 0.018647039413452148,
272
+ 0.01864908790588379,
273
+ 0.018676736831665038,
274
+ 0.018659328460693358,
275
+ 0.018659328460693358,
276
+ 0.018723840713500976,
277
+ 0.018727935791015626,
278
+ 0.01864499282836914,
279
+ 0.01867366409301758,
280
+ 0.018766847610473633,
281
+ 0.01998028755187988,
282
+ 0.019373056411743163,
283
+ 0.019329023361206055,
284
+ 0.018701311111450195,
285
+ 0.018651136398315428,
286
+ 0.0186429443359375,
287
+ 0.018671615600585938,
288
+ 0.018776063919067384,
289
+ 0.018699264526367186,
290
+ 0.01868185615539551,
291
+ 0.01962393569946289,
292
+ 0.018684928894042968,
293
+ 0.018678783416748047,
294
+ 0.018930688858032226,
295
+ 0.018710527420043945,
296
+ 0.018752511978149415,
297
+ 0.01879756736755371,
298
+ 0.01881907272338867,
299
+ 0.01925324821472168,
300
+ 0.019369983673095705,
301
+ 0.019398656845092774,
302
+ 0.019337215423583985,
303
+ 0.019336223602294922,
304
+ 0.01871766471862793,
305
+ 0.018755584716796874,
306
+ 0.03796582412719727,
307
+ 0.018486272811889647,
308
+ 0.019139583587646485,
309
+ 0.018507776260375978,
310
+ 0.018585599899291993,
311
+ 0.018580480575561522,
312
+ 0.018559999465942383,
313
+ 0.018537471771240235,
314
+ 0.018538496017456055,
315
+ 0.018968576431274413,
316
+ 0.018527231216430663,
317
+ 0.018541568756103514,
318
+ 0.01864089584350586,
319
+ 0.018588672637939452,
320
+ 0.01861324882507324,
321
+ 0.01865830421447754,
322
+ 0.018701311111450195,
323
+ 0.018580575942993165,
324
+ 0.019162015914916994,
325
+ 0.018581504821777343,
326
+ 0.01860505676269531,
327
+ 0.01860915184020996,
328
+ 0.018580480575561522,
329
+ 0.018946048736572265,
330
+ 0.019192832946777344,
331
+ 0.019198976516723632,
332
+ 0.019136512756347656,
333
+ 0.019949567794799804,
334
+ 0.01942118453979492,
335
+ 0.01968332862854004,
336
+ 0.019296255111694336,
337
+ 0.018560064315795897,
338
+ 0.01853843116760254,
339
+ 0.018651136398315428,
340
+ 0.018669567108154296,
341
+ 0.01862451171875,
342
+ 0.018554880142211915,
343
+ 0.018571264266967775,
344
+ 0.018557952880859374,
345
+ 0.01855897521972656,
346
+ 0.018807807922363282,
347
+ 0.0186296329498291,
348
+ 0.018689023971557618,
349
+ 0.01864089584350586,
350
+ 0.018990079879760743,
351
+ 0.019331071853637697,
352
+ 0.01861324882507324,
353
+ 0.018523136138916017,
354
+ 0.018561023712158203,
355
+ 0.018530303955078126,
356
+ 0.01861529541015625,
357
+ 0.01855897521972656,
358
+ 0.018555904388427736,
359
+ 0.01859993553161621,
360
+ 0.018524160385131837,
361
+ 0.019128320693969726,
362
+ 0.019105791091918945,
363
+ 0.01923583984375,
364
+ 0.018747392654418944,
365
+ 0.01880473518371582,
366
+ 0.018945024490356444,
367
+ 0.0192542724609375,
368
+ 0.01884671974182129,
369
+ 0.01879859161376953,
370
+ 0.018678783416748047,
371
+ 0.019149824142456053,
372
+ 0.019332096099853514,
373
+ 0.018738176345825194,
374
+ 0.018757696151733397,
375
+ 0.018823104858398436,
376
+ 0.018682880401611326,
377
+ 0.018699296951293944,
378
+ 0.01863983917236328,
379
+ 0.019127296447753905,
380
+ 0.019542015075683594,
381
+ 0.019481599807739256,
382
+ 0.01944576072692871,
383
+ 0.019368959426879884,
384
+ 0.018807807922363282,
385
+ 0.018714624404907225,
386
+ 0.019808256149291992,
387
+ 0.019280895233154297,
388
+ 0.01942630386352539,
389
+ 0.018719743728637696,
390
+ 0.01924300765991211,
391
+ 0.01867366409301758,
392
+ 0.018678783416748047,
393
+ 0.019136512756347656,
394
+ 0.01939455986022949,
395
+ 0.01959321594238281,
396
+ 0.019381248474121093,
397
+ 0.01938739204406738,
398
+ 0.019340288162231444,
399
+ 0.019315711975097655,
400
+ 0.01865727996826172,
401
+ 0.01880678367614746,
402
+ 0.018702335357666015,
403
+ 0.018774015426635742,
404
+ 0.018692096710205077,
405
+ 0.01864192008972168,
406
+ 0.018709503173828124,
407
+ 0.01863680076599121,
408
+ 0.018727935791015626,
409
+ 0.018619392395019533,
410
+ 0.0186562557220459,
411
+ 0.018662399291992187,
412
+ 0.018695167541503906,
413
+ 0.01864499282836914,
414
+ 0.01868185615539551,
415
+ 0.018646015167236327,
416
+ 0.018663423538208008,
417
+ 0.01861734390258789,
418
+ 0.018717695236206054,
419
+ 0.01863884735107422,
420
+ 0.01863577651977539,
421
+ 0.019094528198242186,
422
+ 0.019168256759643554,
423
+ 0.019159040451049804,
424
+ 0.01920614433288574,
425
+ 0.018669567108154296,
426
+ 0.01864192008972168,
427
+ 0.018714624404907225,
428
+ 0.01884876823425293,
429
+ 0.018667520523071288,
430
+ 0.01861631965637207,
431
+ 0.01863167953491211,
432
+ 0.01863270378112793
433
+ ]
434
+ },
435
+ "throughput": {
436
+ "unit": "tokens/s",
437
+ "value": 52.9346621270335
438
+ },
439
+ "energy": null,
440
+ "efficiency": null
441
+ }
442
+ }
443
+ }
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=False/benchmark.log ADDED
File without changes
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=False/benchmark_config.json ADDED
@@ -0,0 +1,102 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "pytorch_generate",
3
+ "backend": {
4
+ "name": "pytorch",
5
+ "version": "2.3.0+cu121",
6
+ "_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
7
+ "task": "text-generation",
8
+ "library": "transformers",
9
+ "model": "google/gemma-2b",
10
+ "processor": "google/gemma-2b",
11
+ "device": "cuda",
12
+ "device_ids": "0",
13
+ "seed": 42,
14
+ "inter_op_num_threads": null,
15
+ "intra_op_num_threads": null,
16
+ "model_kwargs": {},
17
+ "processor_kwargs": {},
18
+ "hub_kwargs": {},
19
+ "no_weights": true,
20
+ "device_map": null,
21
+ "torch_dtype": "float16",
22
+ "eval_mode": true,
23
+ "to_bettertransformer": false,
24
+ "low_cpu_mem_usage": null,
25
+ "attn_implementation": null,
26
+ "cache_implementation": null,
27
+ "autocast_enabled": false,
28
+ "autocast_dtype": null,
29
+ "torch_compile": false,
30
+ "torch_compile_target": "forward",
31
+ "torch_compile_config": {
32
+ "backend": "inductor",
33
+ "mode": "reduce-overhead",
34
+ "fullgraph": true
35
+ },
36
+ "quantization_scheme": null,
37
+ "quantization_config": {},
38
+ "deepspeed_inference": false,
39
+ "deepspeed_inference_config": {},
40
+ "peft_type": null,
41
+ "peft_config": {}
42
+ },
43
+ "scenario": {
44
+ "name": "inference",
45
+ "_target_": "optimum_benchmark.scenarios.inference.scenario.InferenceScenario",
46
+ "iterations": 2,
47
+ "duration": 0,
48
+ "warmup_runs": 10,
49
+ "input_shapes": {
50
+ "batch_size": 1,
51
+ "num_choices": 2,
52
+ "sequence_length": 7
53
+ },
54
+ "new_tokens": null,
55
+ "latency": true,
56
+ "memory": true,
57
+ "energy": false,
58
+ "forward_kwargs": {},
59
+ "generate_kwargs": {
60
+ "max_new_tokens": 128,
61
+ "min_new_tokens": 128,
62
+ "do_sample": false
63
+ },
64
+ "call_kwargs": {}
65
+ },
66
+ "launcher": {
67
+ "name": "process",
68
+ "_target_": "optimum_benchmark.launchers.process.launcher.ProcessLauncher",
69
+ "device_isolation": true,
70
+ "device_isolation_action": "warn",
71
+ "start_method": "spawn"
72
+ },
73
+ "environment": {
74
+ "cpu": " AMD EPYC 7R32",
75
+ "cpu_count": 16,
76
+ "cpu_ram_mb": 66697.29792,
77
+ "system": "Linux",
78
+ "machine": "x86_64",
79
+ "platform": "Linux-5.10.216-204.855.amzn2.x86_64-x86_64-with-glibc2.29",
80
+ "processor": "x86_64",
81
+ "python_version": "3.8.10",
82
+ "gpu": [
83
+ "NVIDIA A10G"
84
+ ],
85
+ "gpu_count": 1,
86
+ "gpu_vram_mb": 24146608128,
87
+ "optimum_benchmark_version": "0.2.1",
88
+ "optimum_benchmark_commit": null,
89
+ "transformers_version": "4.42.0.dev0",
90
+ "transformers_commit": "a51a3fbc4827db26e896292b6409246fe8152cd4",
91
+ "accelerate_version": "0.31.0.dev0",
92
+ "accelerate_commit": null,
93
+ "diffusers_version": null,
94
+ "diffusers_commit": null,
95
+ "optimum_version": "1.21.0.dev0",
96
+ "optimum_commit": null,
97
+ "timm_version": "0.9.16",
98
+ "timm_commit": null,
99
+ "peft_version": "0.11.2.dev0",
100
+ "peft_commit": null
101
+ }
102
+ }
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=False/benchmark_report.json ADDED
@@ -0,0 +1,339 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "prefill": {
3
+ "memory": {
4
+ "unit": "MB",
5
+ "max_ram": 1739.145216,
6
+ "max_global_vram": 6795.296768,
7
+ "max_process_vram": 0.0,
8
+ "max_reserved": 6148.849664,
9
+ "max_allocated": 5033.552896
10
+ },
11
+ "latency": {
12
+ "unit": "s",
13
+ "count": 2,
14
+ "total": 0.03906150245666504,
15
+ "mean": 0.01953075122833252,
16
+ "stdev": 0.0002228479385375983,
17
+ "p50": 0.01953075122833252,
18
+ "p90": 0.0197090295791626,
19
+ "p95": 0.019731314373016357,
20
+ "p99": 0.019749142208099365,
21
+ "values": [
22
+ 0.01975359916687012,
23
+ 0.019307903289794922
24
+ ]
25
+ },
26
+ "throughput": {
27
+ "unit": "tokens/s",
28
+ "value": 358.4091527337344
29
+ },
30
+ "energy": null,
31
+ "efficiency": null
32
+ },
33
+ "decode": {
34
+ "memory": {
35
+ "unit": "MB",
36
+ "max_ram": 1739.18208,
37
+ "max_global_vram": 6795.296768,
38
+ "max_process_vram": 0.0,
39
+ "max_reserved": 6148.849664,
40
+ "max_allocated": 5033.553408
41
+ },
42
+ "latency": {
43
+ "unit": "s",
44
+ "count": 2,
45
+ "total": 4.7793039550781256,
46
+ "mean": 2.3896519775390628,
47
+ "stdev": 0.00408459472656264,
48
+ "p50": 2.3896519775390628,
49
+ "p90": 2.3929196533203125,
50
+ "p95": 2.393328112792969,
51
+ "p99": 2.393654880371094,
52
+ "values": [
53
+ 2.3855673828125,
54
+ 2.393736572265625
55
+ ]
56
+ },
57
+ "throughput": {
58
+ "unit": "tokens/s",
59
+ "value": 53.14581419960094
60
+ },
61
+ "energy": null,
62
+ "efficiency": null
63
+ },
64
+ "per_token": {
65
+ "memory": null,
66
+ "latency": {
67
+ "unit": "s",
68
+ "count": 253,
69
+ "total": 4.779476997375489,
70
+ "mean": 0.01889121342836161,
71
+ "stdev": 0.0012441678618480634,
72
+ "p50": 0.018684928894042968,
73
+ "p90": 0.01933967361450195,
74
+ "p95": 0.0194596866607666,
75
+ "p99": 0.019907501678466798,
76
+ "values": [
77
+ 0.01986867141723633,
78
+ 0.01901158332824707,
79
+ 0.018514944076538087,
80
+ 0.01844428825378418,
81
+ 0.018447359085083007,
82
+ 0.01848320007324219,
83
+ 0.018506752014160157,
84
+ 0.01844428825378418,
85
+ 0.018543615341186523,
86
+ 0.018512895584106445,
87
+ 0.018498559951782227,
88
+ 0.018513919830322266,
89
+ 0.01862758445739746,
90
+ 0.018489343643188477,
91
+ 0.018524160385131837,
92
+ 0.01843302345275879,
93
+ 0.018519039154052733,
94
+ 0.018765823364257812,
95
+ 0.01860403251647949,
96
+ 0.018566144943237304,
97
+ 0.01850060844421387,
98
+ 0.018524160385131837,
99
+ 0.018508800506591795,
100
+ 0.018510847091674804,
101
+ 0.018551807403564453,
102
+ 0.0184770565032959,
103
+ 0.018527231216430663,
104
+ 0.019532800674438477,
105
+ 0.01948057556152344,
106
+ 0.01862553596496582,
107
+ 0.019138559341430664,
108
+ 0.019131391525268555,
109
+ 0.018581504821777343,
110
+ 0.018537471771240235,
111
+ 0.018573312759399413,
112
+ 0.018564096450805666,
113
+ 0.01860095977783203,
114
+ 0.018536447525024414,
115
+ 0.018537471771240235,
116
+ 0.018543615341186523,
117
+ 0.01856716728210449,
118
