Added model files and documentation

README.md

@@ -0,0 +1,69 @@
+# CarD-T: Carcinogen Detection via Transformers
+
+## Overview
+
+CarD-T (Carcinogen Detection via Transformers) is a novel text analytics approach that combines transformer-based machine learning with probabilistic statistical analysis to efficiently nominate carcinogens from scientific texts. This model is designed to address the challenges faced by current systems in managing the burgeoning biomedical literature related to carcinogen identification and classification.
+
+## Model Details
+
+- **Architecture**: Based on Bio-ELECTRA, a 335 million parameter language model
+- **Training Data**: PubMed abstracts featuring known carcinogens from International Agency for Research on Cancer (IARC) groups G1 and G2A
+- **Task**: Named Entity Recognition (NER) for carcinogen identification
+- **Performance**: 
+  - Precision: 0.894
+  - Recall: 0.857
+  - F1 Score: 0.875
+
+## Features
+
+- Efficient nomination of potential carcinogens from scientific literature
+- Context classifier to enhance accuracy and manage computational demands
+- Capable of identifying both chemical and non-chemical carcinogenic factors
+- Trained on a comprehensive dataset of carcinogen-related abstracts from 2000-2024
+
+## Use Cases
+
+- Streamlining toxicogenomic literature reviews
+- Identifying potential carcinogens for further investigation
+- Augmenting existing carcinogen databases with emerging candidates
+
+## Limitations
+
+- Identifies potential candidates, not confirmed carcinogens
+- Analysis limited to abstract-level information
+- May be influenced by publication trends and research focus shifts
+
+
+## Citation
+
+If you use this model in your research, please cite:
+
+O'Neill, J., Reddy, G.A., Dhillon, N., Tripathi, O., Alexandrov, L., & Katira, P. (2024). CarD-T: Interpreting Carcinomic Lexicon via Transformers. BioRxiv.
+
+## License
+
+MIT License
+
+Copyright (c) 2024 Jamey O'Neill
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+
+## Contact
+
+For questions and feedback, please contact Jamey ONeill at [email protected].

checkpoint-10595/config.json

@@ -0,0 +1,48 @@
+{
+  "_name_or_path": "sultan/BioM-ELECTRA-Large-SQuAD2",
+  "architectures": [
+    "ElectraForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "embedding_size": 1024,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "O",
+    "1": "B-cancertype",
+    "2": "B-carcinogen",
+    "3": "B-negative",
+    "4": "I-cancertype",
+    "5": "I-carcinogen",
+    "6": "I-negative"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "B-cancertype": 1,
+    "B-carcinogen": 2,
+    "B-negative": 3,
+    "I-cancertype": 4,
+    "I-carcinogen": 5,
+    "I-negative": 6,
+    "O": 0
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "electra",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "summary_activation": "gelu",
+  "summary_last_dropout": 0.1,
+  "summary_type": "first",
+  "summary_use_proj": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.40.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28895
+}

checkpoint-10595/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1152dab87a3cb1cf2d6d8812b793cf31390c77c430ba4c20736601390303770a
+size 1329781612

checkpoint-10595/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:15a0c6be3b8cad9cacda15dab0ad3bfb05a3f313d063e536180c0ccf651b85f1
+size 2659794471

checkpoint-10595/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17024af815d4eee454cd54e1c9d6acbc87c00976e22d68b0fae1dc0f611fdbe4
+size 14244

checkpoint-10595/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc25d54a6f628bcee11085d2055063ab51e6a482a0b0a73494d5bd691588b7b4
+size 1064

