Hugging Face's logo

ShesterG
/
SHuBERT

Model card Files
xet
 Community
ShesterG commited on
Commit
04f74e6
·
verified ·
1 Parent(s): 3166903

initial commit

Browse files
Files changed (27) hide show
  1. .gitattributes +2 -0
  2. models/byt5_base/config.json +33 -0
  3. models/byt5_base/generation_config.json +7 -0
  4. models/byt5_base/pytorch_model.bin +3 -0
  5. models/byt5_base/special_tokens_map.json +150 -0
  6. models/byt5_base/tokenizer_config.json +157 -0
  7. models/checkpoint-11625/config.json +41 -0
  8. models/checkpoint-11625/generation_config.json +6 -0
  9. models/checkpoint-11625/optimizer.pt +3 -0
  10. models/checkpoint-11625/pytorch_model.bin +3 -0
  11. models/checkpoint-11625/rng_state_0.pth +3 -0
  12. models/checkpoint-11625/rng_state_1.pth +3 -0
  13. models/checkpoint-11625/rng_state_2.pth +3 -0
  14. models/checkpoint-11625/rng_state_3.pth +3 -0
  15. models/checkpoint-11625/rng_state_4.pth +3 -0
  16. models/checkpoint-11625/rng_state_5.pth +3 -0
  17. models/checkpoint-11625/rng_state_6.pth +3 -0
  18. models/checkpoint-11625/rng_state_7.pth +3 -0
  19. models/checkpoint-11625/scheduler.pt +3 -0
  20. models/checkpoint-11625/trainer_state.json +316 -0
  21. models/checkpoint-11625/training_args.bin +3 -0
  22. models/checkpoint_836_400000.pt +3 -0
  23. models/dinov2face.pth +3 -0
  24. models/dinov2hand.pth +3 -0
  25. models/face_landmarker_v2_with_blendshapes.task +3 -0
  26. models/hand_landmarker.task +3 -0
  27. models/yolov8n.pt +3 -0
.gitattributes CHANGED
@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ models/face_landmarker_v2_with_blendshapes.task filter=lfs diff=lfs merge=lfs -text
37
+ models/hand_landmarker.task filter=lfs diff=lfs merge=lfs -text
models/byt5_base/config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google/byt5-base",
3
+ "architectures": [
4
+ "T5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 3968,
7
+ "d_kv": 64,
8
+ "d_model": 1536,
9
+ "decoder_start_token_id": 0,
10
+ "dense_act_fn": "gelu_new",
11
+ "dropout_rate": 0.3,
12
+ "eos_token_id": 1,
13
+ "feed_forward_proj": "gated-gelu",
14
+ "gradient_checkpointing": false,
15
+ "initializer_factor": 1.0,
16
+ "is_encoder_decoder": true,
17
+ "is_gated_act": true,
18
+ "layer_norm_epsilon": 1e-06,
19
+ "model_type": "t5",
20
+ "num_decoder_layers": 6,
21
+ "num_heads": 12,
22
+ "num_layers": 18,
23
+ "output_past": true,
24
+ "pad_token_id": 0,
25
+ "relative_attention_max_distance": 128,
26
+ "relative_attention_num_buckets": 32,
27
+ "tie_word_embeddings": false,
28
+ "tokenizer_class": "ByT5Tokenizer",
29
+ "torch_dtype": "float32",
30
+ "transformers_version": "4.30.2",
31
+ "use_cache": true,
32
+ "vocab_size": 384
33
+ }
models/byt5_base/generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "decoder_start_token_id": 0,
4
+ "eos_token_id": 1,
5
+ "pad_token_id": 0,
6
+ "transformers_version": "4.30.2"
7
+ }