+ 0.018536447525024414,
119
+ 0.018547712326049806,
120
+ 0.018543615341186523,
121
+ 0.018561023712158203,
122
+ 0.018552831649780274,
123
+ 0.018897920608520507,
124
+ 0.019412992477416992,
125
+ 0.019200000762939453,
126
+ 0.01902079963684082,
127
+ 0.019315711975097655,
128
+ 0.019345407485961915,
129
+ 0.018940927505493164,
130
+ 0.018981887817382814,
131
+ 0.018922496795654296,
132
+ 0.019143680572509765,
133
+ 0.01866547203063965,
134
+ 0.018684928894042968,
135
+ 0.018594816207885743,
136
+ 0.01861427116394043,
137
+ 0.018914304733276367,
138
+ 0.01959321594238281,
139
+ 0.019260416030883788,
140
+ 0.018807807922363282,
141
+ 0.018685951232910156,
142
+ 0.018634752273559572,
143
+ 0.018742271423339844,
144
+ 0.018588672637939452,
145
+ 0.018692096710205077,
146
+ 0.0186296329498291,
147
+ 0.018671615600585938,
148
+ 0.01860915184020996,
149
+ 0.0186429443359375,
150
+ 0.018748416900634765,
151
+ 0.018667520523071288,
152
+ 0.018700288772583007,
153
+ 0.018669567108154296,
154
+ 0.018740224838256835,
155
+ 0.018699264526367186,
156
+ 0.018689023971557618,
157
+ 0.018727935791015626,
158
+ 0.01878937530517578,
159
+ 0.018752511978149415,
160
+ 0.018709503173828124,
161
+ 0.018678783416748047,
162
+ 0.018707456588745116,
163
+ 0.018704383850097657,
164
+ 0.018731008529663085,
165
+ 0.018715648651123046,
166
+ 0.018889759063720705,
167
+ 0.018770912170410155,
168
+ 0.018647039413452148,
169
+ 0.01864908790588379,
170
+ 0.018676736831665038,
171
+ 0.018659328460693358,
172
+ 0.018659328460693358,
173
+ 0.018723840713500976,
174
+ 0.018727935791015626,
175
+ 0.01864499282836914,
176
+ 0.01867366409301758,
177
+ 0.018766847610473633,
178
+ 0.01998028755187988,
179
+ 0.019373056411743163,
180
+ 0.019329023361206055,
181
+ 0.018701311111450195,
182
+ 0.018651136398315428,
183
+ 0.0186429443359375,
184
+ 0.018671615600585938,
185
+ 0.018776063919067384,
186
+ 0.018699264526367186,
187
+ 0.01868185615539551,
188
+ 0.01962393569946289,
189
+ 0.018684928894042968,
190
+ 0.018678783416748047,
191
+ 0.018930688858032226,
192
+ 0.018710527420043945,
193
+ 0.018752511978149415,
194
+ 0.01879756736755371,
195
+ 0.01881907272338867,
196
+ 0.01925324821472168,
197
+ 0.019369983673095705,
198
+ 0.019398656845092774,
199
+ 0.019337215423583985,
200
+ 0.019336223602294922,
201
+ 0.01871766471862793,
202
+ 0.018755584716796874,
203
+ 0.03796582412719727,
204
+ 0.018486272811889647,
205
+ 0.019139583587646485,
206
+ 0.018507776260375978,
207
+ 0.018585599899291993,
208
+ 0.018580480575561522,
209
+ 0.018559999465942383,
210
+ 0.018537471771240235,
211
+ 0.018538496017456055,
212
+ 0.018968576431274413,
213
+ 0.018527231216430663,
214
+ 0.018541568756103514,
215
+ 0.01864089584350586,
216
+ 0.018588672637939452,
217
+ 0.01861324882507324,
218
+ 0.01865830421447754,
219
+ 0.018701311111450195,
220
+ 0.018580575942993165,
221
+ 0.019162015914916994,
222
+ 0.018581504821777343,
223
+ 0.01860505676269531,
224
+ 0.01860915184020996,
225
+ 0.018580480575561522,
226
+ 0.018946048736572265,
227
+ 0.019192832946777344,
228
+ 0.019198976516723632,
229
+ 0.019136512756347656,
230
+ 0.019949567794799804,
231
+ 0.01942118453979492,
232
+ 0.01968332862854004,
233
+ 0.019296255111694336,
234
+ 0.018560064315795897,
235
+ 0.01853843116760254,
236
+ 0.018651136398315428,
237
+ 0.018669567108154296,
238
+ 0.01862451171875,
239
+ 0.018554880142211915,
240
+ 0.018571264266967775,
241
+ 0.018557952880859374,
242
+ 0.01855897521972656,
243
+ 0.018807807922363282,
244
+ 0.0186296329498291,
245
+ 0.018689023971557618,
246
+ 0.01864089584350586,
247
+ 0.018990079879760743,
248
+ 0.019331071853637697,
249
+ 0.01861324882507324,
250
+ 0.018523136138916017,
251
+ 0.018561023712158203,
252
+ 0.018530303955078126,
253
+ 0.01861529541015625,
254
+ 0.01855897521972656,
255
+ 0.018555904388427736,
256
+ 0.01859993553161621,
257
+ 0.018524160385131837,
258
+ 0.019128320693969726,
259
+ 0.019105791091918945,
260
+ 0.01923583984375,
261
+ 0.018747392654418944,
262
+ 0.01880473518371582,
263
+ 0.018945024490356444,
264
+ 0.0192542724609375,
265
+ 0.01884671974182129,
266
+ 0.01879859161376953,
267
+ 0.018678783416748047,
268
+ 0.019149824142456053,
269
+ 0.019332096099853514,
270
+ 0.018738176345825194,
271
+ 0.018757696151733397,
272
+ 0.018823104858398436,
273
+ 0.018682880401611326,
274
+ 0.018699296951293944,
275
+ 0.01863983917236328,
276
+ 0.019127296447753905,
277
+ 0.019542015075683594,
278
+ 0.019481599807739256,
279
+ 0.01944576072692871,
280
+ 0.019368959426879884,
281
+ 0.018807807922363282,
282
+ 0.018714624404907225,
283
+ 0.019808256149291992,
284
+ 0.019280895233154297,
285
+ 0.01942630386352539,
286
+ 0.018719743728637696,
287
+ 0.01924300765991211,
288
+ 0.01867366409301758,
289
+ 0.018678783416748047,
290
+ 0.019136512756347656,
291
+ 0.01939455986022949,
292
+ 0.01959321594238281,
293
+ 0.019381248474121093,
294
+ 0.01938739204406738,
295
+ 0.019340288162231444,
296
+ 0.019315711975097655,
297
+ 0.01865727996826172,
298
+ 0.01880678367614746,
299
+ 0.018702335357666015,
300
+ 0.018774015426635742,
301
+ 0.018692096710205077,
302
+ 0.01864192008972168,
303
+ 0.018709503173828124,
304
+ 0.01863680076599121,
305
+ 0.018727935791015626,
306
+ 0.018619392395019533,
307
+ 0.0186562557220459,
308
+ 0.018662399291992187,
309
+ 0.018695167541503906,
310
+ 0.01864499282836914,
311
+ 0.01868185615539551,
312
+ 0.018646015167236327,
313
+ 0.018663423538208008,
314
+ 0.01861734390258789,
315
+ 0.018717695236206054,
316
+ 0.01863884735107422,
317
+ 0.01863577651977539,
318
+ 0.019094528198242186,
319
+ 0.019168256759643554,
320
+ 0.019159040451049804,
321
+ 0.01920614433288574,
322
+ 0.018669567108154296,
323
+ 0.01864192008972168,
324
+ 0.018714624404907225,
325
+ 0.01884876823425293,
326
+ 0.018667520523071288,
327
+ 0.01861631965637207,
328
+ 0.01863167953491211,
329
+ 0.01863270378112793
330
+ ]
331
+ },
332
+ "throughput": {
333
+ "unit": "tokens/s",
334
+ "value": 52.9346621270335
335
+ },
336
+ "energy": null,
337
+ "efficiency": null
338
+ }
339
+ }
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=False/summary.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model": "google/gemma-2b",
3
+ "commit": "a51a3fbc4827db26e896292b6409246fe8152cd4",
4
+ "config": "backend.cache_implementation=null,backend.torch_compile=False",
5
+ "metrics": {
6
+ "prefill.latency.mean": 0.01953075122833252,
7
+ "prefill.throughput.value": 358.4091527337344,
8
+ "decode.latency.mean": 2.3896519775390628,
9
+ "decode.throughput.value": 53.14581419960094,
10
+ "per_token.latency.mean": 0.01889121342836161,
11
+ "per_token.throughput.value": 52.9346621270335
12
+ }
13
+ }
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=True/.hydra/config.yaml ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ name: pytorch_generate
2
+ backend:
3
+ name: pytorch
4
+ version: 2.3.0+cu121
5
+ _target_: optimum_benchmark.backends.pytorch.backend.PyTorchBackend
6
+ task: null
7
+ library: null
8
+ model: google/gemma-2b
9
+ processor: null
10
+ device: cuda
11
+ device_ids: '0'
12
+ seed: 42
13
+ inter_op_num_threads: null
14
+ intra_op_num_threads: null
15
+ model_kwargs: {}
16
+ processor_kwargs: {}
17
+ hub_kwargs: {}
18
+ no_weights: true
19
+ device_map: null
20
+ torch_dtype: float16
21
+ eval_mode: true
22
+ to_bettertransformer: false
23
+ low_cpu_mem_usage: null
24
+ attn_implementation: null
25
+ cache_implementation: null
26
+ autocast_enabled: false
27
+ autocast_dtype: null
28
+ torch_compile: true
29
+ torch_compile_target: forward
30
+ torch_compile_config:
31
+ backend: inductor
32
+ mode: reduce-overhead
33
+ fullgraph: true
34
+ quantization_scheme: null
35
+ quantization_config: {}
36
+ deepspeed_inference: false
37
+ deepspeed_inference_config: {}
38
+ peft_type: null
39
+ peft_config: {}
40
+ scenario:
41
+ name: inference
42
+ _target_: optimum_benchmark.scenarios.inference.scenario.InferenceScenario
43
+ iterations: 2
44
+ duration: 0
45
+ warmup_runs: 10
46
+ input_shapes:
47
+ batch_size: 1
48
+ sequence_length: 7
49
+ new_tokens: null
50
+ latency: true
51
+ memory: true
52
+ energy: false
53
+ forward_kwargs: {}
54
+ generate_kwargs:
55
+ max_new_tokens: 128
56
+ min_new_tokens: 128
57
+ do_sample: false
58
+ call_kwargs: {}
59
+ launcher:
60
+ name: process
61
+ _target_: optimum_benchmark.launchers.process.launcher.ProcessLauncher
62
+ device_isolation: true
63
+ device_isolation_action: warn
64
+ start_method: spawn
65
+ environment:
66
+ cpu: ' AMD EPYC 7R32'
67
+ cpu_count: 16
68
+ cpu_ram_mb: 66697.29792
69
+ system: Linux
70
+ machine: x86_64
71
+ platform: Linux-5.10.216-204.855.amzn2.x86_64-x86_64-with-glibc2.29
72
+ processor: x86_64
73
+ python_version: 3.8.10
74
+ gpu:
75
+ - NVIDIA A10G
76
+ gpu_count: 1
77
+ gpu_vram_mb: 24146608128
78
+ optimum_benchmark_version: 0.2.1
79
+ optimum_benchmark_commit: null
80
+ transformers_version: 4.42.0.dev0
81
+ transformers_commit: a51a3fbc4827db26e896292b6409246fe8152cd4
82
+ accelerate_version: 0.31.0.dev0
83
+ accelerate_commit: null
84
+ diffusers_version: null
85
+ diffusers_commit: null
86
+ optimum_version: 1.21.0.dev0
87
+ optimum_commit: null
88
+ timm_version: 0.9.16
89
+ timm_commit: null
90
+ peft_version: 0.11.2.dev0
91
+ peft_commit: null
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=True/.hydra/hydra.yaml ADDED
@@ -0,0 +1,141 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ hydra:
2
+ run:
3
+ dir: _benchmark/tmpyph4ye2g/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/${hydra.job.override_dirname}
4
+ sweep:
5
+ dir: _benchmark/tmpyph4ye2g/commit=a51a3fbc4827db26e896292b6409246fe8152cd4
6
+ subdir: ${hydra.job.override_dirname}
7
+ launcher:
8
+ _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
9
+ sweeper:
10
+ _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
11
+ max_batch_size: null
12
+ params: null
13
+ help:
14
+ app_name: ${hydra.job.name}
15
+ header: '${hydra.help.app_name} is powered by Hydra.
16
+
17
+ '
18
+ footer: 'Powered by Hydra (https://hydra.cc)
19
+
20
+ Use --hydra-help to view Hydra specific help
21
+
22
+ '
23
+ template: '${hydra.help.header}
24
+
25
+ == Configuration groups ==
26
+
27
+ Compose your configuration from those groups (group=option)
28
+
29
+
30
+ $APP_CONFIG_GROUPS
31
+
32
+
33
+ == Config ==
34
+
35
+ Override anything in the config (foo.bar=value)
36
+
37
+
38
+ $CONFIG
39
+
40
+
41
+ ${hydra.help.footer}
42
+
43
+ '
44
+ hydra_help:
45
+ template: 'Hydra (${hydra.runtime.version})
46
+
47
+ See https://hydra.cc for more info.
48
+
49
+
50
+ == Flags ==
51
+
52
+ $FLAGS_HELP
53
+
54
+
55
+ == Configuration groups ==
56
+
57
+ Compose your configuration from those groups (For example, append hydra/job_logging=disabled
58
+ to command line)
59
+
60
+
61
+ $HYDRA_CONFIG_GROUPS
62
+
63
+
64
+ Use ''--cfg hydra'' to Show the Hydra config.
65
+
66
+ '
67
+ hydra_help: ???
