Upload ProteinVec

config.json

@@ -1,20 +1,120 @@
 {
-  "activation": "relu",
   "architectures": [
     "ProteinVec"
   ],
-  "d_model": 512,
-  "dim_feedforward": 2048,
-  "dropout": 0.1,
-  "embedded_path": "models/protein_vec/src_run/protein_vec_models",
-  "lr0": 0.0001,
-  "nhead": 4,
-  "num_layers": 2,
-  "num_variables": 10,
-  "out_dim": 512,
-  "p_bernoulli": 0.5,
+  "bp_activation": "relu",
+  "bp_d_model": 1024,
+  "bp_dim_feedforward": 2048,
+  "bp_dropout": 0.1,
+  "bp_lr0": 0.0001,
+  "bp_nhead": 4,
+  "bp_num_layers": 4,
+  "bp_num_variables": 10,
+  "bp_out_dim": 512,
+  "bp_p_bernoulli": 0.5,
+  "bp_vocab": 20,
+  "bp_warmup_steps": 500,
+  "cc_activation": "relu",
+  "cc_d_model": 1024,
+  "cc_dim_feedforward": 2048,
+  "cc_dropout": 0.1,
+  "cc_lr0": 0.0001,
+  "cc_nhead": 4,
+  "cc_num_layers": 4,
+  "cc_num_variables": 10,
+  "cc_out_dim": 512,
+  "cc_p_bernoulli": 0.5,
+  "cc_vocab": 20,
+  "cc_warmup_steps": 500,
+  "classifier_dropout": 0.0,
+  "d_ff": 16384,
+  "d_kv": 128,
+  "d_model": 1024,
+  "dense_act_fn": "relu",
+  "dropout_rate": 0.1,
+  "ec_activation": "relu",
+  "ec_d_model": 1024,
+  "ec_dim_feedforward": 2048,
+  "ec_dropout": 0.1,
+  "ec_lr0": 0.0001,
+  "ec_nhead": 4,
+  "ec_num_layers": 2,
+  "ec_num_variables": 10,
+  "ec_out_dim": 512,
+  "ec_p_bernoulli": 0.5,
+  "ec_vocab": 20,
+  "ec_warmup_steps": 500,
+  "eos_token_id": 1,
+  "feed_forward_proj": "relu",
+  "gene3d_activation": "relu",
+  "gene3d_d_model": 1024,
+  "gene3d_dim_feedforward": 2048,
+  "gene3d_dropout": 0.1,
+  "gene3d_lr0": 0.0001,
+  "gene3d_nhead": 4,
+  "gene3d_num_layers": 2,
+  "gene3d_num_variables": 10,
+  "gene3d_out_dim": 512,
+  "gene3d_p_bernoulli": 0.5,
+  "gene3d_vocab": 20,
+  "gene3d_warmup_steps": 500,
+  "initializer_factor": 1.0,
+  "is_gated_act": false,
+  "layer_norm_epsilon": 1e-06,
+  "mf_activation": "relu",
+  "mf_d_model": 1024,
+  "mf_dim_feedforward": 2048,
+  "mf_dropout": 0.1,
+  "mf_lr0": 0.0001,
+  "mf_nhead": 4,
+  "mf_num_layers": 4,
+  "mf_num_variables": 10,
+  "mf_out_dim": 512,
+  "mf_p_bernoulli": 0.5,
+  "mf_vocab": 20,
+  "mf_warmup_steps": 500,
+  "model_type": "t5",
+  "num_decoder_layers": 24,
+  "num_heads": 32,
+  "num_layers": 24,
+  "pad_token_id": 0,
+  "pfam_activation": "relu",
+  "pfam_d_model": 1024,
+  "pfam_dim_feedforward": 2048,
+  "pfam_dropout": 0.1,
+  "pfam_lr0": 0.0001,
+  "pfam_nhead": 4,
+  "pfam_num_layers": 2,
+  "pfam_num_variables": 10,
+  "pfam_out_dim": 512,
+  "pfam_p_bernoulli": 0.5,
+  "pfam_vocab": 20,
+  "pfam_warmup_steps": 500,
+  "relative_attention_max_distance": null,
+  "relative_attention_num_buckets": 32,
+  "tm_activation": "relu",
+  "tm_d_model": 1024,
+  "tm_dim_feedforward": 2048,
+  "tm_dropout": 0.1,
+  "tm_lr0": 0.0001,
+  "tm_nhead": 4,
+  "tm_num_layers": 4,
+  "tm_out_dim": 512,
+  "tm_warmup_steps": 300,
   "torch_dtype": "float32",
-  "transformers_version": "4.38.2",
-  "vocab": 20,
-  "warmup_steps": 500
+  "transformers_version": "4.38.1",
+  "use_cache": true,
+  "vec_activation": "relu",
+  "vec_d_model": 1024,
+  "vec_dim_feedforward": 2048,
+  "vec_dropout": 0.1,
+  "vec_lr0": 0.0001,
+  "vec_nhead": 4,
+  "vec_num_layers": 2,
+  "vec_num_variables": 10,
+  "vec_out_dim": 512,
+  "vec_p_bernoulli": 0.5,
+  "vec_vocab": 20,
+  "vec_warmup_steps": 500,
+  "vocab_size": 128
 }