models/byt5_base/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4651bb6e44016ece08e5c614478da3c6b919191019764d7a43ff78d586e254a6
3
+ size 2326696954
models/byt5_base/special_tokens_map.json ADDED
@@ -0,0 +1,150 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<extra_id_0>",
4
+ "<extra_id_1>",
5
+ "<extra_id_2>",
6
+ "<extra_id_3>",
7
+ "<extra_id_4>",
8
+ "<extra_id_5>",
9
+ "<extra_id_6>",
10
+ "<extra_id_7>",
11
+ "<extra_id_8>",
12
+ "<extra_id_9>",
13
+ "<extra_id_10>",
14
+ "<extra_id_11>",
15
+ "<extra_id_12>",
16
+ "<extra_id_13>",
17
+ "<extra_id_14>",
18
+ "<extra_id_15>",
19
+ "<extra_id_16>",
20
+ "<extra_id_17>",
21
+ "<extra_id_18>",
22
+ "<extra_id_19>",
23
+ "<extra_id_20>",
24
+ "<extra_id_21>",
25
+ "<extra_id_22>",
26
+ "<extra_id_23>",
27
+ "<extra_id_24>",
28
+ "<extra_id_25>",
29
+ "<extra_id_26>",
30
+ "<extra_id_27>",
31
+ "<extra_id_28>",
32
+ "<extra_id_29>",
33
+ "<extra_id_30>",
34
+ "<extra_id_31>",
35
+ "<extra_id_32>",
36
+ "<extra_id_33>",
37
+ "<extra_id_34>",
38
+ "<extra_id_35>",
39
+ "<extra_id_36>",
40
+ "<extra_id_37>",
41
+ "<extra_id_38>",
42
+ "<extra_id_39>",
43
+ "<extra_id_40>",
44
+ "<extra_id_41>",
45
+ "<extra_id_42>",
46
+ "<extra_id_43>",
47
+ "<extra_id_44>",
48
+ "<extra_id_45>",
49
+ "<extra_id_46>",
50
+ "<extra_id_47>",
51
+ "<extra_id_48>",
52
+ "<extra_id_49>",
53
+ "<extra_id_50>",
54
+ "<extra_id_51>",
55
+ "<extra_id_52>",
56
+ "<extra_id_53>",
57
+ "<extra_id_54>",
58
+ "<extra_id_55>",
59
+ "<extra_id_56>",
60
+ "<extra_id_57>",
61
+ "<extra_id_58>",
62
+ "<extra_id_59>",
63
+ "<extra_id_60>",
64
+ "<extra_id_61>",
65
+ "<extra_id_62>",
66
+ "<extra_id_63>",
67
+ "<extra_id_64>",
68
+ "<extra_id_65>",
69
+ "<extra_id_66>",
70
+ "<extra_id_67>",
71
+ "<extra_id_68>",
72
+ "<extra_id_69>",
73
+ "<extra_id_70>",
74
+ "<extra_id_71>",
75
+ "<extra_id_72>",
76
+ "<extra_id_73>",
77
+ "<extra_id_74>",
78
+ "<extra_id_75>",
79
+ "<extra_id_76>",
80
+ "<extra_id_77>",
81
+ "<extra_id_78>",
82
+ "<extra_id_79>",
83
+ "<extra_id_80>",
84
+ "<extra_id_81>",
85
+ "<extra_id_82>",
86
+ "<extra_id_83>",
87
+ "<extra_id_84>",
88
+ "<extra_id_85>",
89
+ "<extra_id_86>",
90
+ "<extra_id_87>",
91
+ "<extra_id_88>",
92
+ "<extra_id_89>",
93
+ "<extra_id_90>",
94
+ "<extra_id_91>",
95
+ "<extra_id_92>",
96
+ "<extra_id_93>",
97
+ "<extra_id_94>",
98
+ "<extra_id_95>",
99
+ "<extra_id_96>",
100
+ "<extra_id_97>",
101
+ "<extra_id_98>",
102
+ "<extra_id_99>",
103
+ "<extra_id_100>",
104
+ "<extra_id_101>",
105
+ "<extra_id_102>",
106
+ "<extra_id_103>",
107
+ "<extra_id_104>",
108
+ "<extra_id_105>",
109
+ "<extra_id_106>",
110
+ "<extra_id_107>",
111
+ "<extra_id_108>",
112
+ "<extra_id_109>",
113
+ "<extra_id_110>",
114
+ "<extra_id_111>",
115
+ "<extra_id_112>",
116
+ "<extra_id_113>",
117
+ "<extra_id_114>",
118
+ "<extra_id_115>",
119
+ "<extra_id_116>",
120
+ "<extra_id_117>",
121
+ "<extra_id_118>",
122
+ "<extra_id_119>",
123
