Upload RNAFM (#1)

- Upload RNAFM (e0927cab4c4c4f0b8d1470fc823c87b2d74b6ddf)


Co-authored-by: Rodenas <qrodenas@users.noreply.huggingface.co>

multimolecule-rnafm-finetuned/eval_metrics_rnafm.csv

@@ -0,0 +1,37 @@
+,step,eval_loss,eval_seq_accuracy,eval_F1,eval_ideal_threshold
+0,1000,0.05654533579945564,0.26735485658861025,0.41148063781321187,0.377992182970047
+1,2000,0.03999479115009308,0.40667867534986835,0.5678379892739934,0.28405293822288513
+2,3000,0.029324576258659363,0.4106969654981294,0.5714511727247558,0.309300035238266
+3,4000,0.022111676633358,0.42275183594291255,0.5634552566842763,0.2923034131526947
+4,5000,0.01701202243566513,0.4226825550782874,0.5901236525047558,0.25730085372924805
+5,6000,0.013379547744989395,0.43619232368019956,0.5925659472422062,0.32337579131126404
+6,7000,0.010823683813214302,0.41776361368989884,0.5878646441073512,0.2581407427787781
+7,8000,0.008882307447493076,0.4336982125536927,0.5927243969948597,0.3128620982170105
+8,9000,0.007649441249668598,0.4209505334626576,0.5834815956984146,0.2866809666156769
+9,10000,0.006518819835036993,0.4351531107108217,0.597099845904619,0.3071558177471161
+10,11000,0.005697860382497311,0.44104198420396284,0.608356326369728,0.24161586165428162
+11,12000,0.005122509319335222,0.44034917555771097,0.6052124690435945,0.26517346501350403
+12,13000,0.0048288097605109215,0.4484550367188582,0.6015997414559263,0.3307763636112213
+13,14000,0.004520080517977476,0.4085492586947485,0.5873728650930723,0.2596055269241333
+14,15000,0.0042026108130812645,0.46529028682277956,0.6141560798548095,0.30816274881362915
+15,16000,0.004035507794469595,0.4568380213385063,0.6124875324157192,0.28661900758743286
+16,17000,0.0039116437546908855,0.4631425800193986,0.6133354859541492,0.33768612146377563
+17,18000,0.003864646190777421,0.44360537619509494,0.6070835639180963,0.34542742371559143
+18,19000,0.0037449176888912916,0.4678536788139116,0.6191800701414923,0.2799220681190491
+19,20000,0.0037235149648040533,0.46050990716364143,0.6152490789684446,0.3276229500770569
+20,21000,0.0036446144804358482,0.4675765553554108,0.6257604867114953,0.30810418725013733
+21,22000,0.003612295025959611,0.4717334072329223,0.6290419281346177,0.3131464123725891
+22,23000,0.003577238880097866,0.4677843979492864,0.6298631766805473,0.29628777503967285
+23,24000,0.003594768000766635,0.4914091727864764,0.6268041237113402,0.36331790685653687
+24,25000,0.0035471622832119465,0.47692947207981157,0.6357695066549773,0.28095003962516785
+25,26000,0.0035547115840017796,0.48510461410558403,0.628321478211103,0.3544747233390808
+26,27000,0.0035257022827863693,0.4830954690314535,0.6359147296158015,0.29656633734703064
+27,28000,0.0035129161551594734,0.49293335180823056,0.6351279644067105,0.36126482486724854
+28,29000,0.003488227492198348,0.49369544131910764,0.6377565535460272,0.3647649884223938
+29,30000,0.003478726837784052,0.4920327005681031,0.6421196967252782,0.31777191162109375
+30,31000,0.0034667213913053274,0.501108493834003,0.6408713098308971,0.34830623865127563
+31,32000,0.0034414229448884726,0.49653595676874046,0.6438972944469233,0.33935099840164185
+32,33000,0.0034377938136458397,0.4927947900789802,0.643843892222804,0.3039156198501587
+33,34000,0.0034172709565609694,0.5103921296937786,0.6450729927007299,0.3888712227344513
+34,35000,0.003412203397601843,0.49521962034086187,0.6462359505134633,0.33549195528030396
+35,36000,0.003401830093935132,0.49009283635859774,0.6467669654289373,0.3117576837539673

multimolecule-rnafm-finetuned/final_model/config.json

@@ -0,0 +1,54 @@
+{
+  "_name_or_path": "multimolecule/rnafm",
+  "architectures": [
+    "RnaFmForTokenPrediction"
+  ],
+  "attention_dropout": 0.1,
+  "bos_token_id": 1,
+  "codon": false,
+  "emb_layer_norm_before": true,
+  "eos_token_id": 2,
+  "head": {
+    "act": null,
+    "bias": true,
+    "dropout": 0.0,
+    "hidden_size": 640,
+    "layer_norm_eps": 1e-12,
+    "num_labels": 2,
+    "output_name": null,
+    "problem_type": "single_label_classification",
+    "transform": null,
+    "transform_act": "gelu"
+  },
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.1,
+  "hidden_size": 640,
+  "initializer_range": 0.02,
+  "intermediate_size": 5120,
+  "layer_norm_eps": 1e-12,
+  "lm_head": {
+    "act": null,
+    "bias": true,
+    "dropout": 0.0,
+    "hidden_size": null,
+    "layer_norm_eps": 1e-12,
+    "output_name": null,
+    "transform": "nonlinear",
+    "transform_act": "gelu"
+  },
+  "mask_token_id": 4,
+  "max_position_embeddings": 1026,
+  "model_type": "rnafm",
+  "null_token_id": 5,
+  "num_attention_heads": 20,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "token_dropout": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.46.3",
+  "unk_token_id": 3,
+  "use_cache": true,
+  "vocab_size": 26
+}

multimolecule-rnafm-finetuned/final_model/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cad389b97d924dfd7ec0d694e02bb3750248c8d5419f12629ec4d7622e22528c
+size 398111136

multimolecule-rnafm-finetuned/final_model/special_tokens_map.json

@@ -0,0 +1,54 @@
+{
+  "additional_special_tokens": [
+    "<null>"
+  ],
+  "bos_token": {
+    "content": "<cls>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "cls_token": {
+    "content": "<cls>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

multimolecule-rnafm-finetuned/final_model/tokenizer_config.json

@@ -0,0 +1,68 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<cls>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<eos>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "<null>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<null>"
+  ],
+  "bos_token": "<cls>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<cls>",
+  "codon": false,
+  "eos_token": "<eos>",
+  "mask_token": "<mask>",
+  "model_max_length": 1024,
+  "nmers": 1,
+  "pad_token": "<pad>",
+  "replace_T_with_U": true,
+  "sep_token": "<eos>",
+  "tokenizer_class": "RnaTokenizer",
+  "unk_token": "<unk>"
+}

multimolecule-rnafm-finetuned/final_model/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94b874b2919304f950adc67dfe7b8ef0aeccc265daf436f4e203b7dc4acf0470
+size 5368

multimolecule-rnafm-finetuned/final_model/vocab.txt

@@ -0,0 +1,26 @@
+<pad>
+<cls>
+<eos>
+<unk>
+<mask>
+<null>
+A
+C
+G
+U
+N
+R
+Y
+S
+W
+K
+M
+B
+D
+H
+V
+.
+X
+*
+-
+I