68
+ hydra_logging:
69
+ version: 1
70
+ root:
71
+ level: ERROR
72
+ disable_existing_loggers: true
73
+ job_logging:
74
+ version: 1
75
+ root:
76
+ level: ERROR
77
+ disable_existing_loggers: true
78
+ env: {}
79
+ mode: MULTIRUN
80
+ searchpath: []
81
+ callbacks: {}
82
+ output_subdir: .hydra
83
+ overrides:
84
+ hydra:
85
+ - hydra/job_logging=disabled
86
+ - hydra/hydra_logging=disabled
87
+ - hydra.sweep.dir=_benchmark/tmpyph4ye2g/commit\=a51a3fbc4827db26e896292b6409246fe8152cd4
88
+ - hydra.run.dir=_benchmark/tmpyph4ye2g/commit\=a51a3fbc4827db26e896292b6409246fe8152cd4/${hydra.job.override_dirname}
89
+ - hydra.mode=MULTIRUN
90
+ task:
91
+ - backend.model=google/gemma-2b
92
+ - backend.cache_implementation=null
93
+ - backend.torch_compile=True
94
+ job:
95
+ name: cli
96
+ chdir: true
97
+ override_dirname: backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=True
98
+ id: '1'
99
+ num: 1
100
+ config_name: generation
101
+ env_set:
102
+ OVERRIDE_BENCHMARKS: '1'
103
+ LOG_LEVEL: WARN
104
+ env_copy: []
105
+ config:
106
+ override_dirname:
107
+ kv_sep: '='
108
+ item_sep: ','
109
+ exclude_keys: []
110
+ runtime:
111
+ version: 1.3.2
112
+ version_base: '1.3'
113
+ cwd: /transformers
114
+ config_sources:
115
+ - path: hydra.conf
116
+ schema: pkg
117
+ provider: hydra
118
+ - path: optimum_benchmark
119
+ schema: pkg
120
+ provider: main
121
+ - path: /transformers/benchmark/config
122
+ schema: file
123
+ provider: command-line
124
+ - path: ''
125
+ schema: structured
126
+ provider: schema
127
+ output_dir: /transformers/_benchmark/tmpyph4ye2g/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=True
128
+ choices:
129
+ backend: pytorch
130
+ launcher: process
131
+ scenario: inference
132
+ hydra/env: default
133
+ hydra/callbacks: null
134
+ hydra/job_logging: disabled
135
+ hydra/hydra_logging: disabled
136
+ hydra/hydra_help: default
137
+ hydra/help: default
138
+ hydra/sweeper: basic
139
+ hydra/launcher: basic
140
+ hydra/output: default
141
+ verbose: false
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=True/.hydra/overrides.yaml ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ - backend.model=google/gemma-2b
2
+ - backend.cache_implementation=null
3
+ - backend.torch_compile=True
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=True/benchmark.log ADDED
@@ -0,0 +1 @@
 
 
1
+ [ISOLATED-PROCESS][2024-05-31 15:18:44,660][process][ERROR] - + Sending traceback to main process
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=null,backend.model=google/gemma-2b,backend.torch_compile=True/benchmark_config.json ADDED
@@ -0,0 +1,102 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "pytorch_generate",
3
+ "backend": {
4
+ "name": "pytorch",
5
+ "version": "2.3.0+cu121",
6
+ "_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
7
+ "task": "text-generation",
8
+ "library": "transformers",
9
+ "model": "google/gemma-2b",
10
+ "processor": "google/gemma-2b",
11
+ "device": "cuda",
12
+ "device_ids": "0",
13
+ "seed": 42,
14
+ "inter_op_num_threads": null,
15
+ "intra_op_num_threads": null,
16
+ "model_kwargs": {},
17
+ "processor_kwargs": {},
18
+ "hub_kwargs": {},
19
+ "no_weights": true,
20
+ "device_map": null,
21
+ "torch_dtype": "float16",
22
+ "eval_mode": true,
23
+ "to_bettertransformer": false,
24
+ "low_cpu_mem_usage": null,
25
+ "attn_implementation": null,
26
+ "cache_implementation": null,
27
+ "autocast_enabled": false,
28
+ "autocast_dtype": null,
29
+ "torch_compile": true,
30
+ "torch_compile_target": "forward",
31
+ "torch_compile_config": {
32
+ "backend": "inductor",
33
+ "mode": "reduce-overhead",
34
+ "fullgraph": true
35
+ },
36
+ "quantization_scheme": null,
37
+ "quantization_config": {},
38
+ "deepspeed_inference": false,
39
+ "deepspeed_inference_config": {},
40
+ "peft_type": null,
41
+ "peft_config": {}
42
+ },
43
+ "scenario": {
44
+ "name": "inference",
45
+ "_target_": "optimum_benchmark.scenarios.inference.scenario.InferenceScenario",
46
+ "iterations": 2,
47
+ "duration": 0,
48
+ "warmup_runs": 10,
49
+ "input_shapes": {
50
+ "batch_size": 1,
51
+ "num_choices": 2,
52
+ "sequence_length": 7
53
+ },
54
+ "new_tokens": null,
55
+ "latency": true,
56
+ "memory": true,
57
+ "energy": false,
58
+ "forward_kwargs": {},
59
+ "generate_kwargs": {
60
+ "max_new_tokens": 128,
61
+ "min_new_tokens": 128,
62
+ "do_sample": false
63
+ },
64
+ "call_kwargs": {}
65
+ },
66
+ "launcher": {
67
+ "name": "process",
68
+ "_target_": "optimum_benchmark.launchers.process.launcher.ProcessLauncher",
69
+ "device_isolation": true,
70
+ "device_isolation_action": "warn",
71
+ "start_method": "spawn"
72
+ },
73
+ "environment": {
74
+ "cpu": " AMD EPYC 7R32",
75
+ "cpu_count": 16,
76
+ "cpu_ram_mb": 66697.29792,
77
+ "system": "Linux",
78
+ "machine": "x86_64",
79
+ "platform": "Linux-5.10.216-204.855.amzn2.x86_64-x86_64-with-glibc2.29",
80
+ "processor": "x86_64",
81
+ "python_version": "3.8.10",
82
+ "gpu": [
83
+ "NVIDIA A10G"
84
+ ],
85
+ "gpu_count": 1,
86
+ "gpu_vram_mb": 24146608128,
87
+ "optimum_benchmark_version": "0.2.1",
88
+ "optimum_benchmark_commit": null,
89
+ "transformers_version": "4.42.0.dev0",
90
+ "transformers_commit": "a51a3fbc4827db26e896292b6409246fe8152cd4",
91
+ "accelerate_version": "0.31.0.dev0",
92
+ "accelerate_commit": null,
93
+ "diffusers_version": null,
94
+ "diffusers_commit": null,
95
+ "optimum_version": "1.21.0.dev0",
96
+ "optimum_commit": null,
97
+ "timm_version": "0.9.16",
98
+ "timm_commit": null,
99
+ "peft_version": "0.11.2.dev0",
100
+ "peft_commit": null
101
+ }
102
+ }
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=False/.hydra/config.yaml ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ name: pytorch_generate
2
+ backend:
3
+ name: pytorch
4
+ version: 2.3.0+cu121
5
+ _target_: optimum_benchmark.backends.pytorch.backend.PyTorchBackend
6
+ task: null
7
+ library: null
8
+ model: google/gemma-2b
9
+ processor: null
10
+ device: cuda
11
+ device_ids: '0'
12
+ seed: 42
13
+ inter_op_num_threads: null
14
+ intra_op_num_threads: null
15
+ model_kwargs: {}
16
+ processor_kwargs: {}
17
+ hub_kwargs: {}
18
+ no_weights: true
19
+ device_map: null
20
+ torch_dtype: float16
21
+ eval_mode: true
22
+ to_bettertransformer: false
23
+ low_cpu_mem_usage: null
24
+ attn_implementation: null
25
+ cache_implementation: static
26
+ autocast_enabled: false
27
+ autocast_dtype: null
28
+ torch_compile: false
29
+ torch_compile_target: forward
30
+ torch_compile_config:
31
+ backend: inductor
32
+ mode: reduce-overhead
33
+ fullgraph: true
34
+ quantization_scheme: null
35
+ quantization_config: {}
36
+ deepspeed_inference: false
37
+ deepspeed_inference_config: {}
38
+ peft_type: null
39
+ peft_config: {}
40
+ scenario:
41
+ name: inference
42
+ _target_: optimum_benchmark.scenarios.inference.scenario.InferenceScenario
43
+ iterations: 2
44
+ duration: 0
45
+ warmup_runs: 10
46
+ input_shapes:
47
+ batch_size: 1
48
+ sequence_length: 7
49
+ new_tokens: null
50
+ latency: true
51
+ memory: true
52
+ energy: false
53
+ forward_kwargs: {}
54
+ generate_kwargs:
55
+ max_new_tokens: 128
56
+ min_new_tokens: 128
57
+ do_sample: false
58
+ call_kwargs: {}
59
+ launcher:
60
+ name: process
61
+ _target_: optimum_benchmark.launchers.process.launcher.ProcessLauncher
62
+ device_isolation: true
63
+ device_isolation_action: warn
64
+ start_method: spawn
65
+ environment:
66
+ cpu: ' AMD EPYC 7R32'
67
+ cpu_count: 16
68
+ cpu_ram_mb: 66697.29792
69
+ system: Linux
70
+ machine: x86_64
71
+ platform: Linux-5.10.216-204.855.amzn2.x86_64-x86_64-with-glibc2.29
72
+ processor: x86_64
73
+ python_version: 3.8.10
74
+ gpu:
75
+ - NVIDIA A10G
76
+ gpu_count: 1
77
+ gpu_vram_mb: 24146608128
78
+ optimum_benchmark_version: 0.2.1
79
+ optimum_benchmark_commit: null
80
+ transformers_version: 4.42.0.dev0
81
+ transformers_commit: a51a3fbc4827db26e896292b6409246fe8152cd4
82
+ accelerate_version: 0.31.0.dev0
83
+ accelerate_commit: null
84
+ diffusers_version: null
85
+ diffusers_commit: null
86
+ optimum_version: 1.21.0.dev0
87
+ optimum_commit: null
88
+ timm_version: 0.9.16
89
+ timm_commit: null
90
+ peft_version: 0.11.2.dev0
91
+ peft_commit: null
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=False/.hydra/hydra.yaml ADDED
@@ -0,0 +1,141 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ hydra:
2
+ run:
3
+ dir: _benchmark/tmpyph4ye2g/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/${hydra.job.override_dirname}
4
+ sweep:
5
+ dir: _benchmark/tmpyph4ye2g/commit=a51a3fbc4827db26e896292b6409246fe8152cd4
6
+ subdir: ${hydra.job.override_dirname}
7
+ launcher:
8
+ _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
9
+ sweeper:
10
+ _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
11
+ max_batch_size: null
12
+ params: null
13
+ help:
14
+ app_name: ${hydra.job.name}
15
+ header: '${hydra.help.app_name} is powered by Hydra.
16
+
17
+ '
18
+ footer: 'Powered by Hydra (https://hydra.cc)
19
+
20
+ Use --hydra-help to view Hydra specific help
21
+
22
+ '
23
+ template: '${hydra.help.header}
24
+
25
+ == Configuration groups ==
26
+
27
+ Compose your configuration from those groups (group=option)
28
+
29
+
30
+ $APP_CONFIG_GROUPS
31
+
32
+
33
+ == Config ==
34
+
35
+ Override anything in the config (foo.bar=value)
36
+
37
+
38
+ $CONFIG
39
+
40
+
41
+ ${hydra.help.footer}
42
+
43
+ '
44
+ hydra_help:
45
+ template: 'Hydra (${hydra.runtime.version})
46
+
47
+ See https://hydra.cc for more info.
48
+
49
+
50
+ == Flags ==
51
+
52
+ $FLAGS_HELP
53
+
54
+
55
+ == Configuration groups ==
56
+
57
+ Compose your configuration from those groups (For example, append hydra/job_logging=disabled
58
+ to command line)
59
+
60
+
61
+ $HYDRA_CONFIG_GROUPS
62
+
63
+
64
+ Use ''--cfg hydra'' to Show the Hydra config.
65
+
66
+ '
67
+ hydra_help: ???
68
+ hydra_logging:
69
+ version: 1
70
+ root:
71
+ level: ERROR
72
+ disable_existing_loggers: true
73
+ job_logging:
74
+ version: 1
75
+ root:
76
+ level: ERROR
77
+ disable_existing_loggers: true
78
+ env: {}
79
+ mode: MULTIRUN
80
+ searchpath: []
81
+ callbacks: {}
82
+ output_subdir: .hydra
83
+ overrides:
84
+ hydra:
85
+ - hydra/job_logging=disabled
86
+ - hydra/hydra_logging=disabled
87
+ - hydra.sweep.dir=_benchmark/tmpyph4ye2g/commit\=a51a3fbc4827db26e896292b6409246fe8152cd4
88
+ - hydra.run.dir=_benchmark/tmpyph4ye2g/commit\=a51a3fbc4827db26e896292b6409246fe8152cd4/${hydra.job.override_dirname}
89
+ - hydra.mode=MULTIRUN
90
+ task:
91
+ - backend.model=google/gemma-2b
92
+ - backend.cache_implementation=static
93
+ - backend.torch_compile=False
94
+ job:
95
+ name: cli
96
+ chdir: true
97
+ override_dirname: backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=False
98
+ id: '2'
99
+ num: 2
100
+ config_name: generation
101
+ env_set:
102
+ OVERRIDE_BENCHMARKS: '1'
103
+ LOG_LEVEL: WARN
104
+ env_copy: []
105
+ config:
106
+ override_dirname:
107
+ kv_sep: '='
108
+ item_sep: ','
109
+ exclude_keys: []
110
+ runtime:
111
+ version: 1.3.2
112
+ version_base: '1.3'
113
+ cwd: /transformers
114
+ config_sources:
115
+ - path: hydra.conf
116
+ schema: pkg
117
+ provider: hydra
118
+ - path: optimum_benchmark
119
+ schema: pkg
120
+ provider: main
121
+ - path: /transformers/benchmark/config
122
+ schema: file
123
+ provider: command-line
124
+ - path: ''
125
+ schema: structured
126
+ provider: schema
127
+ output_dir: /transformers/_benchmark/tmpyph4ye2g/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=False
128
+ choices:
129
+ backend: pytorch
130
+ launcher: process
131
+ scenario: inference
132
+ hydra/env: default
133
+ hydra/callbacks: null
134
+ hydra/job_logging: disabled
135
+ hydra/hydra_logging: disabled
136
+ hydra/hydra_help: default
137
+ hydra/help: default
138
+ hydra/sweeper: basic
139
+ hydra/launcher: basic
140
+ hydra/output: default
141
+ verbose: false
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=False/.hydra/overrides.yaml ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ - backend.model=google/gemma-2b
2
+ - backend.cache_implementation=static
3
+ - backend.torch_compile=False
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=False/benchmark.json ADDED
@@ -0,0 +1,443 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "config": {
3
+ "name": "pytorch_generate",
4
+ "backend": {
5
+ "name": "pytorch",
6
+ "version": "2.3.0+cu121",
7
+ "_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
8
+ "task": "text-generation",
9
+ "library": "transformers",
10
+ "model": "google/gemma-2b",
11
+ "processor": "google/gemma-2b",
12
+ "device": "cuda",
13
+ "device_ids": "0",
14
+ "seed": 42,
15
+ "inter_op_num_threads": null,
16
+ "intra_op_num_threads": null,
17
+ "model_kwargs": {},
18
+ "processor_kwargs": {},
19
+ "hub_kwargs": {},
20
+ "no_weights": true,
21
+ "device_map": null,
22
+ "torch_dtype": "float16",
23
+ "eval_mode": true,
24
+ "to_bettertransformer": false,
25
+ "low_cpu_mem_usage": null,
26
+ "attn_implementation": null,
27
+ "cache_implementation": "static",
28
+ "autocast_enabled": false,
29
+ "autocast_dtype": null,
30
+ "torch_compile": false,
31
+ "torch_compile_target": "forward",
32
+ "torch_compile_config": {
33
+ "backend": "inductor",
34
+ "mode": "reduce-overhead",
35
+ "fullgraph": true
36
+ },
37
+ "quantization_scheme": null,
38
+ "quantization_config": {},
39
+ "deepspeed_inference": false,
40
+ "deepspeed_inference_config": {},
41
+ "peft_type": null,
42
+ "peft_config": {}
43
+ },
44
+ "scenario": {
45
+ "name": "inference",
46
+ "_target_": "optimum_benchmark.scenarios.inference.scenario.InferenceScenario",
47
+ "iterations": 2,
48
+ "duration": 0,
49
+ "warmup_runs": 10,
50
+ "input_shapes": {
51