+ "<extra_id_120>",
124
+ "<extra_id_121>",
125
+ "<extra_id_122>",
126
+ "<extra_id_123>",
127
+ "<extra_id_124>"
128
+ ],
129
+ "eos_token": {
130
+ "content": "</s>",
131
+ "lstrip": false,
132
+ "normalized": true,
133
+ "rstrip": false,
134
+ "single_word": false
135
+ },
136
+ "pad_token": {
137
+ "content": "<pad>",
138
+ "lstrip": false,
139
+ "normalized": true,
140
+ "rstrip": false,
141
+ "single_word": false
142
+ },
143
+ "unk_token": {
144
+ "content": "<unk>",
145
+ "lstrip": false,
146
+ "normalized": true,
147
+ "rstrip": false,
148
+ "single_word": false
149
+ }
150
+ }
models/byt5_base/tokenizer_config.json ADDED
@@ -0,0 +1,157 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<extra_id_0>",
4
+ "<extra_id_1>",
5
+ "<extra_id_2>",
6
+ "<extra_id_3>",
7
+ "<extra_id_4>",
8
+ "<extra_id_5>",
9
+ "<extra_id_6>",
10
+ "<extra_id_7>",
11
+ "<extra_id_8>",
12
+ "<extra_id_9>",
13
+ "<extra_id_10>",
14
+ "<extra_id_11>",
15
+ "<extra_id_12>",
16
+ "<extra_id_13>",
17
+ "<extra_id_14>",
18
+ "<extra_id_15>",
19
+ "<extra_id_16>",
20
+ "<extra_id_17>",
21
+ "<extra_id_18>",
22
+ "<extra_id_19>",
23
+ "<extra_id_20>",
24
+ "<extra_id_21>",
25
+ "<extra_id_22>",
26
+ "<extra_id_23>",
27
+ "<extra_id_24>",
28
+ "<extra_id_25>",
29
+ "<extra_id_26>",
30
+ "<extra_id_27>",
31
+ "<extra_id_28>",
32
+ "<extra_id_29>",
33
+ "<extra_id_30>",
34
+ "<extra_id_31>",
35
+ "<extra_id_32>",
36
+ "<extra_id_33>",
37
+ "<extra_id_34>",
38
+ "<extra_id_35>",
39
+ "<extra_id_36>",
40
+ "<extra_id_37>",
41
+ "<extra_id_38>",
42
+ "<extra_id_39>",
43
+ "<extra_id_40>",
44
+ "<extra_id_41>",
45
+ "<extra_id_42>",
46
+ "<extra_id_43>",
47
+ "<extra_id_44>",
48
+ "<extra_id_45>",
49
+ "<extra_id_46>",
50
+ "<extra_id_47>",
51
+ "<extra_id_48>",
52
+ "<extra_id_49>",
53
+ "<extra_id_50>",
54
+ "<extra_id_51>",
55
+ "<extra_id_52>",
56
+ "<extra_id_53>",
57
+ "<extra_id_54>",
58
+ "<extra_id_55>",
59
+ "<extra_id_56>",
60
+ "<extra_id_57>",
61
+ "<extra_id_58>",
62
+ "<extra_id_59>",
63
+ "<extra_id_60>",
64
+ "<extra_id_61>",
65
+ "<extra_id_62>",
66
+ "<extra_id_63>",
67
+ "<extra_id_64>",
68
+ "<extra_id_65>",
69
+ "<extra_id_66>",
70
+ "<extra_id_67>",
71
+ "<extra_id_68>",
72
+ "<extra_id_69>",
73
+ "<extra_id_70>",
74
+ "<extra_id_71>",
75
+ "<extra_id_72>",
76
+ "<extra_id_73>",
77
+ "<extra_id_74>",
78
+ "<extra_id_75>",
79
+ "<extra_id_76>",
80
+ "<extra_id_77>",
81
+ "<extra_id_78>",
82
+ "<extra_id_79>",
83
+ "<extra_id_80>",
84
+ "<extra_id_81>",
85
+ "<extra_id_82>",
86
+ "<extra_id_83>",
87
+ "<extra_id_84>",
88
+ "<extra_id_85>",
89
+ "<extra_id_86>",
90
+ "<extra_id_87>",
91
+ "<extra_id_88>",
92
+ "<extra_id_89>",
93
+ "<extra_id_90>",
94
+ "<extra_id_91>",
95
+ "<extra_id_92>",
96
+ "<extra_id_93>",
97
+ "<extra_id_94>",
98
+ "<extra_id_95>",
99
+ "<extra_id_96>",
100
+ "<extra_id_97>",
101
+ "<extra_id_98>",
102
+ "<extra_id_99>",
103
+ "<extra_id_100>",
104