checkpoint-10595/trainer_state.json

@@ -0,0 +1,228 @@
+{
+  "best_metric": 0.0534161813557148,
+  "best_model_checkpoint": "/storage/BioM-ELECTRA-Large-SQuAD2_carDB_5e_neg_lg_SGD/checkpoint-4238",
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 10595,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.23596035865974516,
+      "grad_norm": 1.2747831344604492,
+      "learning_rate": 1.9063709296838132e-05,
+      "loss": 0.1626,
+      "step": 500
+    },
+    {
+      "epoch": 0.4719207173194903,
+      "grad_norm": 0.4436088800430298,
+      "learning_rate": 1.8119867862199152e-05,
+      "loss": 0.0822,
+      "step": 1000
+    },
+    {
+      "epoch": 0.7078810759792354,
+      "grad_norm": 1.6280406713485718,
+      "learning_rate": 1.7176026427560172e-05,
+      "loss": 0.0789,
+      "step": 1500
+    },
+    {
+      "epoch": 0.9438414346389806,
+      "grad_norm": 2.336548089981079,
+      "learning_rate": 1.6232184992921192e-05,
+      "loss": 0.0706,
+      "step": 2000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy_score": 0.9812832660700708,
+      "eval_f1": 0.8218279984573852,
+      "eval_loss": 0.05883064866065979,
+      "eval_precision": 0.8056710775047259,
+      "eval_recall": 0.8386462022825659,
+      "eval_runtime": 5.2595,
+      "eval_samples_per_second": 358.209,
+      "eval_steps_per_second": 44.871,
+      "step": 2119
+    },
+    {
+      "epoch": 1.1798017932987257,
+      "grad_norm": 2.0369768142700195,
+      "learning_rate": 1.528834355828221e-05,
+      "loss": 0.0496,
+      "step": 2500
+    },
+    {
+      "epoch": 1.415762151958471,
+      "grad_norm": 0.6217797994613647,
+      "learning_rate": 1.4344502123643229e-05,
+      "loss": 0.0487,
+      "step": 3000
+    },
+    {
+      "epoch": 1.651722510618216,
+      "grad_norm": 0.8187075257301331,
+      "learning_rate": 1.3400660689004247e-05,
+      "loss": 0.0455,
+      "step": 3500
+    },
+    {
+      "epoch": 1.8876828692779613,
+      "grad_norm": 1.2321038246154785,
+      "learning_rate": 1.2456819254365267e-05,
+      "loss": 0.0439,
+      "step": 4000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy_score": 0.9828429938975649,
+      "eval_f1": 0.8423466462832028,
+      "eval_loss": 0.0534161813557148,
+      "eval_precision": 0.8187221396731055,
+      "eval_recall": 0.867375049193231,
+      "eval_runtime": 5.2808,
+      "eval_samples_per_second": 356.764,
+      "eval_steps_per_second": 44.69,
+      "step": 4238
+    },
+    {
+      "epoch": 2.1236432279377064,
+      "grad_norm": 1.0647934675216675,
+      "learning_rate": 1.1512977819726287e-05,
+      "loss": 0.0337,
+      "step": 4500
+    },
+    {
+      "epoch": 2.3596035865974514,
+      "grad_norm": 0.07234682887792587,
+      "learning_rate": 1.0569136385087306e-05,
+      "loss": 0.0247,
+      "step": 5000
+    },
+    {
+      "epoch": 2.595563945257197,
+      "grad_norm": 2.0315685272216797,
+      "learning_rate": 9.625294950448326e-06,
+      "loss": 0.0263,
+      "step": 5500
+    },
+    {
+      "epoch": 2.831524303916942,
+      "grad_norm": 0.034490641206502914,
+      "learning_rate": 8.681453515809346e-06,
+      "loss": 0.027,
+      "step": 6000
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy_score": 0.9833888986371878,
+      "eval_f1": 0.853031465848043,
+      "eval_loss": 0.06263311207294464,
+      "eval_precision": 0.832272557094721,
+      "eval_recall": 0.8748524203069658,
+      "eval_runtime": 5.3057,
+      "eval_samples_per_second": 355.091,
+      "eval_steps_per_second": 44.481,
+      "step": 6357
+    },
+    {
+      "epoch": 3.067484662576687,
+      "grad_norm": 0.1344643086194992,
+      "learning_rate": 7.739499764039643e-06,
+      "loss": 0.0254,
+      "step": 6500
+    },
+    {
+      "epoch": 3.303445021236432,
+      "grad_norm": 0.07355394959449768,
+      "learning_rate": 6.795658329400662e-06,
+      "loss": 0.0163,
+      "step": 7000
+    },
+    {
+      "epoch": 3.5394053798961775,
+      "grad_norm": 0.7838532328605652,
+      "learning_rate": 5.85181689476168e-06,
+      "loss": 0.016,
+      "step": 7500
+    },
+    {
+      "epoch": 3.7753657385559225,
+      "grad_norm": 3.0708000659942627,
+      "learning_rate": 4.9079754601227e-06,
+      "loss": 0.0138,
+      "step": 8000
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy_score": 0.9838763135832798,
+      "eval_f1": 0.8535911602209945,
+      "eval_loss": 0.06983982026576996,
+      "eval_precision": 0.8559556786703602,
+      "eval_recall": 0.8512396694214877,
+      "eval_runtime": 5.271,
+      "eval_samples_per_second": 357.425,
+      "eval_steps_per_second": 44.773,
+      "step": 8476
+    },
+    {
+      "epoch": 4.011326097215668,
+      "grad_norm": 0.10571020096540451,
+      "learning_rate": 3.966021708352997e-06,
+      "loss": 0.0144,
+      "step": 8500
+    },
+    {
+      "epoch": 4.247286455875413,
+      "grad_norm": 0.029509373009204865,
+      "learning_rate": 3.0221802737140165e-06,
+      "loss": 0.0086,
+      "step": 9000
+    },
+    {
+      "epoch": 4.483246814535158,
+      "grad_norm": 1.1270127296447754,
+      "learning_rate": 2.0783388390750357e-06,
+      "loss": 0.0066,
+      "step": 9500
+    },
+    {
+      "epoch": 4.719207173194903,
+      "grad_norm": 0.010910986922681332,
+      "learning_rate": 1.134497404436055e-06,
+      "loss": 0.0097,
+      "step": 10000
+    },
+    {
+      "epoch": 4.955167531854649,
+      "grad_norm": 1.0645971298217773,
+      "learning_rate": 1.906559697970741e-07,
+      "loss": 0.0076,
+      "step": 10500
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy_score": 0.9837593339962176,
+      "eval_f1": 0.8539891556932611,
+      "eval_loss": 0.07769417762756348,
+      "eval_precision": 0.8406404879908502,
+      "eval_recall": 0.8677685950413223,
+      "eval_runtime": 5.252,
+      "eval_samples_per_second": 358.721,
+      "eval_steps_per_second": 44.935,
+      "step": 10595
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 10595,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "total_flos": 1.7692937221754256e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-10595/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07af839673be954537faf350c212a49f3f745cf8247da4cb4f9a1d9f73b78ed7
+size 5048