+ "batch_size": 1,
52
+ "num_choices": 2,
53
+ "sequence_length": 7
54
+ },
55
+ "new_tokens": null,
56
+ "latency": true,
57
+ "memory": true,
58
+ "energy": false,
59
+ "forward_kwargs": {},
60
+ "generate_kwargs": {
61
+ "max_new_tokens": 128,
62
+ "min_new_tokens": 128,
63
+ "do_sample": false
64
+ },
65
+ "call_kwargs": {}
66
+ },
67
+ "launcher": {
68
+ "name": "process",
69
+ "_target_": "optimum_benchmark.launchers.process.launcher.ProcessLauncher",
70
+ "device_isolation": true,
71
+ "device_isolation_action": "warn",
72
+ "start_method": "spawn"
73
+ },
74
+ "environment": {
75
+ "cpu": " AMD EPYC 7R32",
76
+ "cpu_count": 16,
77
+ "cpu_ram_mb": 66697.29792,
78
+ "system": "Linux",
79
+ "machine": "x86_64",
80
+ "platform": "Linux-5.10.216-204.855.amzn2.x86_64-x86_64-with-glibc2.29",
81
+ "processor": "x86_64",
82
+ "python_version": "3.8.10",
83
+ "gpu": [
84
+ "NVIDIA A10G"
85
+ ],
86
+ "gpu_count": 1,
87
+ "gpu_vram_mb": 24146608128,
88
+ "optimum_benchmark_version": "0.2.1",
89
+ "optimum_benchmark_commit": null,
90
+ "transformers_version": "4.42.0.dev0",
91
+ "transformers_commit": "a51a3fbc4827db26e896292b6409246fe8152cd4",
92
+ "accelerate_version": "0.31.0.dev0",
93
+ "accelerate_commit": null,
94
+ "diffusers_version": null,
95
+ "diffusers_commit": null,
96
+ "optimum_version": "1.21.0.dev0",
97
+ "optimum_commit": null,
98
+ "timm_version": "0.9.16",
99
+ "timm_commit": null,
100
+ "peft_version": "0.11.2.dev0",
101
+ "peft_commit": null
102
+ }
103
+ },
104
+ "report": {
105
+ "prefill": {
106
+ "memory": {
107
+ "unit": "MB",
108
+ "max_ram": 1753.206784,
109
+ "max_global_vram": 6793.199616,
110
+ "max_process_vram": 0.0,
111
+ "max_reserved": 6146.752512,
112
+ "max_allocated": 5035.912192
113
+ },
114
+ "latency": {
115
+ "unit": "s",
116
+ "count": 2,
117
+ "total": 0.04486032104492188,
118
+ "mean": 0.02243016052246094,
119
+ "stdev": 0.0003491039276123046,
120
+ "p50": 0.02243016052246094,
121
+ "p90": 0.022709443664550782,
122
+ "p95": 0.022744354057312014,
123
+ "p99": 0.022772282371520996,
124
+ "values": [
125
+ 0.022779264450073243,
126
+ 0.022081056594848634
127
+ ]
128
+ },
129
+ "throughput": {
130
+ "unit": "tokens/s",
131
+ "value": 312.0797995623078
132
+ },
133
+ "energy": null,
134
+ "efficiency": null
135
+ },
136
+ "decode": {
137
+ "memory": {
138
+ "unit": "MB",
139
+ "max_ram": 1753.227264,
140
+ "max_global_vram": 6793.199616,
141
+ "max_process_vram": 0.0,
142
+ "max_reserved": 6146.752512,
143
+ "max_allocated": 5035.912704
144
+ },
145
+ "latency": {
146
+ "unit": "s",
147
+ "count": 2,
148
+ "total": 5.246615478515625,
149
+ "mean": 2.6233077392578124,
150
+ "stdev": 0.01454187011718755,
151
+ "p50": 2.6233077392578124,
152
+ "p90": 2.634941235351562,
153
+ "p95": 2.636395422363281,
154
+ "p99": 2.6375587719726563,
155
+ "values": [
156
+ 2.637849609375,
157
+ 2.608765869140625
158
+ ]
159
+ },
160
+ "throughput": {
161
+ "unit": "tokens/s",
162
+ "value": 48.4121622863549
163
+ },
164
+ "energy": null,
165
+ "efficiency": null
166
+ },
167
+ "per_token": {
168
+ "memory": null,
169
+ "latency": {
170
+ "unit": "s",
171
+ "count": 253,
172
+ "total": 5.24776754379273,
173
+ "mean": 0.020742164204714326,
174
+ "stdev": 0.0014777911921905315,
175
+ "p50": 0.020716543197631835,
176
+ "p90": 0.02107084846496582,
177
+ "p95": 0.021373951721191407,
178
+ "p99": 0.021995028305053704,
179
+ "values": [
180
+ 0.021323776245117186,
181
+ 0.021770240783691407,
182
+ 0.021406719207763672,
183
+ 0.02088243293762207,
184
+ 0.02087116813659668,
185
+ 0.021331968307495116,
186
+ 0.020982784271240236,
187
+ 0.02126335906982422,
188
+ 0.021222400665283202,
189
+ 0.020972543716430665,
190
+ 0.020864000320434572,
191
+ 0.020743167877197266,
192
+ 0.02089369583129883,
193
+ 0.02146816062927246,
194
+ 0.02123161506652832,
195
+ 0.021181440353393553,
196
+ 0.02081279945373535,
197
+ 0.020891647338867187,
198
+ 0.0210513916015625,
199
+ 0.021132287979125978,
200
+ 0.020974592208862306,
201
+ 0.020703231811523438,
202
+ 0.020632575988769532,
203
+ 0.020570112228393556,
204
+ 0.020771839141845702,
205
+ 0.021409791946411134,
206
+ 0.021580799102783203,
207
+ 0.020961280822753905,
208
+ 0.02089574432373047,
209
+ 0.02101759910583496,
210
+ 0.0208855037689209,
211
+ 0.02071347236633301,
212
+ 0.021361663818359376,
213
+ 0.021392383575439454,
214
+ 0.02101043128967285,
215
+ 0.021163007736206055,
216
+ 0.02067967987060547,
217
+ 0.021167104721069335,
218
+ 0.020644863128662108,
219
+ 0.020971519470214844,
220
+ 0.021176319122314453,
221
+ 0.02223411178588867,
222
+ 0.02451968002319336,
223
+ 0.021395456314086913,
224
+ 0.02105548858642578,
225
+ 0.020931583404541015,
226
+ 0.02082713508605957,
227
+ 0.020758527755737305,
228
+ 0.02107084846496582,
229
+ 0.020722688674926756,
230
+ 0.020327423095703127,
231
+ 0.020794368743896483,
232
+ 0.02082815933227539,
233
+ 0.02082611274719238,
234
+ 0.02083839988708496,
235
+ 0.020709375381469726,
236
+ 0.020758527755737305,
237
+ 0.020716543197631835,
238
+ 0.020755456924438476,
239
+ 0.020714496612548827,
240
+ 0.020739072799682616,
241
+ 0.02062131118774414,
242
+ 0.020801536560058592,
243
+ 0.02025267219543457,
244
+ 0.020739072799682616,
245
+ 0.02083839988708496,
246
+ 0.02081177520751953,
247
+ 0.02072166442871094,
248
+ 0.02066227149963379,
249
+ 0.020196352005004883,
250
+ 0.019991552352905274,
251
+ 0.01999564743041992,
252
+ 0.019998720169067383,
253
+ 0.02008678436279297,
254
+ 0.020344831466674804,
255
+ 0.020782079696655274,
256
+ 0.020695039749145508,
257
+ 0.02030182456970215,
258
+ 0.0202608642578125,
259
+ 0.020701183319091796,
260
+ 0.019976192474365235,
261
+ 0.02028441619873047,
262
+ 0.020746240615844725,
263
+ 0.020685823440551757,
264
+ 0.020612096786499022,
265
+ 0.02063871955871582,
266
+ 0.02087014389038086,
267
+ 0.019937280654907227,
268
+ 0.019949567794799804,
269
+ 0.019953664779663087,
270
+ 0.019940351486206053,
271
+ 0.020728832244873048,
272
+ 0.020632575988769532,
273
+ 0.0206878719329834,
274
+ 0.02061516761779785,
275
+ 0.02063974380493164,
276
+ 0.019948543548583983,
277
+ 0.020717567443847656,
278
+ 0.020714496612548827,
279
+ 0.02068889617919922,
280
+ 0.020726783752441406,
281
+ 0.02066022491455078,
282
+ 0.020603904724121092,
283
+ 0.020617216110229493,
284
+ 0.020617216110229493,
285
+ 0.020273151397705077,
286
+ 0.02082815933227539,
287
+ 0.020702207565307617,
288
+ 0.020653055191040038,
289
+ 0.019981311798095702,
290
+ 0.019960832595825196,
291
+ 0.020455423355102538,
292
+ 0.02007756805419922,
293
+ 0.01982054328918457,
294
+ 0.020579328536987306,
295
+ 0.020322303771972656,
296
+ 0.020708351135253905,
297
+ 0.020668415069580077,
298
+ 0.02072166442871094,
299
+ 0.021144575119018554,
300
+ 0.021774335861206053,
301
+ 0.02085478401184082,
302
+ 0.020724735260009765,
303
+ 0.02065920066833496,
304
+ 0.02035916709899902,
305
+ 0.02011955261230469,
306
+ 0.04295475387573242,
307
+ 0.020708351135253905,
308
+ 0.019990528106689453,
309
+ 0.019962879180908204,
310
+ 0.020368383407592772,
311
+ 0.0208721923828125,
312
+ 0.02165350341796875,
313
+ 0.02107084846496582,
314
+ 0.020855808258056642,
315
+ 0.020730880737304686,
316
+ 0.020884479522705078,
317
+ 0.020740095138549804,
318
+ 0.020786176681518553,
319
+ 0.020734975814819336,
320
+ 0.020716543197631835,
321
+ 0.020745216369628908,
322
+ 0.020724735260009765,
323
+ 0.020797439575195312,
324
+ 0.02026188850402832,
325
+ 0.020767744064331056,
326
+ 0.020502527236938475,
327
+ 0.020767744064331056,
328
+ 0.02068992042541504,
329
+ 0.02082508850097656,
330
+ 0.02068070411682129,
331
+ 0.02004991912841797,
332
+ 0.02003763198852539,
333
+ 0.020716543197631835,
334
+ 0.020716543197631835,
335
+ 0.020710399627685547,
336
+ 0.020730880737304686,
337
+ 0.020674560546875,
338
+ 0.02023423957824707,
339
+ 0.02072985649108887,
340
+ 0.02068070411682129,
341
+ 0.020832256317138673,
342
+ 0.020892671585083008,
343
+ 0.020739072799682616,
344
+ 0.020015104293823242,
345
+ 0.019982336044311523,
346
+ 0.019965951919555663,
347
+ 0.019992576599121094,
348
+ 0.019947519302368166,
349
+ 0.02070425605773926,
350
+ 0.020677631378173827,
351
+ 0.02068992042541504,
352
+ 0.020695039749145508,
353
+ 0.02064691162109375,
354
+ 0.02000588798522949,
355
+ 0.020946943283081054,
356
+ 0.020817920684814452,
357
+ 0.020716543197631835,
358
+ 0.020719615936279297,
359
+ 0.020744192123413087,
360
+ 0.019956735610961913,
361
+ 0.019956735610961913,
362
+ 0.020151296615600587,
363
+ 0.020695039749145508,
364
+ 0.019978239059448243,
365
+ 0.020025344848632814,
366
+ 0.02002739143371582,
367
+ 0.020912128448486327,
368
+ 0.020751359939575196,
369
+ 0.021551103591918946,
370
+ 0.020757503509521484,
371
+ 0.020754432678222655,
372
+ 0.02068070411682129,
373
+ 0.02090188789367676,
374
+ 0.020797439575195312,
375
+ 0.020725759506225586,
376
+ 0.020625408172607423,
377
+ 0.0206878719329834,
378
+ 0.0208035831451416,
379
+ 0.020732927322387695,
380
+ 0.02066534423828125,
381
+ 0.020711423873901368,
382
+ 0.020676607131958007,
383
+ 0.020807680130004884,
384
+ 0.020740095138549804,
385
+ 0.020807680130004884,
386
+ 0.020742143630981445,
387
+ 0.020723712921142577,
388
+ 0.020755456924438476,
389
+ 0.02080460739135742,
390
+ 0.020802560806274413,
391
+ 0.02086195182800293,
392
+ 0.020714496612548827,
393
+ 0.020738048553466795,
394
+ 0.020720640182495118,
395
+ 0.020718591690063477,
396
+ 0.020718591690063477,
397
+ 0.020011007308959963,
398
+ 0.020006912231445313,
399
+ 0.019958784103393554,
400
+ 0.019991552352905274,
401
+ 0.019952640533447266,
402
+ 0.019936256408691407,
403
+ 0.020339712142944336,
404
+ 0.020789247512817383,
405
+ 0.020711423873901368,
406
+ 0.02081279945373535,
407
+ 0.020736000061035157,
408
+ 0.02068889617919922,
409
+ 0.020039680480957032,
410
+ 0.020188159942626953,
411
+ 0.0200447998046875,
412
+ 0.01997209548950195,
413
+ 0.019960832595825196,
414
+ 0.020025344848632814,
415
+ 0.020165632247924805,
416
+ 0.020776960372924806,
417
+ 0.02082099151611328,
418
+ 0.020731903076171874,
419
+ 0.020715520858764647,
420
+ 0.020684799194335936,
421
+ 0.020039680480957032,
422
+ 0.020714496612548827,
423
+ 0.020782079696655274,
424
+ 0.01992192077636719,
425
+ 0.019974143981933593,
426
+ 0.019974143981933593,
427
+ 0.020953088760375976,
428
+ 0.020746240615844725,
429
+ 0.020777984619140624,
430
+ 0.02002124786376953,
431
+ 0.019961856842041017,
432
+ 0.019958784103393554
433
+ ]
434
+ },
435
+ "throughput": {
436
+ "unit": "tokens/s",
437
+ "value": 48.21097693232598
438
+ },
439
+ "energy": null,
440
+ "efficiency": null
441
+ }
442
+ }
443
+ }
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=False/benchmark.log ADDED
File without changes
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=False/benchmark_config.json ADDED
@@ -0,0 +1,102 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "pytorch_generate",
3
+ "backend": {
4
+ "name": "pytorch",
5
+ "version": "2.3.0+cu121",
6
+ "_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
7
+ "task": "text-generation",
8
+ "library": "transformers",
9
+ "model": "google/gemma-2b",
10
+ "processor": "google/gemma-2b",
11
+ "device": "cuda",
12
+ "device_ids": "0",
13
+ "seed": 42,
14
+ "inter_op_num_threads": null,
15
+ "intra_op_num_threads": null,
16
+ "model_kwargs": {},
17
+ "processor_kwargs": {},
18
+ "hub_kwargs": {},
19
+ "no_weights": true,
20
+ "device_map": null,
21
+ "torch_dtype": "float16",
22
+ "eval_mode": true,
23
+ "to_bettertransformer": false,
24
+ "low_cpu_mem_usage": null,
25
+ "attn_implementation": null,
26
+ "cache_implementation": "static",
27
+ "autocast_enabled": false,
28
+ "autocast_dtype": null,
29
+ "torch_compile": false,
30
+ "torch_compile_target": "forward",
31
+ "torch_compile_config": {
32
+ "backend": "inductor",
33
+ "mode": "reduce-overhead",
34
+ "fullgraph": true
35
+ },
36
+ "quantization_scheme": null,
37
+ "quantization_config": {},
38
+ "deepspeed_inference": false,
39
+ "deepspeed_inference_config": {},
40
+ "peft_type": null,
41
+ "peft_config": {}
42
+ },
43
+ "scenario": {
44
+ "name": "inference",
45
+ "_target_": "optimum_benchmark.scenarios.inference.scenario.InferenceScenario",
46
+ "iterations": 2,
47
+ "duration": 0,
48
+ "warmup_runs": 10,
49
+ "input_shapes": {
50
+ "batch_size": 1,
51
+ "num_choices": 2,
52
+ "sequence_length": 7
53
+ },
54
+ "new_tokens": null,
55
+ "latency": true,
56
+ "memory": true,
57
+ "energy": false,
58
+ "forward_kwargs": {},
59
+ "generate_kwargs": {
60
+ "max_new_tokens": 128,
61
+ "min_new_tokens": 128,
62
+ "do_sample": false
63
+ },
64
+ "call_kwargs": {}
65
+ },
66
+ "launcher": {
67
+ "name": "process",
68
+ "_target_": "optimum_benchmark.launchers.process.launcher.ProcessLauncher",
69
+ "device_isolation": true,
70
+ "device_isolation_action": "warn",
71
+ "start_method": "spawn"
72
+ },
73
+ "environment": {
74
+ "cpu": " AMD EPYC 7R32",
75
+ "cpu_count": 16,
76
+ "cpu_ram_mb": 66697.29792,
77
+ "system": "Linux",
78
+ "machine": "x86_64",
79
+ "platform": "Linux-5.10.216-204.855.amzn2.x86_64-x86_64-with-glibc2.29",
80
+ "processor": "x86_64",
81
+ "python_version": "3.8.10",
82
+ "gpu": [
83
+ "NVIDIA A10G"
84
+ ],
85
+ "gpu_count": 1,
86
+ "gpu_vram_mb": 24146608128,
87
+ "optimum_benchmark_version": "0.2.1",
88
+ "optimum_benchmark_commit": null,
89
+ "transformers_version": "4.42.0.dev0",
90
+ "transformers_commit": "a51a3fbc4827db26e896292b6409246fe8152cd4",
91
+ "accelerate_version": "0.31.0.dev0",
92
+ "accelerate_commit": null,
93
+ "diffusers_version": null,
94
+ "diffusers_commit": null,
95
+ "optimum_version": "1.21.0.dev0",
96
+ "optimum_commit": null,
97
+ "timm_version": "0.9.16",
98
+ "timm_commit": null,
99
