Review cloud MIC baselines arm64-OWL-ViT

#1
by tarekziade HF Staff - opened
integrations/transformers/expectations/cpu-arm64/owlvit-base-patch32-mic.bfloat16.json ADDED
@@ -0,0 +1,127 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "recorded_with": {
3
+ "contract": "huggingface-mic/owlvit-base-patch32-mic",
4
+ "contract_name": "OWL-ViT",
5
+ "contract_revision": "09d714a0dd1f21dcf214cb144777eb72042a08d0",
6
+ "framework": "transformers",
7
+ "variant": "bfloat16",
8
+ "target": "owlvit-base-patch32-mic",
9
+ "kind": "fixture",
10
+ "revision": "09d714a0dd1f21dcf214cb144777eb72042a08d0",
11
+ "source_model": "google/owlvit-base-patch32",
12
+ "weights": "random",
13
+ "fixture_repo": "huggingface-mic/owlvit-base-patch32-mic",
14
+ "environment": {
15
+ "python": "3.11.14",
16
+ "transformers": "5.23.0.dev0",
17
+ "sentence-transformers": null,
18
+ "torch": "2.15.0",
19
+ "torchvision": "0.29.2",
20
+ "torchaudio": "2.11.1",
21
+ "tokenizers": "0.23.3",
22
+ "numpy": "2.3.5",
23
+ "pillow": "13.0.0",
24
+ "accelerate": "1.15.0",
25
+ "safetensors": "0.8.0",
26
+ "huggingface-hub": "2.4.0",
27
+ "sentencepiece": "0.2.2",
28
+ "protobuf": "7.36.2",
29
+ "librosa": "0.11.0",
30
+ "soundfile": "0.14.0",
31
+ "scipy": "1.17.1"
32
+ },
33
+ "seconds": 2.57,
34
+ "environment_differs": {
35
+ "torch": "2.15.0 (pinned 2.14.0)",
36
+ "torchvision": "0.29.2 (pinned 0.29.1)",
37
+ "torchaudio": "2.11.1 (pinned 2.11.0)",
38
+ "pillow": "13.0.0 (pinned 12.3.0)",
39
+ "huggingface-hub": "2.4.0 (pinned 2.3.0)"
40
+ },
41
+ "platform": "cpu-arm64",
42
+ "template_date": "2026-01-01"
43
+ },
44
+ "outputs": {
45
+ "runtime": {
46
+ "device": "cpu",
47
+ "dtype": "bfloat16",
48
+ "attn_implementation": "sdpa"
49
+ },
50
+ "detections": {
51
+ "boxes": [
52
+ [
53
+ [
54
+ 384.0,
55
+ -384.0,
56
+ 1152.0,
57
+ 384.0
58
+ ],
59
+ [
60
+ 384.0,
61
+ -384.0,
62
+ 1152.0,
63
+ 384.0
64
+ ],
65
+ [
66
+ 384.0,
67
+ -384.0,
68
+ 1152.0,
69
+ 384.0
70
+ ]
71
+ ],
72
+ [
73
+ [
74
+ 384.0,
75
+ -384.0,
76
+ 1152.0,
77
+ 384.0
78
+ ],
79
+ [
80
+ 384.0,
81
+ -384.0,
82
+ 1152.0,
83
+ 384.0
84
+ ],
85
+ [
86
+ 384.0,
87
+ -384.0,
88
+ 1152.0,
89
+ 384.0
90
+ ]
91
+ ]
92
+ ],
93
+ "scores": [
94
+ [
95
+ 1.0,
96
+ 1.0,
97
+ 1.0
98
+ ],
99
+ [
100
+ 1.0,
101
+ 1.0,
102
+ 1.0
103
+ ]
104
+ ],
105
+ "labels": [
106
+ [
107
+ "a photo of a dog",
108
+ "a photo of a dog",
109
+ "a photo of a dog"
110
+ ],
111
+ [
112
+ "a red car",
113
+ "a red car",
114
+ "a red car"
115
+ ]
116
+ ]
117
+ },
118
+ "training": {
119
+ "loss": 10.950000762939453,
120
+ "gradients": {
121
+ "class_head.dense0.weight": true,
122
+ "owlvit.vision_model.encoder.layers.0.self_attn.q_proj.weight": true