+ "<extra_id_101>",
105
+ "<extra_id_102>",
106
+ "<extra_id_103>",
107
+ "<extra_id_104>",
108
+ "<extra_id_105>",
109
+ "<extra_id_106>",
110
+ "<extra_id_107>",
111
+ "<extra_id_108>",
112
+ "<extra_id_109>",
113
+ "<extra_id_110>",
114
+ "<extra_id_111>",
115
+ "<extra_id_112>",
116
+ "<extra_id_113>",
117
+ "<extra_id_114>",
118
+ "<extra_id_115>",
119
+ "<extra_id_116>",
120
+ "<extra_id_117>",
121
+ "<extra_id_118>",
122
+ "<extra_id_119>",
123
+ "<extra_id_120>",
124
+ "<extra_id_121>",
125
+ "<extra_id_122>",
126
+ "<extra_id_123>",
127
+ "<extra_id_124>"
128
+ ],
129
+ "clean_up_tokenization_spaces": true,
130
+ "eos_token": {
131
+ "__type": "AddedToken",
132
+ "content": "</s>",
133
+ "lstrip": false,
134
+ "normalized": true,
135
+ "rstrip": false,
136
+ "single_word": false
137
+ },
138
+ "extra_ids": 125,
139
+ "model_max_length": 1000000000000000019884624838656,
140
+ "pad_token": {
141
+ "__type": "AddedToken",
142
+ "content": "<pad>",
143
+ "lstrip": false,
144
+ "normalized": true,
145
+ "rstrip": false,
146
+ "single_word": false
147
+ },
148
+ "tokenizer_class": "ByT5Tokenizer",
149
+ "unk_token": {
150
+ "__type": "AddedToken",
151
+ "content": "<unk>",
152
+ "lstrip": false,
153
+ "normalized": true,
154
+ "rstrip": false,
155
+ "single_word": false
156
+ }
157
+ }
models/checkpoint-11625/config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/share/data/pals/shester/translation/signlang-hf-translation-2/_outputs/yasl_shubertfinetune_layer0nosoftmax_byt5_4stream/checkpoint-184000",
3
+ "adapter": "linear",
4
+ "architectures": [
5
+ "SignLanguageByT5ForConditionalGeneration"
6
+ ],
7
+ "channels": "face,left_hand,right_hand,body_posture",
8
+ "d_ff": 3968,
9
+ "d_kv": 64,
10
+ "d_model": 1536,
11
+ "decoder_start_token_id": 0,
12
+ "dense_act_fn": "gelu_new",
13
+ "dropout_rate": 0.3,
14
+ "eos_token_id": 1,
15
+ "extraction_layer": 0,
16
+ "face_dim": 384,
17
+ "feed_forward_proj": "gated-gelu",
18
+ "finetune_signhubert": false,
19
+ "gradient_checkpointing": false,
20
+ "hand_dim": 384,
21
+ "initializer_factor": 1.0,
22
+ "is_encoder_decoder": true,
23
+ "is_gated_act": true,
24
+ "layer_norm_epsilon": 1e-06,
25
+ "model_type": "t5",
26
+ "num_decoder_layers": 6,
27
+ "num_heads": 12,
28
+ "num_layers": 18,
29
+ "output_past": true,
30
+ "pad_token_id": 0,
31
+ "pose_dim": 14,
32
+ "relative_attention_max_distance": 128,
33
+ "relative_attention_num_buckets": 32,
34
+ "representations_dim": 768,
35
+ "tie_word_embeddings": false,
36
+ "tokenizer_class": "ByT5Tokenizer",
37
+ "torch_dtype": "float32",
38
+ "transformers_version": "4.30.2",
39
+ "use_cache": true,
40
+ "vocab_size": 384
41
+ }
models/checkpoint-11625/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "decoder_start_token_id": 0,
3
+ "eos_token_id": 1,
4
+ "pad_token_id": 0,
5
+ "transformers_version": "4.30.2"
6
+ }
