Commit
•
7775f30
1
Parent(s):
8504cc2
Upload Isoformer
Browse files- config.json +61 -0
- pytorch_model.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Isoformer"
|
| 4 |
+
],
|
| 5 |
+
"enformer_attn_dim_key": 64,
|
| 6 |
+
"enformer_attn_dropout": 0.05,
|
| 7 |
+
"enformer_depth": 11,
|
| 8 |
+
"enformer_dim": 1536,
|
| 9 |
+
"enformer_dim_divisible_by": 128,
|
| 10 |
+
"enformer_dropout_rate": 0.4,
|
| 11 |
+
"enformer_heads": 8,
|
| 12 |
+
"enformer_num_downsamples": 7,
|
| 13 |
+
"enformer_output_heads": {
|
| 14 |
+
"human": 5313,
|
| 15 |
+
"mouse": 1643
|
| 16 |
+
},
|
| 17 |
+
"enformer_pos_dropout": 0.01,
|
| 18 |
+
"enformer_target_length": -1,
|
| 19 |
+
"enformer_use_checkpointing": false,
|
| 20 |
+
"enformer_use_convnext": false,
|
| 21 |
+
"enformer_use_tf_gamma": false,
|
| 22 |
+
"esm_add_bias_fnn": true,
|
| 23 |
+
"esm_attention_probs_dropout_prob": 0.0,
|
| 24 |
+
"esm_emb_layer_norm_before": false,
|
| 25 |
+
"esm_hidden_dropout_prob": 0.0,
|
| 26 |
+
"esm_hidden_size": 640,
|
| 27 |
+
"esm_intermediate_size": 2560,
|
| 28 |
+
"esm_mask_token_id": 32,
|
| 29 |
+
"esm_max_position_embeddings": 1026,
|
| 30 |
+
"esm_num_attention_heads": 20,
|
| 31 |
+
"esm_num_hidden_layers": 30,
|
| 32 |
+
"esm_pad_token_id": 1,
|
| 33 |
+
"esm_position_embedding_type": "rotary",
|
| 34 |
+
"esm_tie_word_embeddings": false,
|
| 35 |
+
"esm_token_dropout": true,
|
| 36 |
+
"esm_use_cache": false,
|
| 37 |
+
"esm_vocab_size": 33,
|
| 38 |
+
"model_type": "isoformer",
|
| 39 |
+
"nt_add_bias_fnn": false,
|
| 40 |
+
"nt_attention_probs_dropout_prob": 0.0,
|
| 41 |
+
"nt_emb_layer_norm_before": false,
|
| 42 |
+
"nt_hidden_dropout_prob": 0.0,
|
| 43 |
+
"nt_hidden_size": 768,
|
| 44 |
+
"nt_intermediate_size": 3072,
|
| 45 |
+
"nt_mask_token_id": 2,
|
| 46 |
+
"nt_max_position_embeddings": 2050,
|
| 47 |
+
"nt_num_attention_heads": 16,
|
| 48 |
+
"nt_num_hidden_layers": 24,
|
| 49 |
+
"nt_pad_token_id": 1,
|
| 50 |
+
"nt_position_embedding_type": "rotary",
|
| 51 |
+
"nt_tie_word_embeddings": false,
|
| 52 |
+
"nt_token_dropout": false,
|
| 53 |
+
"nt_use_cache": 0.0,
|
| 54 |
+
"nt_vocab_size": 4107,
|
| 55 |
+
"num_heads_omics_cross_attention": 8,
|
| 56 |
+
"num_protein_tokens_per_seq": 1200,
|
| 57 |
+
"num_tokens_per_seq_nuctf": 196608,
|
| 58 |
+
"num_tokens_per_seq_nuctf_rna": 2048,
|
| 59 |
+
"torch_dtype": "float32",
|
| 60 |
+
"transformers_version": "4.29.2"
|
| 61 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1e5b7026887f125f83934f13289e3fcc0611e5e327fe3d03e779f28f28188b6b
|
| 3 |
+
size 2272415646
|