+ "peft_version": "0.11.2.dev0",
100
+ "peft_commit": null
101
+ }
102
+ }
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=False/benchmark_report.json ADDED
@@ -0,0 +1,339 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "prefill": {
3
+ "memory": {
4
+ "unit": "MB",
5
+ "max_ram": 1753.206784,
6
+ "max_global_vram": 6793.199616,
7
+ "max_process_vram": 0.0,
8
+ "max_reserved": 6146.752512,
9
+ "max_allocated": 5035.912192
10
+ },
11
+ "latency": {
12
+ "unit": "s",
13
+ "count": 2,
14
+ "total": 0.04486032104492188,
15
+ "mean": 0.02243016052246094,
16
+ "stdev": 0.0003491039276123046,
17
+ "p50": 0.02243016052246094,
18
+ "p90": 0.022709443664550782,
19
+ "p95": 0.022744354057312014,
20
+ "p99": 0.022772282371520996,
21
+ "values": [
22
+ 0.022779264450073243,
23
+ 0.022081056594848634
24
+ ]
25
+ },
26
+ "throughput": {
27
+ "unit": "tokens/s",
28
+ "value": 312.0797995623078
29
+ },
30
+ "energy": null,
31
+ "efficiency": null
32
+ },
33
+ "decode": {
34
+ "memory": {
35
+ "unit": "MB",
36
+ "max_ram": 1753.227264,
37
+ "max_global_vram": 6793.199616,
38
+ "max_process_vram": 0.0,
39
+ "max_reserved": 6146.752512,
40
+ "max_allocated": 5035.912704
41
+ },
42
+ "latency": {
43
+ "unit": "s",
44
+ "count": 2,
45
+ "total": 5.246615478515625,
46
+ "mean": 2.6233077392578124,
47
+ "stdev": 0.01454187011718755,
48
+ "p50": 2.6233077392578124,
49
+ "p90": 2.634941235351562,
50
+ "p95": 2.636395422363281,
51
+ "p99": 2.6375587719726563,
52
+ "values": [
53
+ 2.637849609375,
54
+ 2.608765869140625
55
+ ]
56
+ },
57
+ "throughput": {
58
+ "unit": "tokens/s",
59
+ "value": 48.4121622863549
60
+ },
61
+ "energy": null,
62
+ "efficiency": null
63
+ },
64
+ "per_token": {
65
+ "memory": null,
66
+ "latency": {
67
+ "unit": "s",
68
+ "count": 253,
69
+ "total": 5.24776754379273,
70
+ "mean": 0.020742164204714326,
71
+ "stdev": 0.0014777911921905315,
72
+ "p50": 0.020716543197631835,
73
+ "p90": 0.02107084846496582,
74
+ "p95": 0.021373951721191407,
75
+ "p99": 0.021995028305053704,
76
+ "values": [
77
+ 0.021323776245117186,
78
+ 0.021770240783691407,
79
+ 0.021406719207763672,
80
+ 0.02088243293762207,
81
+ 0.02087116813659668,
82
+ 0.021331968307495116,
83
+ 0.020982784271240236,
84
+ 0.02126335906982422,
85
+ 0.021222400665283202,
86
+ 0.020972543716430665,
87
+ 0.020864000320434572,
88
+ 0.020743167877197266,
89
+ 0.02089369583129883,
90
+ 0.02146816062927246,
91
+ 0.02123161506652832,
92
+ 0.021181440353393553,
93
+ 0.02081279945373535,
94
+ 0.020891647338867187,
95
+ 0.0210513916015625,
96
+ 0.021132287979125978,
97
+ 0.020974592208862306,
98
+ 0.020703231811523438,
99
+ 0.020632575988769532,
100
+ 0.020570112228393556,
101
+ 0.020771839141845702,
102
+ 0.021409791946411134,
103
+ 0.021580799102783203,
104
+ 0.020961280822753905,
105
+ 0.02089574432373047,
106
+ 0.02101759910583496,
107
+ 0.0208855037689209,
108
+ 0.02071347236633301,
109
+ 0.021361663818359376,
110
+ 0.021392383575439454,
111
+ 0.02101043128967285,
112
+ 0.021163007736206055,
113
+ 0.02067967987060547,
114
+ 0.021167104721069335,
115
+ 0.020644863128662108,
116
+ 0.020971519470214844,
117
+ 0.021176319122314453,
118
+ 0.02223411178588867,
119
+ 0.02451968002319336,
120
+ 0.021395456314086913,
121
+ 0.02105548858642578,
122
+ 0.020931583404541015,
123
+ 0.02082713508605957,
124
+ 0.020758527755737305,
125
+ 0.02107084846496582,
126
+ 0.020722688674926756,
127
+ 0.020327423095703127,
128
+ 0.020794368743896483,
129
+ 0.02082815933227539,
130
+ 0.02082611274719238,
131
+ 0.02083839988708496,
132
+ 0.020709375381469726,
133
+ 0.020758527755737305,
134
+ 0.020716543197631835,
135
+ 0.020755456924438476,
136
+ 0.020714496612548827,
137
+ 0.020739072799682616,
138
+ 0.02062131118774414,
139
+ 0.020801536560058592,
140
+ 0.02025267219543457,
141
+ 0.020739072799682616,
142
+ 0.02083839988708496,
143
+ 0.02081177520751953,
144
+ 0.02072166442871094,
145
+ 0.02066227149963379,
146
+ 0.020196352005004883,
147
+ 0.019991552352905274,
148
+ 0.01999564743041992,
149
+ 0.019998720169067383,
150
+ 0.02008678436279297,
151
+ 0.020344831466674804,
152
+ 0.020782079696655274,
153
+ 0.020695039749145508,
154
+ 0.02030182456970215,
155
+ 0.0202608642578125,
156
+ 0.020701183319091796,
157
+ 0.019976192474365235,
158
+ 0.02028441619873047,
159
+ 0.020746240615844725,
160
+ 0.020685823440551757,
161
+ 0.020612096786499022,
162
+ 0.02063871955871582,
163
+ 0.02087014389038086,
164
+ 0.019937280654907227,
165
+ 0.019949567794799804,
166
+ 0.019953664779663087,
167
+ 0.019940351486206053,
168
+ 0.020728832244873048,
169
+ 0.020632575988769532,
170
+ 0.0206878719329834,
171
+ 0.02061516761779785,
172
+ 0.02063974380493164,
173
+ 0.019948543548583983,
174
+ 0.020717567443847656,
175
+ 0.020714496612548827,
176
+ 0.02068889617919922,
177
+ 0.020726783752441406,
178
+ 0.02066022491455078,
179
+ 0.020603904724121092,
180
+ 0.020617216110229493,
181
+ 0.020617216110229493,
182
+ 0.020273151397705077,
183
+ 0.02082815933227539,
184
+ 0.020702207565307617,
185
+ 0.020653055191040038,
186
+ 0.019981311798095702,
187
+ 0.019960832595825196,
188
+ 0.020455423355102538,
189
+ 0.02007756805419922,
190
+ 0.01982054328918457,
191
+ 0.020579328536987306,
192
+ 0.020322303771972656,
193
+ 0.020708351135253905,
194
+ 0.020668415069580077,
195
+ 0.02072166442871094,
196
+ 0.021144575119018554,
197
+ 0.021774335861206053,
198
+ 0.02085478401184082,
199
+ 0.020724735260009765,
200
+ 0.02065920066833496,
201
+ 0.02035916709899902,
202
+ 0.02011955261230469,
203
+ 0.04295475387573242,
204
+ 0.020708351135253905,
205
+ 0.019990528106689453,
206
+ 0.019962879180908204,
207
+ 0.020368383407592772,
208
+ 0.0208721923828125,
209
+ 0.02165350341796875,
210
+ 0.02107084846496582,
211
+ 0.020855808258056642,
212
+ 0.020730880737304686,
213
+ 0.020884479522705078,
214
+ 0.020740095138549804,
215
+ 0.020786176681518553,
216
+ 0.020734975814819336,
217
+ 0.020716543197631835,
218
+ 0.020745216369628908,
219
+ 0.020724735260009765,
220
+ 0.020797439575195312,
221
+ 0.02026188850402832,
222
+ 0.020767744064331056,
223
+ 0.020502527236938475,
224
+ 0.020767744064331056,
225
+ 0.02068992042541504,
226
+ 0.02082508850097656,
227
+ 0.02068070411682129,
228
+ 0.02004991912841797,
229
+ 0.02003763198852539,
230
+ 0.020716543197631835,
231
+ 0.020716543197631835,
232
+ 0.020710399627685547,
233
+ 0.020730880737304686,
234
+ 0.020674560546875,
235
+ 0.02023423957824707,
236
+ 0.02072985649108887,
237
+ 0.02068070411682129,
238
+ 0.020832256317138673,
239
+ 0.020892671585083008,
240
+ 0.020739072799682616,
241
+ 0.020015104293823242,
242
+ 0.019982336044311523,
243
+ 0.019965951919555663,
244
+ 0.019992576599121094,
245
+ 0.019947519302368166,
246
+ 0.02070425605773926,
247
+ 0.020677631378173827,
248
+ 0.02068992042541504,
249
+ 0.020695039749145508,
250
+ 0.02064691162109375,
251
+ 0.02000588798522949,
252
+ 0.020946943283081054,
253
+ 0.020817920684814452,
254
+ 0.020716543197631835,
255
+ 0.020719615936279297,
256
+ 0.020744192123413087,
257
+ 0.019956735610961913,
258
+ 0.019956735610961913,
259
+ 0.020151296615600587,
260
+ 0.020695039749145508,
261
+ 0.019978239059448243,
262
+ 0.020025344848632814,
263
+ 0.02002739143371582,
264
+ 0.020912128448486327,
265
+ 0.020751359939575196,
266
+ 0.021551103591918946,
267
+ 0.020757503509521484,
268
+ 0.020754432678222655,
269
+ 0.02068070411682129,
270
+ 0.02090188789367676,
271
+ 0.020797439575195312,
272
+ 0.020725759506225586,
273
+ 0.020625408172607423,
274
+ 0.0206878719329834,
275
+ 0.0208035831451416,
276
+ 0.020732927322387695,
277
+ 0.02066534423828125,
278
+ 0.020711423873901368,
279
+ 0.020676607131958007,
280
+ 0.020807680130004884,
281
+ 0.020740095138549804,
282
+ 0.020807680130004884,
283
+ 0.020742143630981445,
284
+ 0.020723712921142577,
285
+ 0.020755456924438476,
286
+ 0.02080460739135742,
287
+ 0.020802560806274413,
288
+ 0.02086195182800293,
289
+ 0.020714496612548827,
290
+ 0.020738048553466795,
291
+ 0.020720640182495118,
292
+ 0.020718591690063477,
293
+ 0.020718591690063477,
294
+ 0.020011007308959963,
295
+ 0.020006912231445313,
296
+ 0.019958784103393554,
297
+ 0.019991552352905274,
298
+ 0.019952640533447266,
299
+ 0.019936256408691407,
300
+ 0.020339712142944336,
301
+ 0.020789247512817383,
302
+ 0.020711423873901368,
303
+ 0.02081279945373535,
304
+ 0.020736000061035157,
305
+ 0.02068889617919922,
306
+ 0.020039680480957032,
307
+ 0.020188159942626953,
308
+ 0.0200447998046875,
309
+ 0.01997209548950195,
310
+ 0.019960832595825196,
311
+ 0.020025344848632814,
312
+ 0.020165632247924805,
313
+ 0.020776960372924806,
314
+ 0.02082099151611328,
315
+ 0.020731903076171874,
316
+ 0.020715520858764647,
317
+ 0.020684799194335936,
318
+ 0.020039680480957032,
319
+ 0.020714496612548827,
320
+ 0.020782079696655274,
321
+ 0.01992192077636719,
322
+ 0.019974143981933593,
323
+ 0.019974143981933593,
324
+ 0.020953088760375976,
325
+ 0.020746240615844725,
326
+ 0.020777984619140624,
327
+ 0.02002124786376953,
328
+ 0.019961856842041017,
329
+ 0.019958784103393554
330
+ ]
331
+ },
332
+ "throughput": {
333
+ "unit": "tokens/s",
334
+ "value": 48.21097693232598
335
+ },
336
+ "energy": null,
337
+ "efficiency": null
338
+ }
339
+ }
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=False/summary.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model": "google/gemma-2b",
3
+ "commit": "a51a3fbc4827db26e896292b6409246fe8152cd4",
4
+ "config": "backend.cache_implementation=static,backend.torch_compile=False",
5
+ "metrics": {
6
+ "prefill.latency.mean": 0.02243016052246094,
7
+ "prefill.throughput.value": 312.0797995623078,
8
+ "decode.latency.mean": 2.6233077392578124,
9
+ "decode.throughput.value": 48.4121622863549,
10
+ "per_token.latency.mean": 0.020742164204714326,
11
+ "per_token.throughput.value": 48.21097693232598
12
+ }
13
+ }
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=True/.hydra/config.yaml ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ name: pytorch_generate
2
+ backend:
3
+ name: pytorch
4
+ version: 2.3.0+cu121
5
+ _target_: optimum_benchmark.backends.pytorch.backend.PyTorchBackend
6
+ task: null
7
+ library: null
8
+ model: google/gemma-2b
9
+ processor: null
10
+ device: cuda
11
+ device_ids: '0'
12
+ seed: 42
13
+ inter_op_num_threads: null
14
+ intra_op_num_threads: null
15
+ model_kwargs: {}
16
+ processor_kwargs: {}
17
+ hub_kwargs: {}
18
+ no_weights: true
19
+ device_map: null
20
+ torch_dtype: float16
21
+ eval_mode: true
22
+ to_bettertransformer: false
23
+ low_cpu_mem_usage: null
24
+ attn_implementation: null
25
+ cache_implementation: static
26
+ autocast_enabled: false
27
+ autocast_dtype: null
28
+ torch_compile: true
29
+ torch_compile_target: forward
30
+ torch_compile_config:
31
+ backend: inductor
32
+ mode: reduce-overhead
33
+ fullgraph: true
34
+ quantization_scheme: null
35
+ quantization_config: {}
36
+ deepspeed_inference: false
37
+ deepspeed_inference_config: {}
38
+ peft_type: null
39
+ peft_config: {}
40
+ scenario:
41
+ name: inference
42
+ _target_: optimum_benchmark.scenarios.inference.scenario.InferenceScenario
43
+ iterations: 2
44
+ duration: 0
45
+ warmup_runs: 10
46
+ input_shapes:
47
+ batch_size: 1
48
+ sequence_length: 7
49
+ new_tokens: null
50
+ latency: true
51
+ memory: true
52
+ energy: false
53
+ forward_kwargs: {}
54
+ generate_kwargs:
55
+ max_new_tokens: 128
56
+ min_new_tokens: 128
57
+ do_sample: false
58
+ call_kwargs: {}
59
+ launcher:
60
+ name: process
61
+ _target_: optimum_benchmark.launchers.process.launcher.ProcessLauncher
62
+ device_isolation: true
63
+ device_isolation_action: warn
64
+ start_method: spawn
65
+ environment:
66
+ cpu: ' AMD EPYC 7R32'
67
+ cpu_count: 16
68
+ cpu_ram_mb: 66697.29792
69
+ system: Linux
70
+ machine: x86_64
71
+ platform: Linux-5.10.216-204.855.amzn2.x86_64-x86_64-with-glibc2.29
72
+ processor: x86_64
73
+ python_version: 3.8.10
74
+ gpu:
75
+ - NVIDIA A10G
76
+ gpu_count: 1
77
+ gpu_vram_mb: 24146608128
78
+ optimum_benchmark_version: 0.2.1
79
+ optimum_benchmark_commit: null
80
+ transformers_version: 4.42.0.dev0
81
+ transformers_commit: a51a3fbc4827db26e896292b6409246fe8152cd4
82
+ accelerate_version: 0.31.0.dev0
83
+ accelerate_commit: null
84
+ diffusers_version: null
85
+ diffusers_commit: null
86
+ optimum_version: 1.21.0.dev0
87
+ optimum_commit: null
88
+ timm_version: 0.9.16
89
+ timm_commit: null
90
+ peft_version: 0.11.2.dev0
91
+ peft_commit: null
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=True/.hydra/hydra.yaml ADDED
@@ -0,0 +1,141 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ hydra:
2
+ run:
3
+ dir: _benchmark/tmpyph4ye2g/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/${hydra.job.override_dirname}
4
+ sweep:
5
+ dir: _benchmark/tmpyph4ye2g/commit=a51a3fbc4827db26e896292b6409246fe8152cd4
6
+ subdir: ${hydra.job.override_dirname}
7
+ launcher:
8
+ _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
9
+ sweeper:
10
+ _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
11
+ max_batch_size: null
12
+ params: null
13
+ help:
14
+ app_name: ${hydra.job.name}
15
+ header: '${hydra.help.app_name} is powered by Hydra.
16
+
17
+ '
18
+ footer: 'Powered by Hydra (https://hydra.cc)
19
+
20
+ Use --hydra-help to view Hydra specific help
21
+
22
+ '
23
+ template: '${hydra.help.header}
24
+
25
+ == Configuration groups ==
26
+
27
+ Compose your configuration from those groups (group=option)
28
+
29
+
30
+ $APP_CONFIG_GROUPS
31
+
32
+
33
+ == Config ==
34
+
35
+ Override anything in the config (foo.bar=value)
36
+
37
+
38
+ $CONFIG
39
+
40
+
41
+ ${hydra.help.footer}
42
+
43
+ '
44
+ hydra_help:
45
+ template: 'Hydra (${hydra.runtime.version})
46
+
47
+ See https://hydra.cc for more info.