models/checkpoint-11625/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:31d300d0675fb03d6aa09294231d4fd263008cd4e4f508fc34a8a2dce74b9eb7
3
+ size 5361549242
models/checkpoint-11625/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cb605058abf7fd50b80f055fe8a6e271ed64f04ef7b1b450c7a364f88495526a
3
+ size 2683976334
models/checkpoint-11625/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:08ea98bfa0ed33496185ef6cc28cafc143e531cfbe7b7e074392f9a3b7a5dcc4
3
+ size 15984
models/checkpoint-11625/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2cd916d759973bf7731364295a464821092019bafc68fc09ee0e68bb53fd7b98
3
+ size 15984
models/checkpoint-11625/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a040dd4933f71369f6db78fbba791c16f3f4355f0400ccafa5c2a470c8dfa23
3
+ size 15984
models/checkpoint-11625/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a902d12473668da2330b3c3d09194cdfd3a8539b74be6787756ee5e5e6420659
3
+ size 15984
models/checkpoint-11625/rng_state_4.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3b03403fd7dd38a12788f28bdf43e7cc27b5d475e24957765dc10d9a1c99cb6e
3
+ size 15984
models/checkpoint-11625/rng_state_5.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:96a1aeaef7afbe40243305d22ffefc472fa52f62e274fd8c22b7abefcc326421
3
+ size 15984
models/checkpoint-11625/rng_state_6.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c7b185de823a33fd4167145a9c90da5ea93fa947842c8abdf0c26b35a04ebc0
3
+ size 15984
models/checkpoint-11625/rng_state_7.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87294a19873c95c6fb3ce4e1d7737c1417eda728402cb4f433c1d40207e8d360
3
+ size 15984
models/checkpoint-11625/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd20d76f422c1104d0bfa0682f9c61df13cf41b35b4a1edcf1c54d84b188d51b
3
+ size 1064
models/checkpoint-11625/trainer_state.json ADDED
@@ -0,0 +1,316 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 29.97126821077391,
3
+ "best_model_checkpoint": "/share/data/pals/shester/translation/signlang-hf-translation-2/_outputs/gsl_merged_sentencesindependent_test/checkpoint-11625",
4
+ "epoch": 30.02064849169221,
5
+ "global_step": 11625,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 2.0,
12
+ "learning_rate": 0.0005,
13
+ "loss": 2.1869,
14
+ "step": 775
15
+ },
16
+ {
17
+ "epoch": 2.0,
18
+ "eval_bleu-1": 50.309218203033836,
19
+ "eval_bleu-2": 36.14451108729435,
20
+ "eval_bleu-3": 27.208363540058304,
21
+ "eval_bleu-4": 20.749772116970437,
22
+ "eval_loss": 2.169796943664551,
23
+ "eval_runtime": 77.6393,
24
+ "eval_sacrebleu": 20.749772116970426,
25
+ "eval_samples_per_second": 20.827,
26
+ "eval_steps_per_second": 0.657,
27
+ "eval_wer": 67.85005950019834,
28
+ "step": 775
29
+ },
30
+ {
31
+ "epoch": 4.0,
32
+ "learning_rate": 0.0005,
33
+ "loss": 2.1398,
34
+ "step": 1550
35
+ },
36
+ {
37
+ "epoch": 4.0,
38
+ "eval_bleu-1": 53.2211566747073,
39
+ "eval_bleu-2": 39.042439243817604,
40
+ "eval_bleu-3": 30.004808472168353,
41
+ "eval_bleu-4": 23.412065127614813,
