tarekziade HF Staff commited on
Commit
d68fa05
·
1 Parent(s): b17382a

Review cloud MIC baselines arm64-Florence-2 (#2)

Browse files

- Review cloud MIC baselines arm64-Florence-2 (e73598ba5c9d6ef7e1609e44c59923240a88a3c2)

integrations/transformers/expectations/cpu-arm64/Florence-2-base-mic.bfloat16.json ADDED
@@ -0,0 +1,82 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "recorded_with": {
3
+ "contract": "huggingface-mic/Florence-2-base-mic",
4
+ "contract_name": "Florence-2",
5
+ "contract_revision": "b17382af64c8a6577317297944f7238905c5e4ba",
6
+ "framework": "transformers",
7
+ "variant": "bfloat16",
8
+ "target": "Florence-2-base-mic",
9
+ "kind": "fixture",
10
+ "revision": "b17382af64c8a6577317297944f7238905c5e4ba",
11
+ "source_model": "florence-community/Florence-2-base",
12
+ "weights": "overfit",
13
+ "fixture_repo": "huggingface-mic/Florence-2-base-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": 10.04,
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
+ "attn_implementations": {
50
+ "text_config": "sdpa",
51
+ "vision_config": "sdpa"
52
+ }
53
+ },
54
+ "inference_logits": [
55
+ [
56
+ [
57
+ -3.9375,
58
+ -6.84375,
59
+ 2.21875,
60
+ -6.78125
61
+ ]
62
+ ],
63
+ [
64
+ [
65
+ -3.640625,
66
+ -5.5625,
67
+ 2.140625,
68
+ -5.4375
69
+ ]
70
+ ]
71
+ ],
72
+ "training": {
73
+ "loss": 0.013443506322801113,
74
+ "gradients": {
75
+ "model.language_model.encoder.layers.0.self_attn.k_proj.weight": true,
76
+ "model.language_model.decoder.layers.0.self_attn.k_proj.weight": true,
77
+ "model.multi_modal_projector.image_projection.weight": true
78
+ },
79
+ "updated": true
80
+ }
81
+ }
82
+ }
integrations/transformers/expectations/cpu-arm64/Florence-2-base-mic.eager.json ADDED
@@ -0,0 +1,82 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "recorded_with": {
3
+ "contract": "huggingface-mic/Florence-2-base-mic",
4
+ "contract_name": "Florence-2",
5
+ "contract_revision": "b17382af64c8a6577317297944f7238905c5e4ba",
6
+ "framework": "transformers",
7
+ "variant": "eager",
8
+ "target": "Florence-2-base-mic",
9
+ "kind": "fixture",
10
+ "revision": "b17382af64c8a6577317297944f7238905c5e4ba",
11
+ "source_model": "florence-community/Florence-2-base",
12
+ "weights": "overfit",
13
+ "fixture_repo": "huggingface-mic/Florence-2-base-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": 5.31,
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
+ "attn_implementations": {
50
+ "text_config": "eager",
51
+ "vision_config": "eager"
52
+ }
53
+ },
54
+ "inference_logits": [
55
+ [
56
+ [
57
+ -3.933049440383911,
58
+ -6.834774494171143,
59
+ 2.224327802658081,
60
+ -6.791712760925293
61
+ ]
62
+ ],
63
+ [
64
+ [
65
+ -3.6436893939971924,
66
+ -5.550571441650391,
67
+ 2.1495935916900635,
68
+ -5.428072452545166
69
+ ]
70
+ ]
71
+ ],
72
+ "training": {
73
+ "loss": 0.01346452534198761,
74
+ "gradients": {
75
+ "model.language_model.encoder.layers.0.self_attn.k_proj.weight": true,
76
+ "model.language_model.decoder.layers.0.self_attn.k_proj.weight": true,
77
+ "model.multi_modal_projector.image_projection.weight": true
78
+ },
79
+ "updated": true
80
+ }
81
+ }
82
+ }
integrations/transformers/expectations/cpu-arm64/Florence-2-base-mic.json ADDED
@@ -0,0 +1,82 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "recorded_with": {
3
+ "contract": "huggingface-mic/Florence-2-base-mic",
4
+ "contract_name": "Florence-2",
5
+ "contract_revision": "b17382af64c8a6577317297944f7238905c5e4ba",
6
+ "framework": "transformers",
7
+ "variant": "default",
8
+ "target": "Florence-2-base-mic",
9
+ "kind": "fixture",
10
+ "revision": "b17382af64c8a6577317297944f7238905c5e4ba",
11
+ "source_model": "florence-community/Florence-2-base",
12
+ "weights": "overfit",
13
+ "fixture_repo": "huggingface-mic/Florence-2-base-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": 5.28,
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
+ "attn_implementations": {
50
+ "text_config": "sdpa",
51
+ "vision_config": "sdpa"
52
+ }
53
+ },
54
+ "inference_logits": [
55
+ [
56
+ [
57
+ -3.933049440383911,
58
+ -6.834774494171143,
59
+ 2.224327802658081,
60
+ -6.791713237762451
61
+ ]
62
+ ],
63
+ [
64
+ [
65
+ -3.643688917160034,
66
+ -5.550570964813232,
67
+ 2.1495935916900635,
68
+ -5.428072452545166
69
+ ]
70
+ ]
71
+ ],
72
+ "training": {
73
+ "loss": 0.013464555144309998,
74
+ "gradients": {
75
+ "model.language_model.encoder.layers.0.self_attn.k_proj.weight": true,
76
+ "model.language_model.decoder.layers.0.self_attn.k_proj.weight": true,
77
+ "model.multi_modal_projector.image_projection.weight": true
78
+ },
79
+ "updated": true
80
+ }
81
+ }
82
+ }