checkpoint-3204/config.json

@@ -0,0 +1,50 @@
+{
+  "_name_or_path": "sultan/BioM-ELECTRA-Large-SQuAD2",
+  "architectures": [
+    "ElectraForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "embedding_size": 1024,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "O",
+    "1": "B-antineoplastic",
+    "2": "B-cancertype",
+    "3": "B-carcinogen",
+    "4": "B-negative",
+    "5": "I-cancertype",
+    "6": "I-carcinogen",
+    "7": "I-negative"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "B-antineoplastic": 1,
+    "B-cancertype": 2,
+    "B-carcinogen": 3,
+    "B-negative": 4,
+    "I-cancertype": 5,
+    "I-carcinogen": 6,
+    "I-negative": 7,
+    "O": 0
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "electra",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "summary_activation": "gelu",
+  "summary_last_dropout": 0.1,
+  "summary_type": "first",
+  "summary_use_proj": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.40.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28895
+}

checkpoint-3204/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ca26bb96ef03d72f27f204e9972313d8190bf4c74542e7b5ba3c035bd9011e4
+size 1329785712

checkpoint-3204/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f27b65e1ecc59b25c4d7e5c7daf5576fa34b0dfa99c71f5de7be4333aec324da
+size 2659802663

checkpoint-3204/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:223e37654453351bb53133440aabdb815af7d246c62afead675019647840ae7f
+size 14244

checkpoint-3204/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a90c28e7909084a367599a7cdab33ed7a9cb38d9886e04e2a407c17d10355fba
+size 1064