42
+ "eval_loss": 2.1439623832702637,
43
+ "eval_runtime": 71.956,
44
+ "eval_sacrebleu": 23.412065127614802,
45
+ "eval_samples_per_second": 22.472,
46
+ "eval_steps_per_second": 0.709,
47
+ "eval_wer": 63.97593547534047,
48
+ "step": 1550
49
+ },
50
+ {
51
+ "epoch": 6.0,
52
+ "learning_rate": 0.0005,
53
+ "loss": 2.1184,
54
+ "step": 2325
55
+ },
56
+ {
57
+ "epoch": 6.0,
58
+ "eval_bleu-1": 54.72246687202718,
59
+ "eval_bleu-2": 40.62506031297361,
60
+ "eval_bleu-3": 31.44563117560889,
61
+ "eval_bleu-4": 24.726633214061028,
62
+ "eval_loss": 2.1261134147644043,
63
+ "eval_runtime": 72.1839,
64
+ "eval_sacrebleu": 24.726633214061028,
65
+ "eval_samples_per_second": 22.401,
66
+ "eval_steps_per_second": 0.707,
67
+ "eval_wer": 61.820706069020225,
68
+ "step": 2325
69
+ },
70
+ {
71
+ "epoch": 8.0,
72
+ "learning_rate": 0.0005,
73
+ "loss": 2.1025,
74
+ "step": 3100
75
+ },
76
+ {
77
+ "epoch": 8.0,
78
+ "eval_bleu-1": 55.754253322892644,
79
+ "eval_bleu-2": 41.79377630362248,
80
+ "eval_bleu-3": 32.66421259141236,
81
+ "eval_bleu-4": 25.99836335930296,
82
+ "eval_loss": 2.113940715789795,
83
+ "eval_runtime": 71.6669,
84
+ "eval_sacrebleu": 25.99836335930296,
85
+ "eval_samples_per_second": 22.563,
86
+ "eval_steps_per_second": 0.712,
87
+ "eval_wer": 59.771254793071535,
88
+ "step": 3100
89
+ },
90
+ {
91
+ "epoch": 10.0,
92
+ "learning_rate": 0.0005,
93
+ "loss": 2.0896,
94
+ "step": 3875
95
+ },
96
+ {
97
+ "epoch": 10.0,
98
+ "eval_bleu-1": 56.6433519912154,
99
+ "eval_bleu-2": 42.75430114819028,
100
+ "eval_bleu-3": 33.618453012644544,
101
+ "eval_bleu-4": 26.883103502976656,
102
+ "eval_loss": 2.1048648357391357,
103
+ "eval_runtime": 71.6327,
104
+ "eval_sacrebleu": 26.883103502976653,
105
+ "eval_samples_per_second": 22.573,
106
+ "eval_steps_per_second": 0.712,
107
+ "eval_wer": 59.1498082771387,
108
+ "step": 3875
109
+ },
110
+ {
111
+ "epoch": 12.0,
112
+ "learning_rate": 0.0005,
113
+ "loss": 2.0782,
114
+ "step": 4650
115
+ },
116
+ {
117
+ "epoch": 12.0,
118
+ "eval_bleu-1": 57.36710667142347,
119
+ "eval_bleu-2": 43.43036904164615,
120
+ "eval_bleu-3": 34.180841419607404,
121
+ "eval_bleu-4": 27.351050094998076,
122
+ "eval_loss": 2.102236747741699,
123
+ "eval_runtime": 70.5957,
124
+ "eval_sacrebleu": 27.351050094998076,
125
+ "eval_samples_per_second": 22.905,
126
+ "eval_steps_per_second": 0.722,
127
+ "eval_wer": 58.67380669046675,
128
+ "step": 4650
129
+ },
130
+ {
131
+ "epoch": 14.0,
132
+ "learning_rate": 0.0005,
133
+ "loss": 2.0679,
134
+ "step": 5425
135
+ },
136
+ {
137
+ "epoch": 14.0,
138
+ "eval_bleu-1": 57.54398436111605,
139
+ "eval_bleu-2": 43.783579541899066,
140
+ "eval_bleu-3": 34.5851936909885,
141
+ "eval_bleu-4": 27.78839897218679,
142
+ "eval_loss": 2.1011126041412354,
143
+ "eval_runtime": 72.1101,
144
+ "eval_sacrebleu": 27.78839897218679,
145
+ "eval_samples_per_second": 22.424,
146
+ "eval_steps_per_second": 0.707,