model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6a2e2a081a78c4d5ce1d72a5e545e211aa6a83140cd6954b097ceff1a2e52d49
-size 4996418440
+oid sha256:ac4ed24c74dce0afa06faf9efdd2e32fa179002491ddc2e8eeac83cb9e93403f
+size 4995336712

model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:365bbe7727ae454be8bb8fdb173a49f0ca762a3afb3b4b6a7055a02f45eb92b3
-size 623719592
+oid sha256:8753bd949ac67f613ff84915e80694d153a9dcae910c3b587522a20360baadc7
+size 667829856

model.safetensors.index.json

@@ -1,6 +1,6 @@
 {
   "metadata": {
-    "total_size": 5620074496
+    "total_size": 5663102976
   },
   "weight_map": {
     "moe.encoder.layers.0.linear1.bias": "model-00001-of-00002.safetensors",
@@ -297,30 +297,30 @@
     "moe.model_aspect_tmvec.encoder.layers.1.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
     "moe.model_aspect_tmvec.encoder.layers.2.linear1.bias": "model-00001-of-00002.safetensors",
     "moe.model_aspect_tmvec.encoder.layers.2.linear1.weight": "model-00001-of-00002.safetensors",
-    "moe.model_aspect_tmvec.encoder.layers.2.linear2.bias": "model-00001-of-00002.safetensors",
-    "moe.model_aspect_tmvec.encoder.layers.2.linear2.weight": "model-00001-of-00002.safetensors",
-    "moe.model_aspect_tmvec.encoder.layers.2.norm1.bias": "model-00001-of-00002.safetensors",
-    "moe.model_aspect_tmvec.encoder.layers.2.norm1.weight": "model-00001-of-00002.safetensors",
-    "moe.model_aspect_tmvec.encoder.layers.2.norm2.bias": "model-00001-of-00002.safetensors",
-    "moe.model_aspect_tmvec.encoder.layers.2.norm2.weight": "model-00001-of-00002.safetensors",
+    "moe.model_aspect_tmvec.encoder.layers.2.linear2.bias": "model-00002-of-00002.safetensors",
+    "moe.model_aspect_tmvec.encoder.layers.2.linear2.weight": "model-00002-of-00002.safetensors",
+    "moe.model_aspect_tmvec.encoder.layers.2.norm1.bias": "model-00002-of-00002.safetensors",
+    "moe.model_aspect_tmvec.encoder.layers.2.norm1.weight": "model-00002-of-00002.safetensors",
+    "moe.model_aspect_tmvec.encoder.layers.2.norm2.bias": "model-00002-of-00002.safetensors",
+    "moe.model_aspect_tmvec.encoder.layers.2.norm2.weight": "model-00002-of-00002.safetensors",
     "moe.model_aspect_tmvec.encoder.layers.2.self_attn.in_proj_bias": "model-00001-of-00002.safetensors",
     "moe.model_aspect_tmvec.encoder.layers.2.self_attn.in_proj_weight": "model-00001-of-00002.safetensors",
     "moe.model_aspect_tmvec.encoder.layers.2.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
     "moe.model_aspect_tmvec.encoder.layers.2.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
-    "moe.model_aspect_tmvec.encoder.layers.3.linear1.bias": "model-00001-of-00002.safetensors",
-    "moe.model_aspect_tmvec.encoder.layers.3.linear1.weight": "model-00001-of-00002.safetensors",