48
+
49
+
50
+ == Flags ==
51
+
52
+ $FLAGS_HELP
53
+
54
+
55
+ == Configuration groups ==
56
+
57
+ Compose your configuration from those groups (For example, append hydra/job_logging=disabled
58
+ to command line)
59
+
60
+
61
+ $HYDRA_CONFIG_GROUPS
62
+
63
+
64
+ Use ''--cfg hydra'' to Show the Hydra config.
65
+
66
+ '
67
+ hydra_help: ???
68
+ hydra_logging:
69
+ version: 1
70
+ root:
71
+ level: ERROR
72
+ disable_existing_loggers: true
73
+ job_logging:
74
+ version: 1
75
+ root:
76
+ level: ERROR
77
+ disable_existing_loggers: true
78
+ env: {}
79
+ mode: MULTIRUN
80
+ searchpath: []
81
+ callbacks: {}
82
+ output_subdir: .hydra
83
+ overrides:
84
+ hydra:
85
+ - hydra/job_logging=disabled
86
+ - hydra/hydra_logging=disabled
87
+ - hydra.sweep.dir=_benchmark/tmpyph4ye2g/commit\=a51a3fbc4827db26e896292b6409246fe8152cd4
88
+ - hydra.run.dir=_benchmark/tmpyph4ye2g/commit\=a51a3fbc4827db26e896292b6409246fe8152cd4/${hydra.job.override_dirname}
89
+ - hydra.mode=MULTIRUN
90
+ task:
91
+ - backend.model=google/gemma-2b
92
+ - backend.cache_implementation=static
93
+ - backend.torch_compile=True
94
+ job:
95
+ name: cli
96
+ chdir: true
97
+ override_dirname: backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=True
98
+ id: '3'
99
+ num: 3
100
+ config_name: generation
101
+ env_set:
102
+ OVERRIDE_BENCHMARKS: '1'
103
+ LOG_LEVEL: WARN
104
+ env_copy: []
105
+ config:
106
+ override_dirname:
107
+ kv_sep: '='
108
+ item_sep: ','
109
+ exclude_keys: []
110
+ runtime:
111
+ version: 1.3.2
112
+ version_base: '1.3'
113
+ cwd: /transformers
114
+ config_sources:
115
+ - path: hydra.conf
116
+ schema: pkg
117
+ provider: hydra
118
+ - path: optimum_benchmark
119
+ schema: pkg
120
+ provider: main
121
+ - path: /transformers/benchmark/config
122
+ schema: file
123
+ provider: command-line
124
+ - path: ''
125
+ schema: structured
126
+ provider: schema
127
+ output_dir: /transformers/_benchmark/tmpyph4ye2g/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=True
128
+ choices:
129
+ backend: pytorch
130
+ launcher: process
131
+ scenario: inference
132
+ hydra/env: default
133
+ hydra/callbacks: null
134
+ hydra/job_logging: disabled
135
+ hydra/hydra_logging: disabled
136
+ hydra/hydra_help: default
137
+ hydra/help: default
138
+ hydra/sweeper: basic
139
+ hydra/launcher: basic
140
+ hydra/output: default
141
+ verbose: false
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=True/.hydra/overrides.yaml ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ - backend.model=google/gemma-2b
2
+ - backend.cache_implementation=static
3
+ - backend.torch_compile=True
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=True/benchmark.json ADDED
@@ -0,0 +1,443 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "config": {
3
+ "name": "pytorch_generate",
4
+ "backend": {
5
+ "name": "pytorch",
6
+ "version": "2.3.0+cu121",
7
+ "_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
8
+ "task": "text-generation",
9
+ "library": "transformers",
10
+ "model": "google/gemma-2b",
11
+ "processor": "google/gemma-2b",
12
+ "device": "cuda",
13
+ "device_ids": "0",
14
+ "seed": 42,
15
+ "inter_op_num_threads": null,
16
+ "intra_op_num_threads": null,
17
+ "model_kwargs": {},
18
+ "processor_kwargs": {},
19
+ "hub_kwargs": {},
20
+ "no_weights": true,
21
+ "device_map": null,
22
+ "torch_dtype": "float16",
23
+ "eval_mode": true,
24
+ "to_bettertransformer": false,
25
+ "low_cpu_mem_usage": null,
26
+ "attn_implementation": null,
27
+ "cache_implementation": "static",
28
+ "autocast_enabled": false,
29
+ "autocast_dtype": null,
30
+ "torch_compile": true,
31
+ "torch_compile_target": "forward",
32
+ "torch_compile_config": {
33
+ "backend": "inductor",
34
+ "mode": "reduce-overhead",
35
+ "fullgraph": true
36
+ },
37
+ "quantization_scheme": null,
38
+ "quantization_config": {},
39
+ "deepspeed_inference": false,
40
+ "deepspeed_inference_config": {},
41
+ "peft_type": null,
42
+ "peft_config": {}
43
+ },
44
+ "scenario": {
45
+ "name": "inference",
46
+ "_target_": "optimum_benchmark.scenarios.inference.scenario.InferenceScenario",
47
+ "iterations": 2,
48
+ "duration": 0,
49
+ "warmup_runs": 10,
50
+ "input_shapes": {
51
+ "batch_size": 1,
52
+ "num_choices": 2,
53
+ "sequence_length": 7
54
+ },
55
+ "new_tokens": null,
56
+ "latency": true,
57
+ "memory": true,
58
+ "energy": false,
59
+ "forward_kwargs": {},
60
+ "generate_kwargs": {
61
+ "max_new_tokens": 128,
62
+ "min_new_tokens": 128,
63
+ "do_sample": false
64
+ },
65
+ "call_kwargs": {}
66
+ },
67
+ "launcher": {
68
+ "name": "process",
69
+ "_target_": "optimum_benchmark.launchers.process.launcher.ProcessLauncher",
70
+ "device_isolation": true,
71
+ "device_isolation_action": "warn",
72
+ "start_method": "spawn"
73
+ },
74
+ "environment": {
75
+ "cpu": " AMD EPYC 7R32",
76
+ "cpu_count": 16,
77
+ "cpu_ram_mb": 66697.29792,
78
+ "system": "Linux",
79
+ "machine": "x86_64",
80
+ "platform": "Linux-5.10.216-204.855.amzn2.x86_64-x86_64-with-glibc2.29",
81
+ "processor": "x86_64",
82
+ "python_version": "3.8.10",
83
+ "gpu": [
84
+ "NVIDIA A10G"
85
+ ],
86
+ "gpu_count": 1,
87
+ "gpu_vram_mb": 24146608128,
88
+ "optimum_benchmark_version": "0.2.1",
89
+ "optimum_benchmark_commit": null,
90
+ "transformers_version": "4.42.0.dev0",
91
+ "transformers_commit": "a51a3fbc4827db26e896292b6409246fe8152cd4",
92
+ "accelerate_version": "0.31.0.dev0",
93
+ "accelerate_commit": null,
94
+ "diffusers_version": null,
95
+ "diffusers_commit": null,
96
+ "optimum_version": "1.21.0.dev0",
97
+ "optimum_commit": null,
98
+ "timm_version": "0.9.16",
99
+ "timm_commit": null,
100
+ "peft_version": "0.11.2.dev0",
101
+ "peft_commit": null
102
+ }
103
+ },
104
+ "report": {
105
+ "prefill": {
106
+ "memory": {
107
+ "unit": "MB",
108
+ "max_ram": 1917.6448,
109
+ "max_global_vram": 6094.848,
110
+ "max_process_vram": 0.0,
111
+ "max_reserved": 5379.19488,
112
+ "max_allocated": 5020.227584
113
+ },
114
+ "latency": {
115
+ "unit": "s",
116
+ "count": 2,
117
+ "total": 0.02828486347198486,
118
+ "mean": 0.01414243173599243,
119
+ "stdev": 0.00024083185195922837,
120
+ "p50": 0.01414243173599243,
121
+ "p90": 0.014335097217559814,
122
+ "p95": 0.014359180402755738,
123
+ "p99": 0.014378446950912475,
124
+ "values": [
125
+ 0.01438326358795166,
126
+ 0.013901599884033203
127
+ ]
128
+ },
129
+ "throughput": {
130
+ "unit": "tokens/s",
131
+ "value": 494.96438311842996
132
+ },
133
+ "energy": null,
134
+ "efficiency": null
135
+ },
136
+ "decode": {
137
+ "memory": {
138
+ "unit": "MB",
139
+ "max_ram": 2261.282816,
140
+ "max_global_vram": 8902.934528,
141
+ "max_process_vram": 0.0,
142
+ "max_reserved": 5379.19488,
143
+ "max_allocated": 5026.065408
144
+ },
145
+ "latency": {
146
+ "unit": "s",
147
+ "count": 2,
148
+ "total": 3.1440765380859377,
149
+ "mean": 1.5720382690429688,
150
+ "stdev": 0.016047058105468692,
151
+ "p50": 1.5720382690429688,
152
+ "p90": 1.5848759155273437,
153
+ "p95": 1.5864806213378906,
154
+ "p99": 1.587764385986328,
155
+ "values": [
156
+ 1.5880853271484374,
157
+ 1.5559912109375
158
+ ]
159
+ },
160
+ "throughput": {
161
+ "unit": "tokens/s",
162
+ "value": 80.78683738234663
163
+ },
164
+ "energy": null,
165
+ "efficiency": null
166
+ },
167
+ "per_token": {
168
+ "memory": null,
169
+ "latency": {
170
+ "unit": "s",
171
+ "count": 253,
172
+ "total": 3.145101314544681,
173
+ "mean": 0.012431230492271454,
174
+ "stdev": 0.0009004109051801406,
175
+ "p50": 0.012269503593444824,
176
+ "p90": 0.012711731147766114,
177
+ "p95": 0.012742655754089355,
178
+ "p99": 0.012799139900207519,
179
+ "values": [
180
+ 0.012650495529174804,
181
+ 0.01268838405609131,
182
+ 0.01267302417755127,
183
+ 0.012722175598144531,
184
+ 0.012668928146362305,
185
+ 0.012759039878845215,
186
+ 0.012886015892028809,
187
+ 0.012727295875549317,
188
+ 0.012696576118469239,
189
+ 0.01274675178527832,
190
+ 0.012687359809875488,
191
+ 0.012735487937927246,
192
+ 0.012660736083984376,
193
+ 0.012703743934631348,
194
+ 0.012687359809875488,
195
+ 0.012703743934631348,
196
+ 0.012710911750793457,
197
+ 0.012724224090576173,
198
+ 0.01267199993133545,
199
+ 0.01274675178527832,
200
+ 0.012690431594848632,
201
+ 0.012777471542358398,
202
+ 0.012693504333496093,
203
+ 0.012747776031494141,
204
+ 0.012677120208740235,
205
+ 0.012717056274414062,
206
+ 0.012650495529174804,
207
+ 0.01278771209716797,
208
+ 0.012727295875549317,
209
+ 0.012790783882141114,
210
+ 0.012695551872253418,
211
+ 0.012742655754089355,
212
+ 0.012650495529174804,
213
+ 0.01268838405609131,
214
+ 0.01266585636138916,
215
+ 0.012603391647338867,
216
+ 0.012642304420471191,
217
+ 0.012694560050964355,
218
+ 0.01262384033203125,
219
+ 0.012711935997009278,
220
+ 0.012589056015014649,
221
+ 0.012701696395874023,
222
+ 0.012691455841064453,
223
+ 0.012734463691711426,
224
+ 0.012661760330200195,
225
+ 0.01275391960144043,
226
+ 0.012683263778686523,
227
+ 0.012702719688415527,
228
+ 0.012732416152954102,
229
+ 0.012808192253112792,
230
+ 0.012731391906738282,
231
+ 0.012742655754089355,
232
+ 0.012692480087280274,
233
+ 0.012724224090576173,
234
+ 0.012683263778686523,
235
+ 0.012698623657226562,
236
+ 0.01265561580657959,
237
+ 0.012682239532470703,
238
+ 0.012632063865661621,
239
+ 0.012750847816467285,
240
+ 0.012643327713012695,
241
+ 0.012698623657226562,
242
+ 0.012659711837768555,
243
+ 0.012713024139404297,
244
+ 0.012572607994079589,
245
+ 0.012608511924743653,
246
+ 0.012516351699829101,
247
+ 0.012624896049499512,
248
+ 0.012257280349731446,
249
+ 0.012431360244750977,
250
+ 0.012498944282531739,
251
+ 0.012498944282531739,
252
+ 0.012628992080688477,
253
+ 0.012311552047729492,
254
+ 0.012232704162597656,
255
+ 0.012273664474487305,
256
+ 0.012214271545410157,
257
+ 0.012273664474487305,
258
+ 0.012231679916381836,
259
+ 0.012280832290649414,
260
+ 0.012257280349731446,
261
+ 0.012304384231567383,
262
+ 0.012225536346435547,
263
+ 0.012273664474487305,
264
+ 0.012233728408813477,
265
+ 0.012291071891784668,
266
+ 0.012230655670166016,
267
+ 0.01226956844329834,
268
+ 0.012224512100219726,
269
+ 0.012270591735839843,
270
+ 0.012220416069030762,
271
+ 0.012285951614379884,
272
+ 0.012228608131408691,
273
+ 0.012275712013244629,
274
+ 0.012237824440002442,
275
+ 0.012273664474487305,
276
+ 0.0122357759475708,
277
+ 0.012288000106811523,
278
+ 0.012221440315246582,
279
+ 0.012280832290649414,
280
+ 0.012237824440002442,
281
+ 0.012267519950866699,
282
+ 0.01223475170135498,
283
+ 0.012263423919677734,
284
+ 0.012233728408813477,
285
+ 0.012264448165893555,
286
+ 0.012228608131408691,
287
+ 0.01226854419708252,
288
+ 0.01221939182281494,
289
+ 0.012270591735839843,
290
+ 0.012222463607788087,
291
+ 0.012322815895080566,
292
+ 0.012254207611083985,
293
+ 0.012275712013244629,
294
+ 0.012225536346435547,
295
+ 0.012283904075622559,
296
+ 0.012239871978759765,
297
+ 0.012263423919677734,
298
+ 0.012220416069030762,
299
+ 0.012285951614379884,
300
+ 0.012237824440002442,
301
+ 0.012279808044433594,
302
+ 0.012226559638977052,
303
+ 0.012267519950866699,
304
+ 0.012229632377624512,
305
+ 0.012265472412109376,
306
+ 0.026366975784301756,
307
+ 0.012244992256164551,
308
+ 0.012266495704650878,
309
+ 0.012233728408813477,
310
+ 0.012283904075622559,
311
+ 0.012221440315246582,
312
+ 0.012275712013244629,
313
+ 0.012247039794921874,
314
+ 0.012272640228271485,
315
+ 0.012240896224975586,
316
+ 0.012281855583190919,
317
+ 0.012232704162597656,
318
+ 0.012275712013244629,
319
+ 0.012225536346435547,
320
+ 0.012277759552001954,
321
+ 0.01221939182281494,
322
+ 0.012282879829406738,
323
+ 0.01221939182281494,
324
+ 0.012265472412109376,
325
+ 0.012224512100219726,
326
+ 0.01226137638092041,
327
+ 0.01223475170135498,
328
+ 0.012271615982055664,
329
+ 0.012239871978759765,
330
+ 0.012271615982055664,
331
+ 0.012247039794921874,
332
+ 0.012281855583190919,
333
+ 0.012247039794921874,
334
+ 0.01228492832183838,
335
+ 0.012229632377624512,
336
+ 0.012265472412109376,
337
+ 0.012225536346435547,
338
+ 0.012274687767028808,
339
+ 0.012213248252868653,
340
+ 0.012271615982055664,
341
+ 0.0122357759475708,
342
+ 0.012264448165893555,
343
+ 0.012231679916381836,
344
+ 0.012277759552001954,
345
+ 0.012226559638977052,
346
+ 0.012263423919677734,
347
+ 0.01225830364227295,
348
+ 0.012267519950866699,
349
+ 0.012223487854003906,
350
+ 0.012278783798217773,
351
+ 0.012233728408813477,