147
+ "eval_wer": 58.66719555731852,
148
+ "step": 5425
149
+ },
150
+ {
151
+ "epoch": 16.02,
152
+ "learning_rate": 0.0005,
153
+ "loss": 2.0584,
154
+ "step": 6200
155
+ },
156
+ {
157
+ "epoch": 16.02,
158
+ "eval_bleu-1": 57.56515601474742,
159
+ "eval_bleu-2": 43.81704921199358,
160
+ "eval_bleu-3": 34.676312912246864,
161
+ "eval_bleu-4": 27.854480124541276,
162
+ "eval_loss": 2.102771282196045,
163
+ "eval_runtime": 55.4215,
164
+ "eval_sacrebleu": 27.854480124541286,
165
+ "eval_samples_per_second": 29.176,
166
+ "eval_steps_per_second": 0.92,
167
+ "eval_wer": 58.49530609546476,
168
+ "step": 6200
169
+ },
170
+ {
171
+ "epoch": 18.02,
172
+ "learning_rate": 0.0005,
173
+ "loss": 2.0493,
174
+ "step": 6975
175
+ },
176
+ {
177
+ "epoch": 18.02,
178
+ "eval_bleu-1": 58.31903945111492,
179
+ "eval_bleu-2": 44.49232668400205,
180
+ "eval_bleu-3": 35.16701039076612,
181
+ "eval_bleu-4": 28.215887680578,
182
+ "eval_loss": 2.106776237487793,
183
+ "eval_runtime": 55.6637,
184
+ "eval_sacrebleu": 28.215887680578007,
185
+ "eval_samples_per_second": 29.049,
186
+ "eval_steps_per_second": 0.916,
187
+ "eval_wer": 57.80113711490149,
188
+ "step": 6975
189
+ },
190
+ {
191
+ "epoch": 20.02,
192
+ "learning_rate": 0.0005,
193
+ "loss": 2.0401,
194
+ "step": 7750
195
+ },
196
+ {
197
+ "epoch": 20.02,
198
+ "eval_bleu-1": 58.11736257346271,
199
+ "eval_bleu-2": 44.6065064549101,
200
+ "eval_bleu-3": 35.501910391662186,
201
+ "eval_bleu-4": 28.68772645989158,
202
+ "eval_loss": 2.1100265979766846,
203
+ "eval_runtime": 121.7006,
204
+ "eval_sacrebleu": 28.687726459891582,
205
+ "eval_samples_per_second": 13.287,
206
+ "eval_steps_per_second": 0.419,
207
+ "eval_wer": 57.96641544360703,
208
+ "step": 7750
209
+ },
210
+ {
211
+ "epoch": 22.02,
212
+ "learning_rate": 0.0005,
213
+ "loss": 2.0317,
214
+ "step": 8525
215
+ },
216
+ {
217
+ "epoch": 22.02,
218
+ "eval_bleu-1": 56.11123826962692,
219
+ "eval_bleu-2": 42.90971903145265,
220
+ "eval_bleu-3": 34.051516980448184,
221
+ "eval_bleu-4": 27.48591624207356,
222
+ "eval_loss": 2.109617233276367,
223
+ "eval_runtime": 121.4034,
224
+ "eval_sacrebleu": 27.48591624207356,
225
+ "eval_samples_per_second": 13.319,
226
+ "eval_steps_per_second": 0.42,
227
+ "eval_wer": 62.045484596059765,
228
+ "step": 8525
229
+ },
230
+ {
231
+ "epoch": 24.02,
232
+ "learning_rate": 0.0005,
233
+ "loss": 2.0228,
234
+ "step": 9300
235
+ },
236
+ {
237
+ "epoch": 24.02,
238
+ "eval_bleu-1": 55.77956212812663,
239
+ "eval_bleu-2": 42.8248230377153,
240
+ "eval_bleu-3": 34.044724350184055,
241
+ "eval_bleu-4": 27.49157131564679,
242
+ "eval_loss": 2.1198818683624268,
243
+ "eval_runtime": 139.0903,
244
+ "eval_sacrebleu": 27.491571315646805,
245
+ "eval_samples_per_second": 11.626,
246
+ "eval_steps_per_second": 0.367,
247
+ "eval_wer": 63.14293269866455,
248
+ "step": 9300
249
+ },
250
+ {
251
+ "epoch": 26.02,
252