checkpoint-3204/trainer_state.json

@@ -0,0 +1,87 @@
+{
+  "best_metric": 0.07832363992929459,
+  "best_model_checkpoint": "/storage/BioM-ELECTRA-Large-SQuAD2_carDB_5e_neg_lg_SGD/checkpoint-3204",
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 3204,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.3121098626716604,
+      "grad_norm": 3.2046010494232178,
+      "learning_rate": 4.690387016229713e-05,
+      "loss": 0.1789,
+      "step": 500
+    },
+    {
+      "epoch": 0.6242197253433208,
+      "grad_norm": 1.161939263343811,
+      "learning_rate": 4.378277153558052e-05,
+      "loss": 0.1113,
+      "step": 1000
+    },
+    {
+      "epoch": 0.9363295880149812,
+      "grad_norm": 2.1341910362243652,
+      "learning_rate": 4.066167290886392e-05,
+      "loss": 0.1054,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy_score": 0.9690997325938397,
+      "eval_f1": 0.7794316644113668,
+      "eval_loss": 0.08875501155853271,
+      "eval_precision": 0.7747886241352806,
+      "eval_recall": 0.7841306884480747,
+      "eval_runtime": 4.093,
+      "eval_samples_per_second": 348.153,
+      "eval_steps_per_second": 43.733,
+      "step": 1602
+    },
+    {
+      "epoch": 1.2484394506866416,
+      "grad_norm": 0.97569340467453,
+      "learning_rate": 3.754057428214732e-05,
+      "loss": 0.079,
+      "step": 2000
+    },
+    {
+      "epoch": 1.5605493133583022,
+      "grad_norm": 0.8073092699050903,
+      "learning_rate": 3.441947565543071e-05,
+      "loss": 0.0685,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8726591760299627,
+      "grad_norm": 2.6909968852996826,
+      "learning_rate": 3.129837702871411e-05,
+      "loss": 0.0671,
+      "step": 3000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy_score": 0.973506982271962,
+      "eval_f1": 0.814260389994091,
+      "eval_loss": 0.07832363992929459,
+      "eval_precision": 0.8248204309656824,
+      "eval_recall": 0.8039673278879813,
+      "eval_runtime": 4.1937,
+      "eval_samples_per_second": 339.796,
+      "eval_steps_per_second": 42.683,
+      "step": 3204
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 8010,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "total_flos": 4245597826238592.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-3204/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36534a66cd54fa62d55a78ae7da66791486f97dbf75f55360327b517cd4e0be3
+size 5048

checkpoint-8010/config.json

@@ -0,0 +1,50 @@
+{
+  "_name_or_path": "sultan/BioM-ELECTRA-Large-SQuAD2",
+  "architectures": [
+    "ElectraForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "embedding_size": 1024,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "O",
+    "1": "B-antineoplastic",
+    "2": "B-cancertype",
+    "3": "B-carcinogen",
+    "4": "B-negative",
+    "5": "I-cancertype",
+    "6": "I-carcinogen",
+    "7": "I-negative"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "B-antineoplastic": 1,
+    "B-cancertype": 2,
+    "B-carcinogen": 3,
+    "B-negative": 4,
+    "I-cancertype": 5,
+    "I-carcinogen": 6,
+    "I-negative": 7,
+    "O": 0
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "electra",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "summary_activation": "gelu",
+  "summary_last_dropout": 0.1,
+  "summary_type": "first",
+  "summary_use_proj": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.40.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28895
+}

checkpoint-8010/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28ff692670e07e3443424d23688d7f8c5f7ab04b3a81f11285f1ae8c826d24d3
+size 1329785712

checkpoint-8010/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75b61d9f1a1c4a763eefe45bc88b93a6caaff5621848dba6693e1c84f9e02b5f
+size 2659802663

checkpoint-8010/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24255887216b6ed59d1dea7f442c2bf9f0d4baf66cc6133b44860e5e7a22bdb9
+size 14244

checkpoint-8010/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b119d907611dd4a92192cad349abbf6251e9f178cd9bdccbc47a3fd72a25015
+size 1064

