Hugging Face's logo

BGI-Research
/
OrionGeno

Model card Files
xet
 Community
caixudong2001 commited on
Commit
e1fd73c
·
verified ·
1 Parent(s): 2724963

Add other invertebrates checkpoint

Browse files
Files changed (5) hide show
  1. checkpoints/oriongeno_other_invertebrates/config.json +92 -0
  2. checkpoints/oriongeno_other_invertebrates/pytorch_model.bin +3 -0
  3. checkpoints/oriongeno_other_invertebrates/special_tokens_map.json +20 -0
  4. checkpoints/oriongeno_other_invertebrates/tokenizer_config.json +62 -0
  5. checkpoints/oriongeno_other_invertebrates/vocab.txt +15 -0
checkpoints/oriongeno_other_invertebrates/config.json ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "OrionGenoForJointLabeling"
4
+ ],
5
+ "attention_type": "flash_attention_2",
6
+ "attn_cfg": {},
7
+ "attn_layer_idx": [],
8
+ "bidirectional": true,
9
+ "bidirectional_strategy": "add",
10
+ "bidirectional_weight_tie": true,
11
+ "bos_token_id": 14,
12
+ "d_intermediate": 0,
13
+ "d_model": 768,
14
+ "dtype": "bfloat16",
15
+ "eos_token_id": 13,
16
+ "fused_add_norm": true,
17
+ "initializer_cfg": null,
18
+ "keys_to_ignore_at_inference": [
19
+ "gene_logits",
20
+ "repeat_logits",
21
+ "boundary_logits",
22
+ "gene_cross_entropy",
23
+ "repeat_cross_entropy",
24
+ "boundary_focal_loss",
25
+ "hidden_states"
26
+ ],
27
+ "loss_weights": {
28
+ "boundary_focal_loss": 0.25,
29
+ "repeat_cross_entropy": 0.1,
30
+ "state_cross_entropy": 1.0
31
+ },
32
+ "mask_token_id": 10,
33
+ "model_type": "orion_geno",
34
+ "n_layer": 24,
35
+ "norm_epsilon": 1e-05,
36
+ "pad_token_id": 9,
37
+ "pad_vocab_size_multiple": 1,
38
+ "residual_in_fp32": true,
39
+ "rms_norm": true,
40
+ "settings": {
41
+ "channel_growth": 128,
42
+ "decoder_channel_sizes": [
43
+ 768,
44
+ 768,
45
+ 640,
46
+ 512,
47
+ 384
48
+ ],
49
+ "encoder_output_channels": 384,
50
+ "feature_scales": [
51
+ 1,
52
+ 2,
53
+ 4,
54
+ 8
55
+ ],
56
+ "gene_label_count": 20,
57
+ "input_channels": 5,
58
+ "loss": {
59
+ "boundary_label_ids": [
60
+ 7,
61
+ 8,
62
+ 9,
63
+ 10,
64
+ 11,
65
+ 12,
66
+ 13,
67
+ 14,
68
+ 17,
69
+ 19
70
+ ],
71
+ "boundary_loss_gamma": 2.0,
72
+ "boundary_positive_weight": 4.0,
73
+ "intergenic_label_id": 0,
74
+ "state_boundary_weight": 1.6,
75
+ "state_default_weight": 1.0,
76
+ "state_intergenic_weight": 0.25,
77
+ "state_label_smoothing": 0.02,
78
+ "state_near_boundary_distance": 8,
79
+ "state_near_boundary_factor": 1.2
80
+ },
81
+ "pooling_kernel_size": 2,
82
+ "repeat_label_count": 2,
83
+ "use_decoder_skip_connection": false,
84
+ "use_repeat_head": true
85
+ },
86
+ "ssm_cfg": {
87
+ "layer": "Mamba2"
88
+ },
89
+ "transformers_version": "4.57.3",
90
+ "use_second_lm_head": false,
91
+ "vocab_size": 15
92
+ }
checkpoints/oriongeno_other_invertebrates/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81fc34d738eae282d9d41e5f1d2612c611f7a491dc18a9814908587f0536d733
3
+ size 244045419
checkpoints/oriongeno_other_invertebrates/special_tokens_map.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<bos>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": "<cls>",
10
+ "eos_token": {
11
+ "content": "<eos>",
12
+ "lstrip": false,
13
+ "normalized": false,
14
+ "rstrip": false,
15
+ "single_word": false
16
+ },
17
+ "mask_token": "<mask>",
18
+ "pad_token": "<pad>",
19
+ "unk_token": "<unk>"
20
+ }
checkpoints/oriongeno_other_invertebrates/tokenizer_config.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "9": {
4
+ "content": "<pad>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "10": {
12
+ "content": "<mask>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "11": {
20
+ "content": "<unk>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "12": {
28
+ "content": "<cls>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "13": {
36
+ "content": "<eos>",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "14": {
44
+ "content": "<bos>",
45
+ "lstrip": false,
46
+ "normalized": false,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": true
50
+ }
51
+ },
52
+ "bos_token": "<bos>",
53
+ "clean_up_tokenization_spaces": true,
54
+ "cls_token": "<cls>",
55
+ "eos_token": "<eos>",
56
+ "extra_special_tokens": {},
57
+ "mask_token": "<mask>",
58
+ "model_max_length": 51200,
59
+ "pad_token": "<pad>",
60
+ "tokenizer_class": "DnaTokenizer",
61
+ "unk_token": "<unk>"
62
+ }
checkpoints/oriongeno_other_invertebrates/vocab.txt ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ A
2
+ C
3
+ G
4
+ T
5
+ N
6
+ a
7
+ c
8
+ g
9
+ t
10
+ <pad>
11
+ <mask>
12
+ <unk>
13
+ <cls>
14
+ <eos>
15
+ <bos>