123
+ },
124
+ "updated": true
125
+ }
126
+ }
127
+ }
integrations/transformers/expectations/cpu-arm64/owlvit-base-patch32-mic.eager.json ADDED
@@ -0,0 +1,127 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "recorded_with": {
3
+ "contract": "huggingface-mic/owlvit-base-patch32-mic",
4
+ "contract_name": "OWL-ViT",
5
+ "contract_revision": "09d714a0dd1f21dcf214cb144777eb72042a08d0",
6
+ "framework": "transformers",
7
+ "variant": "eager",
8
+ "target": "owlvit-base-patch32-mic",
9
+ "kind": "fixture",
10
+ "revision": "09d714a0dd1f21dcf214cb144777eb72042a08d0",
11
+ "source_model": "google/owlvit-base-patch32",
12
+ "weights": "random",
13
+ "fixture_repo": "huggingface-mic/owlvit-base-patch32-mic",
14
+ "environment": {
15
+ "python": "3.11.14",
16
+ "transformers": "5.23.0.dev0",
17
+ "sentence-transformers": null,
18
+ "torch": "2.15.0",
19
+ "torchvision": "0.29.2",
20
+ "torchaudio": "2.11.1",
21
+ "tokenizers": "0.23.3",
22
+ "numpy": "2.3.5",
23
+ "pillow": "13.0.0",
24
+ "accelerate": "1.15.0",
25
+ "safetensors": "0.8.0",
26
+ "huggingface-hub": "2.4.0",
27
+ "sentencepiece": "0.2.2",
28
+ "protobuf": "7.36.2",
29
+ "librosa": "0.11.0",
30
+ "soundfile": "0.14.0",
31
+ "scipy": "1.17.1"
32
+ },
33
+ "seconds": 2.38,
34
+ "environment_differs": {
35
+ "torch": "2.15.0 (pinned 2.14.0)",
36
+ "torchvision": "0.29.2 (pinned 0.29.1)",
37
+ "torchaudio": "2.11.1 (pinned 2.11.0)",
38
+ "pillow": "13.0.0 (pinned 12.3.0)",
39
+ "huggingface-hub": "2.4.0 (pinned 2.3.0)"
40
+ },
41
+ "platform": "cpu-arm64",
42
+ "template_date": "2026-01-01"
43
+ },
44
+ "outputs": {
45
+ "runtime": {
46
+ "device": "cpu",
47
+ "dtype": "float32",
48
+ "attn_implementation": "eager"
49
+ },
50
+ "detections": {
51
+ "boxes": [
52
+ [
53
+ [
54
+ 384.0,
55
+ -384.0,
56
+ 1152.0,
57
+ 384.0
58
+ ],
59
+ [
60
+ 384.0,
61
+ -384.0,
62
+ 1152.0,
63
+ 384.0
64
+ ],
65
+ [
66
+ 384.0,
67
+ -384.0,
68
+ 1152.0,
69
+ 384.0
70
+ ]
71
+ ],
72
+ [
73
+ [
74
+ 384.0,
75
+ -384.0,
76
+ 1152.0,
77
+ 384.0
78
+ ],
79
+ [
80
+ 384.0,
81
+ -384.0,
82
+ 1152.0,
83
+ 384.0
84
+ ],
85
+ [
86
+ 384.0,
87
+ -384.0,
88
+ 1152.0,
89
+ 384.0
90
+ ]
91
+ ]
92
+ ],
93
+ "scores": [
94
+ [
95
+ 1.0,
96
+ 1.0,
97
+ 1.0
98
+ ],
99
+ [
100
+ 1.0,
101
+ 1.0,
102
+ 1.0
103
+ ]
104
+ ],
105
+ "labels": [
106
+ [
107
+ "a photo of a dog",
108
+ "a photo of a dog",
109
+ "a photo of a dog"
110
+ ],
111
+ [
112
+ "a red car",
113
+ "a red car",
114
+ "a red car"
115
+ ]
116
+ ]
117
+ },
118
+ "training": {
119
+ "loss": 10.527908325195312,
120
+ "gradients": {
121
+ "class_head.dense0.weight": true,
122
+ "owlvit.vision_model.encoder.layers.0.self_attn.q_proj.weight": true
123
+ },
124
+ "updated": true