checkpoint-8010/trainer_state.json

@@ -0,0 +1,193 @@
+{
+  "best_metric": 0.07832363992929459,
+  "best_model_checkpoint": "/storage/BioM-ELECTRA-Large-SQuAD2_carDB_5e_neg_lg_SGD/checkpoint-3204",
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 8010,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.3121098626716604,
+      "grad_norm": 3.2046010494232178,
+      "learning_rate": 4.690387016229713e-05,
+      "loss": 0.1789,
+      "step": 500
+    },
+    {
+      "epoch": 0.6242197253433208,
+      "grad_norm": 1.161939263343811,
+      "learning_rate": 4.378277153558052e-05,
+      "loss": 0.1113,
+      "step": 1000
+    },
+    {
+      "epoch": 0.9363295880149812,
+      "grad_norm": 2.1341910362243652,
+      "learning_rate": 4.066167290886392e-05,
+      "loss": 0.1054,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy_score": 0.9690997325938397,
+      "eval_f1": 0.7794316644113668,
+      "eval_loss": 0.08875501155853271,
+      "eval_precision": 0.7747886241352806,
+      "eval_recall": 0.7841306884480747,
+      "eval_runtime": 4.093,
+      "eval_samples_per_second": 348.153,
+      "eval_steps_per_second": 43.733,
+      "step": 1602
+    },
+    {
+      "epoch": 1.2484394506866416,
+      "grad_norm": 0.97569340467453,
+      "learning_rate": 3.754057428214732e-05,
+      "loss": 0.079,
+      "step": 2000
+    },
+    {
+      "epoch": 1.5605493133583022,
+      "grad_norm": 0.8073092699050903,
+      "learning_rate": 3.441947565543071e-05,
+      "loss": 0.0685,
+      "step": 2500
+    },
+    {
+      "epoch": 1.8726591760299627,
+      "grad_norm": 2.6909968852996826,
+      "learning_rate": 3.129837702871411e-05,
+      "loss": 0.0671,
+      "step": 3000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy_score": 0.973506982271962,
+      "eval_f1": 0.814260389994091,
+      "eval_loss": 0.07832363992929459,
+      "eval_precision": 0.8248204309656824,
+      "eval_recall": 0.8039673278879813,
+      "eval_runtime": 4.1937,
+      "eval_samples_per_second": 339.796,
+      "eval_steps_per_second": 42.683,
+      "step": 3204
+    },
+    {
+      "epoch": 2.184769038701623,
+      "grad_norm": 0.14074726402759552,
+      "learning_rate": 2.818352059925094e-05,
+      "loss": 0.0534,
+      "step": 3500
+    },
+    {
+      "epoch": 2.4968789013732833,
+      "grad_norm": 0.8002931475639343,
+      "learning_rate": 2.5062421972534333e-05,
+      "loss": 0.043,
+      "step": 4000
+    },
+    {
+      "epoch": 2.808988764044944,
+      "grad_norm": 1.326310157775879,
+      "learning_rate": 2.194132334581773e-05,
+      "loss": 0.0433,
+      "step": 4500
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy_score": 0.9748935327324948,
+      "eval_f1": 0.8186646433990896,
+      "eval_loss": 0.09053385257720947,
+      "eval_precision": 0.7989633469085524,
+      "eval_recall": 0.839362115908207,
+      "eval_runtime": 4.09,
+      "eval_samples_per_second": 348.41,
+      "eval_steps_per_second": 43.765,
+      "step": 4806
+    },
+    {
+      "epoch": 3.1210986267166043,
+      "grad_norm": 0.9473728537559509,
+      "learning_rate": 1.8820224719101125e-05,
+      "loss": 0.0318,
+      "step": 5000
+    },
+    {
+      "epoch": 3.4332084893882646,
+      "grad_norm": 0.391970157623291,
+      "learning_rate": 1.569912609238452e-05,
+      "loss": 0.0228,
+      "step": 5500
+    },
+    {
+      "epoch": 3.7453183520599254,
+      "grad_norm": 5.073329448699951,
+      "learning_rate": 1.258426966292135e-05,
+      "loss": 0.0224,
+      "step": 6000
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy_score": 0.9751906506883232,
+      "eval_f1": 0.8262295081967214,
+      "eval_loss": 0.09821399301290512,
+      "eval_precision": 0.819433817903596,
+      "eval_recall": 0.8331388564760793,
+      "eval_runtime": 4.1062,
+      "eval_samples_per_second": 347.04,
+      "eval_steps_per_second": 43.593,
+      "step": 6408
+    },
+    {
+      "epoch": 4.057428214731585,
+      "grad_norm": 0.055937759578228,
+      "learning_rate": 9.463171036204745e-06,
+      "loss": 0.0204,
+      "step": 6500
+    },
+    {
+      "epoch": 4.369538077403246,
+      "grad_norm": 0.6860196590423584,
+      "learning_rate": 6.342072409488139e-06,
+      "loss": 0.0116,
+      "step": 7000
+    },
+    {
+      "epoch": 4.681647940074907,
+      "grad_norm": 0.6975510716438293,
+      "learning_rate": 3.2209737827715358e-06,
+      "loss": 0.011,
+      "step": 7500
+    },
+    {
+      "epoch": 4.9937578027465666,
+      "grad_norm": 0.33860084414482117,
+      "learning_rate": 9.987515605493134e-08,
+      "loss": 0.0091,
+      "step": 8000
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy_score": 0.9755125284738041,
+      "eval_f1": 0.8289676425269646,
+      "eval_loss": 0.11285488307476044,
+      "eval_precision": 0.8210606638687524,
+      "eval_recall": 0.8370283936211591,
+      "eval_runtime": 4.1141,
+      "eval_samples_per_second": 346.37,
+      "eval_steps_per_second": 43.509,
+      "step": 8010
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 8010,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "total_flos": 1.064813249753856e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-8010/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36534a66cd54fa62d55a78ae7da66791486f97dbf75f55360327b517cd4e0be3
+size 5048