352
+ 0.01226854419708252,
353
+ 0.012222463607788087,
354
+ 0.012272640228271485,
355
+ 0.012233728408813477,
356
+ 0.012272640228271485,
357
+ 0.012236800193786621,
358
+ 0.01226854419708252,
359
+ 0.012214271545410157,
360
+ 0.01226854419708252,
361
+ 0.01221939182281494,
362
+ 0.012272640228271485,
363
+ 0.012246015548706055,
364
+ 0.012263423919677734,
365
+ 0.012228608131408691,
366
+ 0.01226035213470459,
367
+ 0.012223487854003906,
368
+ 0.012273664474487305,
369
+ 0.012217344284057617,
370
+ 0.012267519950866699,
371
+ 0.012220416069030762,
372
+ 0.012263423919677734,
373
+ 0.012224512100219726,
374
+ 0.01226956844329834,
375
+ 0.012229632377624512,
376
+ 0.012270591735839843,
377
+ 0.012228608131408691,
378
+ 0.012273664474487305,
379
+ 0.012248064041137695,
380
+ 0.012274687767028808,
381
+ 0.012223487854003906,
382
+ 0.012272640228271485,
383
+ 0.012229632377624512,
384
+ 0.012262399673461915,
385
+ 0.01223475170135498,
386
+ 0.012259327888488769,
387
+ 0.012232704162597656,
388
+ 0.01226854419708252,
389
+ 0.012229632377624512,
390
+ 0.012274687767028808,
391
+ 0.012232704162597656,
392
+ 0.01225830364227295,
393
+ 0.012229632377624512,
394
+ 0.01226854419708252,
395
+ 0.012257280349731446,
396
+ 0.012280832290649414,
397
+ 0.012231743812561036,
398
+ 0.012269503593444824,
399
+ 0.012228608131408691,
400
+ 0.012270591735839843,
401
+ 0.012228608131408691,
402
+ 0.012266495704650878,
403
+ 0.012229632377624512,
404
+ 0.01226137638092041,
405
+ 0.012226559638977052,
406
+ 0.012255231857299804,
407
+ 0.012224512100219726,
408
+ 0.012271615982055664,
409
+ 0.012229632377624512,
410
+ 0.012255231857299804,
411
+ 0.012241920471191407,
412
+ 0.012278783798217773,
413
+ 0.012222463607788087,
414
+ 0.01226854419708252,
415
+ 0.012225536346435547,
416
+ 0.012274687767028808,
417
+ 0.012224512100219726,
418
+ 0.01226854419708252,
419
+ 0.012223487854003906,
420
+ 0.01226854419708252,
421
+ 0.012218367576599122,
422
+ 0.012283904075622559,
423
+ 0.012211199760437011,
424
+ 0.012267519950866699,
425
+ 0.012218367576599122,
426
+ 0.012263423919677734,
427
+ 0.012247039794921874,
428
+ 0.012292096138000488,
429
+ 0.01222758388519287,
430
+ 0.012275712013244629,
431
+ 0.012225536346435547,
432
+ 0.012262399673461915
433
+ ]
434
+ },
435
+ "throughput": {
436
+ "unit": "tokens/s",
437
+ "value": 80.44255961802848
438
+ },
439
+ "energy": null,
440
+ "efficiency": null
441
+ }
442
+ }
443
+ }
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=True/benchmark.log ADDED
File without changes
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=True/benchmark_config.json ADDED
@@ -0,0 +1,102 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "name": "pytorch_generate",
3
+ "backend": {
4
+ "name": "pytorch",
5
+ "version": "2.3.0+cu121",
6
+ "_target_": "optimum_benchmark.backends.pytorch.backend.PyTorchBackend",
7
+ "task": "text-generation",
8
+ "library": "transformers",
9
+ "model": "google/gemma-2b",
10
+ "processor": "google/gemma-2b",
11
+ "device": "cuda",
12
+ "device_ids": "0",
13
+ "seed": 42,
14
+ "inter_op_num_threads": null,
15
+ "intra_op_num_threads": null,
16
+ "model_kwargs": {},
17
+ "processor_kwargs": {},
18
+ "hub_kwargs": {},
19
+ "no_weights": true,
20
+ "device_map": null,
21
+ "torch_dtype": "float16",
22
+ "eval_mode": true,
23
+ "to_bettertransformer": false,
24
+ "low_cpu_mem_usage": null,
25
+ "attn_implementation": null,
26
+ "cache_implementation": "static",
27
+ "autocast_enabled": false,
28
+ "autocast_dtype": null,
29
+ "torch_compile": true,
30
+ "torch_compile_target": "forward",
31
+ "torch_compile_config": {
32
+ "backend": "inductor",
33
+ "mode": "reduce-overhead",
34
+ "fullgraph": true
35
+ },
36
+ "quantization_scheme": null,
37
+ "quantization_config": {},
38
+ "deepspeed_inference": false,
39
+ "deepspeed_inference_config": {},
40
+ "peft_type": null,
41
+ "peft_config": {}
42
+ },
43
+ "scenario": {
44
+ "name": "inference",
45
+ "_target_": "optimum_benchmark.scenarios.inference.scenario.InferenceScenario",
46
+ "iterations": 2,
47
+ "duration": 0,
48
+ "warmup_runs": 10,
49
+ "input_shapes": {
50
+ "batch_size": 1,
51
+ "num_choices": 2,
52
+ "sequence_length": 7
53
+ },
54
+ "new_tokens": null,
55
+ "latency": true,
56
+ "memory": true,
57
+ "energy": false,
58
+ "forward_kwargs": {},
59
+ "generate_kwargs": {
60
+ "max_new_tokens": 128,
61
+ "min_new_tokens": 128,
62
+ "do_sample": false
63
+ },
64
+ "call_kwargs": {}
65
+ },
66
+ "launcher": {
67
+ "name": "process",
68
+ "_target_": "optimum_benchmark.launchers.process.launcher.ProcessLauncher",
69
+ "device_isolation": true,
70
+ "device_isolation_action": "warn",
71
+ "start_method": "spawn"
72
+ },
73
+ "environment": {
74
+ "cpu": " AMD EPYC 7R32",
75
+ "cpu_count": 16,
76
+ "cpu_ram_mb": 66697.29792,
77
+ "system": "Linux",
78
+ "machine": "x86_64",
79
+ "platform": "Linux-5.10.216-204.855.amzn2.x86_64-x86_64-with-glibc2.29",
80
+ "processor": "x86_64",
81
+ "python_version": "3.8.10",
82
+ "gpu": [
83
+ "NVIDIA A10G"
84
+ ],
85
+ "gpu_count": 1,
86
+ "gpu_vram_mb": 24146608128,
87
+ "optimum_benchmark_version": "0.2.1",
88
+ "optimum_benchmark_commit": null,
89
+ "transformers_version": "4.42.0.dev0",
90
+ "transformers_commit": "a51a3fbc4827db26e896292b6409246fe8152cd4",
91
+ "accelerate_version": "0.31.0.dev0",
92
+ "accelerate_commit": null,
93
+ "diffusers_version": null,
94
+ "diffusers_commit": null,
95
+ "optimum_version": "1.21.0.dev0",
96
+ "optimum_commit": null,
97
+ "timm_version": "0.9.16",
98
+ "timm_commit": null,
99
+ "peft_version": "0.11.2.dev0",
100
+ "peft_commit": null
101
+ }
102
+ }
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=True/benchmark_report.json ADDED
@@ -0,0 +1,339 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "prefill": {
3
+ "memory": {
4
+ "unit": "MB",
5
+ "max_ram": 1917.6448,
6
+ "max_global_vram": 6094.848,
7
+ "max_process_vram": 0.0,
8
+ "max_reserved": 5379.19488,
9
+ "max_allocated": 5020.227584
10
+ },
11
+ "latency": {
12
+ "unit": "s",
13
+ "count": 2,
14
+ "total": 0.02828486347198486,
15
+ "mean": 0.01414243173599243,
16
+ "stdev": 0.00024083185195922837,
17
+ "p50": 0.01414243173599243,
18
+ "p90": 0.014335097217559814,
19
+ "p95": 0.014359180402755738,
20
+ "p99": 0.014378446950912475,
21
+ "values": [
22
+ 0.01438326358795166,
23
+ 0.013901599884033203
24
+ ]
25
+ },
26
+ "throughput": {
27
+ "unit": "tokens/s",
28
+ "value": 494.96438311842996
29
+ },
30
+ "energy": null,
31
+ "efficiency": null
32
+ },
33
+ "decode": {
34
+ "memory": {
35
+ "unit": "MB",
36
+ "max_ram": 2261.282816,
37
+ "max_global_vram": 8902.934528,
38
+ "max_process_vram": 0.0,
39
+ "max_reserved": 5379.19488,
40
+ "max_allocated": 5026.065408
41
+ },
42
+ "latency": {
43
+ "unit": "s",
44
+ "count": 2,
45
+ "total": 3.1440765380859377,
46
+ "mean": 1.5720382690429688,
47
+ "stdev": 0.016047058105468692,
48
+ "p50": 1.5720382690429688,
49
+ "p90": 1.5848759155273437,
50
+ "p95": 1.5864806213378906,
51
+ "p99": 1.587764385986328,
52
+ "values": [
53
+ 1.5880853271484374,
54
+ 1.5559912109375
55
+ ]
56
+ },
57
+ "throughput": {
58
+ "unit": "tokens/s",
59
+ "value": 80.78683738234663
60
+ },
61
+ "energy": null,
62
+ "efficiency": null
63
+ },
64
+ "per_token": {
65
+ "memory": null,
66
+ "latency": {
67
+ "unit": "s",
68
+ "count": 253,
69
+ "total": 3.145101314544681,
70
+ "mean": 0.012431230492271454,
71
+ "stdev": 0.0009004109051801406,
72
+ "p50": 0.012269503593444824,
73
+ "p90": 0.012711731147766114,
74
+ "p95": 0.012742655754089355,
75
+ "p99": 0.012799139900207519,
76
+ "values": [
77
+ 0.012650495529174804,
78
+ 0.01268838405609131,
79
+ 0.01267302417755127,
80
+ 0.012722175598144531,
81
+ 0.012668928146362305,
82
+ 0.012759039878845215,
83
+ 0.012886015892028809,
84
+ 0.012727295875549317,
85
+ 0.012696576118469239,
86
+ 0.01274675178527832,
87
+ 0.012687359809875488,
88
+ 0.012735487937927246,
89
+ 0.012660736083984376,
90
+ 0.012703743934631348,
91
+ 0.012687359809875488,
92
+ 0.012703743934631348,
93
+ 0.012710911750793457,
94
+ 0.012724224090576173,
95
+ 0.01267199993133545,
96
+ 0.01274675178527832,
97
+ 0.012690431594848632,
98
+ 0.012777471542358398,
99
+ 0.012693504333496093,
100
+ 0.012747776031494141,
101
+ 0.012677120208740235,
102
+ 0.012717056274414062,
103
+ 0.012650495529174804,
104
+ 0.01278771209716797,
105
+ 0.012727295875549317,
106
+ 0.012790783882141114,
107
+ 0.012695551872253418,
108
+ 0.012742655754089355,
109
+ 0.012650495529174804,
110
+ 0.01268838405609131,
111
+ 0.01266585636138916,
112
+ 0.012603391647338867,
113
+ 0.012642304420471191,
114
+ 0.012694560050964355,
115
+ 0.01262384033203125,
116
+ 0.012711935997009278,
117
+ 0.012589056015014649,
118
+ 0.012701696395874023,
119
+ 0.012691455841064453,
120
+ 0.012734463691711426,
121
+ 0.012661760330200195,
122
+ 0.01275391960144043,
123
+ 0.012683263778686523,
124
+ 0.012702719688415527,
125
+ 0.012732416152954102,
126
+ 0.012808192253112792,
127
+ 0.012731391906738282,
128
+ 0.012742655754089355,
129
+ 0.012692480087280274,
130
+ 0.012724224090576173,
131
+ 0.012683263778686523,
132
+ 0.012698623657226562,
133
+ 0.01265561580657959,
134
+ 0.012682239532470703,
135
+ 0.012632063865661621,
136
+ 0.012750847816467285,
137
+ 0.012643327713012695,
138
+ 0.012698623657226562,
139
+ 0.012659711837768555,
140
+ 0.012713024139404297,
141
+ 0.012572607994079589,
142
+ 0.012608511924743653,
143
+ 0.012516351699829101,
144
+ 0.012624896049499512,
145
+ 0.012257280349731446,
146
+ 0.012431360244750977,
147
+ 0.012498944282531739,
148
+ 0.012498944282531739,
149
+ 0.012628992080688477,
150
+ 0.012311552047729492,
151
+ 0.012232704162597656,
152
+ 0.012273664474487305,
153
+ 0.012214271545410157,
154
+ 0.012273664474487305,
155
+ 0.012231679916381836,
156
+ 0.012280832290649414,
157
+ 0.012257280349731446,
158
+ 0.012304384231567383,
159
+ 0.012225536346435547,
160
+ 0.012273664474487305,
161
+ 0.012233728408813477,
162
+ 0.012291071891784668,
163
+ 0.012230655670166016,
164
+ 0.01226956844329834,
165
+ 0.012224512100219726,
166
+ 0.012270591735839843,
167
+ 0.012220416069030762,
168
+ 0.012285951614379884,
169
+ 0.012228608131408691,
170
+ 0.012275712013244629,
171
+ 0.012237824440002442,
172
+ 0.012273664474487305,
173
+ 0.0122357759475708,
174
+ 0.012288000106811523,
175
+ 0.012221440315246582,
176
+ 0.012280832290649414,
177
+ 0.012237824440002442,
178
+ 0.012267519950866699,
179
+ 0.01223475170135498,
180
+ 0.012263423919677734,
181
+ 0.012233728408813477,
182
+ 0.012264448165893555,
183
+ 0.012228608131408691,
184
+ 0.01226854419708252,
185
+ 0.01221939182281494,
186
+ 0.012270591735839843,
187
+ 0.012222463607788087,
188
+ 0.012322815895080566,
189
+ 0.012254207611083985,
190
+ 0.012275712013244629,
191
+ 0.012225536346435547,
192
+ 0.012283904075622559,
193
+ 0.012239871978759765,
194
+ 0.012263423919677734,
195
+ 0.012220416069030762,
196
+ 0.012285951614379884,
197
+ 0.012237824440002442,
198
+ 0.012279808044433594,
199
+ 0.012226559638977052,
200
+ 0.012267519950866699,
201
+ 0.012229632377624512,
202
+ 0.012265472412109376,
203
+ 0.026366975784301756,
204
+ 0.012244992256164551,
205
+ 0.012266495704650878,
206
+ 0.012233728408813477,
207
+ 0.012283904075622559,
208
+ 0.012221440315246582,
209
+ 0.012275712013244629,
210
+ 0.012247039794921874,
211
+ 0.012272640228271485,
212
+ 0.012240896224975586,
213
+ 0.012281855583190919,
214
+ 0.012232704162597656,
215
+ 0.012275712013244629,
216
+ 0.012225536346435547,
217
+ 0.012277759552001954,
218
+ 0.01221939182281494,
219
+ 0.012282879829406738,
220
+ 0.01221939182281494,
221
+ 0.012265472412109376,
222
+ 0.012224512100219726,
223
+ 0.01226137638092041,
224
+ 0.01223475170135498,
225
+ 0.012271615982055664,
226
+ 0.012239871978759765,
227
+ 0.012271615982055664,
228
+ 0.012247039794921874,
229
+ 0.012281855583190919,
230
+ 0.012247039794921874,
231
+ 0.01228492832183838,
232
+ 0.012229632377624512,
233
+ 0.012265472412109376,
234
+ 0.012225536346435547,
235
+ 0.012274687767028808,
236
+ 0.012213248252868653,
237
+ 0.012271615982055664,
238
+ 0.0122357759475708,
239
+ 0.012264448165893555,
240
+ 0.012231679916381836,
241
+ 0.012277759552001954,
242
+ 0.012226559638977052,
243
+ 0.012263423919677734,
244
+ 0.01225830364227295,
245
+ 0.012267519950866699,
246
+ 0.012223487854003906,
247
+ 0.012278783798217773,
248
+ 0.012233728408813477,
249
+ 0.01226854419708252,
250
+ 0.012222463607788087,