125
+ }
126
+ }
127
+ }
integrations/transformers/expectations/cpu-arm64/owlvit-base-patch32-mic.json ADDED
@@ -0,0 +1,127 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "recorded_with": {
3
+ "contract": "huggingface-mic/owlvit-base-patch32-mic",
4
+ "contract_name": "OWL-ViT",
5
+ "contract_revision": "09d714a0dd1f21dcf214cb144777eb72042a08d0",
6
+ "framework": "transformers",
7
+ "variant": "default",
8
+ "target": "owlvit-base-patch32-mic",
9
+ "kind": "fixture",
10
+ "revision": "09d714a0dd1f21dcf214cb144777eb72042a08d0",
11
+ "source_model": "google/owlvit-base-patch32",
12
+ "weights": "random",
13
+ "fixture_repo": "huggingface-mic/owlvit-base-patch32-mic",
14
+ "environment": {
15
+ "python": "3.11.14",
16
+ "transformers": "5.23.0.dev0",
17
+ "sentence-transformers": null,
18
+ "torch": "2.15.0",
19
+ "torchvision": "0.29.2",
20
+ "torchaudio": "2.11.1",
21
+ "tokenizers": "0.23.3",
22
+ "numpy": "2.3.5",
23
+ "pillow": "13.0.0",
24
+ "accelerate": "1.15.0",
25
+ "safetensors": "0.8.0",
26
+ "huggingface-hub": "2.4.0",
27
+ "sentencepiece": "0.2.2",
28
+ "protobuf": "7.36.2",
29
+ "librosa": "0.11.0",
30
+ "soundfile": "0.14.0",
31
+ "scipy": "1.17.1"
32
+ },
33
+ "seconds": 2.46,
34
+ "environment_differs": {
35
+ "torch": "2.15.0 (pinned 2.14.0)",
36
+ "torchvision": "0.29.2 (pinned 0.29.1)",
37
+ "torchaudio": "2.11.1 (pinned 2.11.0)",
38
+ "pillow": "13.0.0 (pinned 12.3.0)",
39
+ "huggingface-hub": "2.4.0 (pinned 2.3.0)"
40
+ },
41
+ "platform": "cpu-arm64",
42
+ "template_date": "2026-01-01"
43
+ },
44
+ "outputs": {
45
+ "runtime": {
46
+ "device": "cpu",
47
+ "dtype": "float32",
48
+ "attn_implementation": "sdpa"
49
+ },
50
+ "detections": {
51
+ "boxes": [
52
+ [
53
+ [
54
+ 384.0,
55
+ -384.0,
56
+ 1152.0,
57
+ 384.0
58
+ ],
59
+ [
60
+ 384.0,
61
+ -384.0,
62
+ 1152.0,
63
+ 384.0
64
+ ],
65
+ [
66
+ 384.0,
67
+ -384.0,
68
+ 1152.0,
69
+ 384.0
70
+ ]
71
+ ],
72
+ [
73
+ [
74
+ 384.0,
75
+ -384.0,
76
+ 1152.0,
77
+ 384.0
78
+ ],
79
+ [
80
+ 384.0,
81
+ -384.0,
82
+ 1152.0,
83
+ 384.0
84
+ ],
85
+ [
86
+ 384.0,
87
+ -384.0,
88
+ 1152.0,
89
+ 384.0
90
+ ]
91
+ ]
92
+ ],
93
+ "scores": [
94
+ [
95
+ 1.0,
96
+ 1.0,
97
+ 1.0
98
+ ],
99
+ [
100
+ 1.0,
101
+ 1.0,
102
+ 1.0
103
+ ]
104
+ ],
105
+ "labels": [
106
+ [
107
+ "a photo of a dog",
108
+ "a photo of a dog",
109
+ "a photo of a dog"
110
+ ],
111
+ [
112
+ "a red car",
113
+ "a red car",
114
+ "a red car"
115
+ ]
116
+ ]
117
+ },
118
+ "training": {
119
+ "loss": 10.527918815612793,
120
+ "gradients": {
121
+ "class_head.dense0.weight": true,
122
+ "owlvit.vision_model.encoder.layers.0.self_attn.q_proj.weight": true
123
+ },
124
+ "updated": true
125
+ }
126
+ }
127
+ }