config.json

@@ -0,0 +1,48 @@
+{
+  "_name_or_path": "sultan/BioM-ELECTRA-Large-SQuAD2",
+  "architectures": [
+    "ElectraForTokenClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "embedding_size": 1024,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "O",
+    "1": "B-cancertype",
+    "2": "B-carcinogen",
+    "3": "B-negative",
+    "4": "I-cancertype",
+    "5": "I-carcinogen",
+    "6": "I-negative"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "B-cancertype": 1,
+    "B-carcinogen": 2,
+    "B-negative": 3,
+    "I-cancertype": 4,
+    "I-carcinogen": 5,
+    "I-negative": 6,
+    "O": 0
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "electra",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "summary_activation": "gelu",
+  "summary_last_dropout": 0.1,
+  "summary_type": "first",
+  "summary_use_proj": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.40.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28895
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2fc3d970de77d2af227efab4238d737a8226de685bea1a93bc42848c723b9e9a
+size 1329781612

runs/Apr25_14-35-27_hpcf/events.out.tfevents.1714080930.hpcf.137937.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64afb7a940a161fa73270408d5c7bdfd08cd268948f591dde724000e76873b33
+size 11444

runs/Apr26_09-58-52_hpcf/events.out.tfevents.1714150736.hpcf.183226.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de80d861e08c3f13afa8fec93cdc5f4e42c442309baa5f920daa4b9cdaa03721
+size 12443

runs/Apr26_10-20-51_hpcf/events.out.tfevents.1714152055.hpcf.183659.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:78ad061897c41c39551a6a3fb59cf510f60d389690b2792b223efbe66fc4cf5b
+size 5268

runs/Apr26_10-21-30_hpcf/events.out.tfevents.1714152093.hpcf.186484.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7036534514f03fdd01de11863b01a94b6e43cbfeba22e6d634a073eeb2b2b5a4
+size 12443

special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

tokenizer_config.json

@@ -0,0 +1,61 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[SEP]",
+      "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
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "ignore_mismatched_sizes": true,
+  "mask_token": "[MASK]",
+  "max_len": 512,
+  "model_max_length": 512,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "padding": true,
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "ElectraTokenizer",
+  "truncation": true,
+  "unk_token": "[UNK]"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07af839673be954537faf350c212a49f3f745cf8247da4cb4f9a1d9f73b78ed7
+size 5048