251
+ 0.012272640228271485,
252
+ 0.012233728408813477,
253
+ 0.012272640228271485,
254
+ 0.012236800193786621,
255
+ 0.01226854419708252,
256
+ 0.012214271545410157,
257
+ 0.01226854419708252,
258
+ 0.01221939182281494,
259
+ 0.012272640228271485,
260
+ 0.012246015548706055,
261
+ 0.012263423919677734,
262
+ 0.012228608131408691,
263
+ 0.01226035213470459,
264
+ 0.012223487854003906,
265
+ 0.012273664474487305,
266
+ 0.012217344284057617,
267
+ 0.012267519950866699,
268
+ 0.012220416069030762,
269
+ 0.012263423919677734,
270
+ 0.012224512100219726,
271
+ 0.01226956844329834,
272
+ 0.012229632377624512,
273
+ 0.012270591735839843,
274
+ 0.012228608131408691,
275
+ 0.012273664474487305,
276
+ 0.012248064041137695,
277
+ 0.012274687767028808,
278
+ 0.012223487854003906,
279
+ 0.012272640228271485,
280
+ 0.012229632377624512,
281
+ 0.012262399673461915,
282
+ 0.01223475170135498,
283
+ 0.012259327888488769,
284
+ 0.012232704162597656,
285
+ 0.01226854419708252,
286
+ 0.012229632377624512,
287
+ 0.012274687767028808,
288
+ 0.012232704162597656,
289
+ 0.01225830364227295,
290
+ 0.012229632377624512,
291
+ 0.01226854419708252,
292
+ 0.012257280349731446,
293
+ 0.012280832290649414,
294
+ 0.012231743812561036,
295
+ 0.012269503593444824,
296
+ 0.012228608131408691,
297
+ 0.012270591735839843,
298
+ 0.012228608131408691,
299
+ 0.012266495704650878,
300
+ 0.012229632377624512,
301
+ 0.01226137638092041,
302
+ 0.012226559638977052,
303
+ 0.012255231857299804,
304
+ 0.012224512100219726,
305
+ 0.012271615982055664,
306
+ 0.012229632377624512,
307
+ 0.012255231857299804,
308
+ 0.012241920471191407,
309
+ 0.012278783798217773,
310
+ 0.012222463607788087,
311
+ 0.01226854419708252,
312
+ 0.012225536346435547,
313
+ 0.012274687767028808,
314
+ 0.012224512100219726,
315
+ 0.01226854419708252,
316
+ 0.012223487854003906,
317
+ 0.01226854419708252,
318
+ 0.012218367576599122,
319
+ 0.012283904075622559,
320
+ 0.012211199760437011,
321
+ 0.012267519950866699,
322
+ 0.012218367576599122,
323
+ 0.012263423919677734,
324
+ 0.012247039794921874,
325
+ 0.012292096138000488,
326
+ 0.01222758388519287,
327
+ 0.012275712013244629,
328
+ 0.012225536346435547,
329
+ 0.012262399673461915
330
+ ]
331
+ },
332
+ "throughput": {
333
+ "unit": "tokens/s",
334
+ "value": 80.44255961802848
335
+ },
336
+ "energy": null,
337
+ "efficiency": null
338
+ }
339
+ }
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/backend.cache_implementation=static,backend.model=google/gemma-2b,backend.torch_compile=True/summary.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model": "google/gemma-2b",
3
+ "commit": "a51a3fbc4827db26e896292b6409246fe8152cd4",
4
+ "config": "backend.cache_implementation=static,backend.torch_compile=True",
5
+ "metrics": {
6
+ "prefill.latency.mean": 0.01414243173599243,
7
+ "prefill.throughput.value": 494.96438311842996,
8
+ "decode.latency.mean": 1.5720382690429688,
9
+ "decode.throughput.value": 80.78683738234663,
10
+ "per_token.latency.mean": 0.012431230492271454,
11
+ "per_token.throughput.value": 80.44255961802848
12
+ }
13
+ }
2024-05-31/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/multirun.yaml ADDED
@@ -0,0 +1,232 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ hydra:
2
+ run:
3
+ dir: _benchmark/tmpyph4ye2g/commit=a51a3fbc4827db26e896292b6409246fe8152cd4/${hydra.job.override_dirname}
4
+ sweep:
5
+ dir: _benchmark/tmpyph4ye2g/commit=a51a3fbc4827db26e896292b6409246fe8152cd4
6
+ subdir: ${hydra.job.override_dirname}
7
+ launcher:
8
+ _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
9
+ sweeper:
10
+ _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
11
+ max_batch_size: null
12
+ params: null
13
+ help:
14
+ app_name: ${hydra.job.name}
15
+ header: '${hydra.help.app_name} is powered by Hydra.
16
+
17
+ '
18
+ footer: 'Powered by Hydra (https://hydra.cc)
19
+
20
+ Use --hydra-help to view Hydra specific help
21
+
22
+ '
23
+ template: '${hydra.help.header}
24
+
25
+ == Configuration groups ==
26
+
27
+ Compose your configuration from those groups (group=option)
28
+
29
+
30
+ $APP_CONFIG_GROUPS
31
+
32
+
33
+ == Config ==
34
+
35
+ Override anything in the config (foo.bar=value)
36
+
37
+
38
+ $CONFIG
39
+
40
+
41
+ ${hydra.help.footer}
42
+
43
+ '
44
+ hydra_help:
45
+ template: 'Hydra (${hydra.runtime.version})
46
+
47
+ See https://hydra.cc for more info.
48
+
49
+
50
+ == Flags ==
51
+
52
+ $FLAGS_HELP
53
+
54
+
55
+ == Configuration groups ==
56
+
57
+ Compose your configuration from those groups (For example, append hydra/job_logging=disabled
58
+ to command line)
59
+
60
+
61
+ $HYDRA_CONFIG_GROUPS
62
+
63
+
64
+ Use ''--cfg hydra'' to Show the Hydra config.
65
+
66
+ '
67
+ hydra_help: ???
68
+ hydra_logging:
69
+ version: 1
70
+ root:
71
+ level: ERROR
72
+ disable_existing_loggers: true
73
+ job_logging:
74
+ version: 1
75
+ root:
76
+ level: ERROR
77
+ disable_existing_loggers: true
78
+ env: {}
79
+ mode: MULTIRUN
80
+ searchpath: []
81
+ callbacks: {}
82
+ output_subdir: .hydra
83
+ overrides:
84
+ hydra:
85
+ - hydra/job_logging=disabled
86
+ - hydra/hydra_logging=disabled
87
+ - hydra.sweep.dir=_benchmark/tmpyph4ye2g/commit\=a51a3fbc4827db26e896292b6409246fe8152cd4
88
+ - hydra.run.dir=_benchmark/tmpyph4ye2g/commit\=a51a3fbc4827db26e896292b6409246fe8152cd4/${hydra.job.override_dirname}
89
+ - hydra.mode=MULTIRUN
90
+ task:
91
+ - backend.model=google/gemma-2b
92
+ - backend.cache_implementation=null,static
93
+ - backend.torch_compile=false,true
94
+ job:
95
+ name: cli
96
+ chdir: true
97
+ override_dirname: backend.cache_implementation=null,static,backend.model=google/gemma-2b,backend.torch_compile=false,true
98
+ id: ???
99
+ num: ???
100
+ config_name: generation
101
+ env_set:
102
+ OVERRIDE_BENCHMARKS: '1'
103
+ LOG_LEVEL: WARN
104
+ env_copy: []
105
+ config:
106
+ override_dirname:
107
+ kv_sep: '='
108
+ item_sep: ','
109
+ exclude_keys: []
110
+ runtime:
111
+ version: 1.3.2
112
+ version_base: '1.3'
113
+ cwd: /transformers
114
+ config_sources:
115
+ - path: hydra.conf
116
+ schema: pkg
117
+ provider: hydra
118
+ - path: optimum_benchmark
119
+ schema: pkg
120
+ provider: main
121
+ - path: /transformers/benchmark/config
122
+ schema: file
123
+ provider: command-line
124
+ - path: ''
125
+ schema: structured
126
+ provider: schema
127
+ output_dir: ???
128
+ choices:
129
+ backend: pytorch
130
+ launcher: process
131
+ scenario: inference
132
+ hydra/env: default
133
+ hydra/callbacks: null
134
+ hydra/job_logging: disabled
135
+ hydra/hydra_logging: disabled
136
+ hydra/hydra_help: default
137
+ hydra/help: default
138
+ hydra/sweeper: basic
139
+ hydra/launcher: basic
140
+ hydra/output: default
141
+ verbose: false
142
+ name: pytorch_generate
143
+ backend:
144
+ name: pytorch
145
+ version: 2.3.0+cu121
146
+ _target_: optimum_benchmark.backends.pytorch.backend.PyTorchBackend
147
+ task: null
148
+ library: null
149
+ model: google/gemma-2b
150
+ processor: null
151
+ device: cuda
152
+ device_ids: '0'
153
+ seed: 42
154
+ inter_op_num_threads: null
155
+ intra_op_num_threads: null
156
+ model_kwargs: {}
157
+ processor_kwargs: {}
158
+ hub_kwargs: {}
159
+ no_weights: true
160
+ device_map: null
161
+ torch_dtype: float16
162
+ eval_mode: true
163
+ to_bettertransformer: false
164
+ low_cpu_mem_usage: null
165
+ attn_implementation: null
166
+ cache_implementation: static
167
+ autocast_enabled: false
168
+ autocast_dtype: null
169
+ torch_compile: true
170
+ torch_compile_target: forward
171
+ torch_compile_config:
172
+ backend: inductor
173
+ mode: reduce-overhead
174
+ fullgraph: true
175
+ quantization_scheme: null
176
+ quantization_config: {}
177
+ deepspeed_inference: false
178
+ deepspeed_inference_config: {}
179
+ peft_type: null
180
+ peft_config: {}
181
+ scenario:
182
+ name: inference
183
+ _target_: optimum_benchmark.scenarios.inference.scenario.InferenceScenario
184
+ iterations: 2
185
+ duration: 0
186
+ warmup_runs: 10
187
+ input_shapes:
188
+ batch_size: 1
189
+ sequence_length: 7
190
+ new_tokens: null
191
+ latency: true
192
+ memory: true
193
+ energy: false
194
+ forward_kwargs: {}
195
+ generate_kwargs:
196
+ max_new_tokens: 128
197
+ min_new_tokens: 128
198
+ do_sample: false
199
+ call_kwargs: {}
200
+ launcher:
201
+ name: process
202
+ _target_: optimum_benchmark.launchers.process.launcher.ProcessLauncher
203
+ device_isolation: true
204
+ device_isolation_action: warn
205
+ start_method: spawn
206
+ environment:
207
+ cpu: ' AMD EPYC 7R32'
208
+ cpu_count: 16
209
+ cpu_ram_mb: 66697.29792
210
+ system: Linux
211
+ machine: x86_64
212
+ platform: Linux-5.10.216-204.855.amzn2.x86_64-x86_64-with-glibc2.29
213
+ processor: x86_64
214
+ python_version: 3.8.10
215
+ gpu:
216
+ - NVIDIA A10G
217
+ gpu_count: 1
218
+ gpu_vram_mb: 24146608128
219
+ optimum_benchmark_version: 0.2.1
220
+ optimum_benchmark_commit: null
221
+ transformers_version: 4.42.0.dev0
222
+ transformers_commit: a51a3fbc4827db26e896292b6409246fe8152cd4
223
+ accelerate_version: 0.31.0.dev0
224
+ accelerate_commit: null
225
+ diffusers_version: null
226
+ diffusers_commit: null
227
+ optimum_version: 1.21.0.dev0
228
+ optimum_commit: null
229
+ timm_version: 0.9.16
230
+ timm_commit: null
231
+ peft_version: 0.11.2.dev0
232
+ peft_commit: null
2024-05-31/summaries.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "model": "google/gemma-2b",
4
+ "commit": "a51a3fbc4827db26e896292b6409246fe8152cd4",
5
+ "config": "backend.cache_implementation=null,backend.torch_compile=False",
6
+ "metrics": {
7
+ "prefill.latency.mean": 0.01953075122833252,
8
+ "prefill.throughput.value": 358.4091527337344,
9
+ "decode.latency.mean": 2.3896519775390628,
10
+ "decode.throughput.value": 53.14581419960094,
11
+ "per_token.latency.mean": 0.01889121342836161,
12
+ "per_token.throughput.value": 52.9346621270335
13
+ }
14
+ },
15
+ {
16
+ "model": "google/gemma-2b",
17
+ "commit": "a51a3fbc4827db26e896292b6409246fe8152cd4",
18
+ "config": "backend.cache_implementation=static,backend.torch_compile=False",
19
+ "metrics": {
20
+ "prefill.latency.mean": 0.02243016052246094,
21
+ "prefill.throughput.value": 312.0797995623078,
22
+ "decode.latency.mean": 2.6233077392578124,
23
+ "decode.throughput.value": 48.4121622863549,
24
+ "per_token.latency.mean": 0.020742164204714326,
25
+ "per_token.throughput.value": 48.21097693232598
26
+ }
27
+ },
28
+ {
29
+ "model": "google/gemma-2b",
30
+ "commit": "a51a3fbc4827db26e896292b6409246fe8152cd4",
31
+ "config": "backend.cache_implementation=static,backend.torch_compile=True",
32
+ "metrics": {
33
+ "prefill.latency.mean": 0.01414243173599243,
34
+ "prefill.throughput.value": 494.96438311842996,
35
+ "decode.latency.mean": 1.5720382690429688,
36
+ "decode.throughput.value": 80.78683738234663,
37
+ "per_token.latency.mean": 0.012431230492271454,
38
+ "per_token.throughput.value": 80.44255961802848
39
+ }
40
+ }
41
+ ]
2024-05-31/summary.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "google/gemma-2b": {
3
+ "backend.cache_implementation=null,backend.torch_compile=False": {
4
+ "a51a3fbc4827db26e896292b6409246fe8152cd4": {
5
+ "metrics": {
6
+ "prefill.latency.mean": 0.01953075122833252,
7
+ "prefill.throughput.value": 358.4091527337344,
8
+ "decode.latency.mean": 2.3896519775390628,
9
+ "decode.throughput.value": 53.14581419960094,
10
+ "per_token.latency.mean": 0.01889121342836161,
11
+ "per_token.throughput.value": 52.9346621270335
12
+ }
13
+ }
14
+ },
15
+ "backend.cache_implementation=static,backend.torch_compile=False": {
16
+ "a51a3fbc4827db26e896292b6409246fe8152cd4": {
17
+ "metrics": {
18
+ "prefill.latency.mean": 0.02243016052246094,
19
+ "prefill.throughput.value": 312.0797995623078,
20
+ "decode.latency.mean": 2.6233077392578124,
21
+ "decode.throughput.value": 48.4121622863549,
22
+ "per_token.latency.mean": 0.020742164204714326,
23
+ "per_token.throughput.value": 48.21097693232598
24
+ }
25
+ }
26
+ },
27
+ "backend.cache_implementation=static,backend.torch_compile=True": {
28
+ "a51a3fbc4827db26e896292b6409246fe8152cd4": {
29
+ "metrics": {
30
+ "prefill.latency.mean": 0.01414243173599243,
31
+ "prefill.throughput.value": 494.96438311842996,
32
+ "decode.latency.mean": 1.5720382690429688,
33
+ "decode.throughput.value": 80.78683738234663,
34
+ "per_token.latency.mean": 0.012431230492271454,
35
+ "per_token.throughput.value": 80.44255961802848
36
+ }
37
+ }
38
+ }
39
+ }
40
+ }