-    "moe.model_aspect_tmvec.encoder.layers.3.linear2.bias": "model-00001-of-00002.safetensors",
-    "moe.model_aspect_tmvec.encoder.layers.3.linear2.weight": "model-00001-of-00002.safetensors",
-    "moe.model_aspect_tmvec.encoder.layers.3.norm1.bias": "model-00001-of-00002.safetensors",
-    "moe.model_aspect_tmvec.encoder.layers.3.norm1.weight": "model-00001-of-00002.safetensors",
-    "moe.model_aspect_tmvec.encoder.layers.3.norm2.bias": "model-00001-of-00002.safetensors",
-    "moe.model_aspect_tmvec.encoder.layers.3.norm2.weight": "model-00001-of-00002.safetensors",
-    "moe.model_aspect_tmvec.encoder.layers.3.self_attn.in_proj_bias": "model-00001-of-00002.safetensors",
-    "moe.model_aspect_tmvec.encoder.layers.3.self_attn.in_proj_weight": "model-00001-of-00002.safetensors",
-    "moe.model_aspect_tmvec.encoder.layers.3.self_attn.out_proj.bias": "model-00001-of-00002.safetensors",
-    "moe.model_aspect_tmvec.encoder.layers.3.self_attn.out_proj.weight": "model-00001-of-00002.safetensors",
-    "moe.model_aspect_tmvec.mlp.bias": "model-00001-of-00002.safetensors",
-    "moe.model_aspect_tmvec.mlp.weight": "model-00001-of-00002.safetensors",
+    "moe.model_aspect_tmvec.encoder.layers.3.linear1.bias": "model-00002-of-00002.safetensors",
+    "moe.model_aspect_tmvec.encoder.layers.3.linear1.weight": "model-00002-of-00002.safetensors",
+    "moe.model_aspect_tmvec.encoder.layers.3.linear2.bias": "model-00002-of-00002.safetensors",
+    "moe.model_aspect_tmvec.encoder.layers.3.linear2.weight": "model-00002-of-00002.safetensors",
+    "moe.model_aspect_tmvec.encoder.layers.3.norm1.bias": "model-00002-of-00002.safetensors",
+    "moe.model_aspect_tmvec.encoder.layers.3.norm1.weight": "model-00002-of-00002.safetensors",
+    "moe.model_aspect_tmvec.encoder.layers.3.norm2.bias": "model-00002-of-00002.safetensors",
+    "moe.model_aspect_tmvec.encoder.layers.3.norm2.weight": "model-00002-of-00002.safetensors",
+    "moe.model_aspect_tmvec.encoder.layers.3.self_attn.in_proj_bias": "model-00002-of-00002.safetensors",
+    "moe.model_aspect_tmvec.encoder.layers.3.self_attn.in_proj_weight": "model-00002-of-00002.safetensors",
+    "moe.model_aspect_tmvec.encoder.layers.3.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "moe.model_aspect_tmvec.encoder.layers.3.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "moe.model_aspect_tmvec.mlp.bias": "model-00002-of-00002.safetensors",
+    "moe.model_aspect_tmvec.mlp.weight": "model-00002-of-00002.safetensors",
     "t5.encoder.block.0.layer.0.SelfAttention.k.weight": "model-00001-of-00002.safetensors",
     "t5.encoder.block.0.layer.0.SelfAttention.o.weight": "model-00001-of-00002.safetensors",
     "t5.encoder.block.0.layer.0.SelfAttention.q.weight": "model-00001-of-00002.safetensors",