+ "learning_rate": 0.0005,
253
+ "loss": 2.0145,
254
+ "step": 10075
255
+ },
256
+ {
257
+ "epoch": 26.02,
258
+ "eval_bleu-1": 58.14441655148197,
259
+ "eval_bleu-2": 44.64384618349215,
260
+ "eval_bleu-3": 35.550465547095214,
261
+ "eval_bleu-4": 28.763128628752966,
262
+ "eval_loss": 2.122112989425659,
263
+ "eval_runtime": 53.5948,
264
+ "eval_sacrebleu": 28.76312862875297,
265
+ "eval_samples_per_second": 30.171,
266
+ "eval_steps_per_second": 0.952,
267
+ "eval_wer": 57.7085812508264,
268
+ "step": 10075
269
+ },
270
+ {
271
+ "epoch": 28.02,
272
+ "learning_rate": 0.0005,
273
+ "loss": 2.0058,
274
+ "step": 10850
275
+ },
276
+ {
277
+ "epoch": 28.02,
278
+ "eval_bleu-1": 57.257595157723195,
279
+ "eval_bleu-2": 44.08830411429965,
280
+ "eval_bleu-3": 35.20513160246965,
281
+ "eval_bleu-4": 28.566682599296062,
282
+ "eval_loss": 2.1236658096313477,
283
+ "eval_runtime": 127.9309,
284
+ "eval_sacrebleu": 28.566682599296072,
285
+ "eval_samples_per_second": 12.64,
286
+ "eval_steps_per_second": 0.399,
287
+ "eval_wer": 60.419145841597256,
288
+ "step": 10850
289
+ },
290
+ {
291
+ "epoch": 30.02,
292
+ "learning_rate": 0.0005,
293
+ "loss": 1.9975,
294
+ "step": 11625
295
+ },
296
+ {
297
+ "epoch": 30.02,
298
+ "eval_bleu-1": 58.98909694240342,
299
+ "eval_bleu-2": 45.66022502382035,
300
+ "eval_bleu-3": 36.70586989359544,
301
+ "eval_bleu-4": 29.97126821077391,
302
+ "eval_loss": 2.1420726776123047,
303
+ "eval_runtime": 55.6751,
304
+ "eval_sacrebleu": 29.97126821077392,
305
+ "eval_samples_per_second": 29.044,
306
+ "eval_steps_per_second": 0.916,
307
+ "eval_wer": 57.24580193045088,
308
+ "step": 11625
309
+ }
310
+ ],
311
+ "max_steps": 400000,
312
+ "num_train_epochs": 1034,
313
+ "total_flos": 0.0,
314
+ "trial_name": null,
315
+ "trial_params": null
316
+ }
models/checkpoint-11625/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0addc13b3d794209c846b20ff9913edc88c196b467c3be77cfe9c8169038bb99
3
+ size 4728
models/checkpoint_836_400000.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9087f47602401c07c44a9c70164ca1387373938ac4642149126a43e61c20cd7d
3
+ size 1057496908
models/dinov2face.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ebf07e5720a2ccbee559a89f2a8afd7b85ba14bf9ff156842cefb3de5be59bcb
3
+ size 112850895
models/dinov2hand.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6ffaf9fbf57b906a5f288b8dbbbeb8651a5cf7a37623fa284cb8e56bf640d85
3
+ size 112850895
models/face_landmarker_v2_with_blendshapes.task ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e0044e474913045fff3ace464ab14db1646f5f49b42f7c38cacc5f85a5bb8e9
3
+ size 11577701
models/hand_landmarker.task ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fbc2a30080c3c557093b5ddfc334698132eb341044ccee322ccf8bcf3607cde1
3
+ size 7819105
models/yolov8n.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f59b3d833e2ff32e194b5bb8e08d211dc7c5bdf144b90d2c8412c47ccfc83b36
3
+ size 6549796