Upload 15 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+eval_nbest_predictions.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,3 +1,52 @@
----
-license: apache-2.0
----
+---
+tags:
+- generated_from_trainer
+datasets:
+- squad_v2
+model-index:
+- name: matbert-lr3-seed16
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# matbert-lr3-seed16
+
+This model is a fine-tuned version of [/scratch/project_2006183/matilda/finetune-msb-qa/finetuning-scripts/matbert-base-uncased](https://huggingface.co//scratch/project_2006183/matilda/finetune-msb-qa/finetuning-scripts/matbert-base-uncased) on the squad_v2 dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 3e-05
+- train_batch_size: 12
+- eval_batch_size: 8
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 2.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.20.1
+- Pytorch 1.12.0+cu116
+- Datasets 2.3.2
+- Tokenizers 0.12.1

all_results.json

@@ -0,0 +1,22 @@
+{
+    "epoch": 2.0,
+    "eval_HasAns_exact": 69.88866396761134,
+    "eval_HasAns_f1": 76.60602930598823,
+    "eval_HasAns_total": 5928,
+    "eval_NoAns_exact": 77.44322960470984,
+    "eval_NoAns_f1": 77.44322960470984,
+    "eval_NoAns_total": 5945,
+    "eval_best_exact": 73.67977764676156,
+    "eval_best_exact_thresh": 0.0,
+    "eval_best_f1": 77.03365128660806,
+    "eval_best_f1_thresh": 0.0,
+    "eval_exact": 73.67135517560853,
+    "eval_f1": 77.02522881545504,
+    "eval_samples": 12290,
+    "eval_total": 11873,
+    "train_loss": 1.043660425004505,
+    "train_runtime": 7588.3192,
+    "train_samples": 133054,
+    "train_samples_per_second": 35.068,
+    "train_steps_per_second": 2.922
+}

config.json

@@ -0,0 +1,27 @@
+{
+  "_name_or_path": "/scratch/project_2006183/matilda/finetune-msb-qa/finetuning-scripts/matbert-base-uncased",
+  "architectures": [
+    "BertForQuestionAnswering"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "total_flos": 4.5890014442789274e+17,
+  "transformers_version": "4.20.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

eval_nbest_predictions.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c3b3dc7e8b566fc14ccef8811f0f4744dc9f6f5969dbba385b8ce433a2514c4
+size 55549023

eval_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 2.0,
+    "eval_HasAns_exact": 69.88866396761134,
+    "eval_HasAns_f1": 76.60602930598823,
+    "eval_HasAns_total": 5928,
+    "eval_NoAns_exact": 77.44322960470984,
+    "eval_NoAns_f1": 77.44322960470984,
+    "eval_NoAns_total": 5945,
+    "eval_best_exact": 73.67977764676156,
+    "eval_best_exact_thresh": 0.0,
+    "eval_best_f1": 77.03365128660806,
+    "eval_best_f1_thresh": 0.0,
+    "eval_exact": 73.67135517560853,
+    "eval_f1": 77.02522881545504,
+    "eval_samples": 12290,
+    "eval_total": 11873
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eea4e3d45c502736d913e8db69bf35cf975901c449dc6e998001a7b9734d1672
+size 435643185

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 +1,15 @@
+{
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "name_or_path": "/scratch/project_2006183/matilda/finetune-msb-qa/finetuning-scripts/matbert-base-uncased",
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "special_tokens_map_file": null,
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.0,
+    "train_loss": 1.043660425004505,
+    "train_runtime": 7588.3192,
+    "train_samples": 133054,
+    "train_samples_per_second": 35.068,
+    "train_steps_per_second": 2.922
+}

trainer_state.json

@@ -0,0 +1,289 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "global_step": 22176,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.05,
+      "learning_rate": 2.9323593073593074e-05,
+      "loss": 2.3904,
+      "step": 500
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 2.8647186147186148e-05,
+      "loss": 1.7398,
+      "step": 1000
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 2.797077922077922e-05,
+      "loss": 1.5769,
+      "step": 1500
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 2.7294372294372295e-05,
+      "loss": 1.4762,
+      "step": 2000
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 2.661796536796537e-05,
+      "loss": 1.458,
+      "step": 2500
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 2.594155844155844e-05,
+      "loss": 1.3717,
+      "step": 3000
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 2.5265151515151516e-05,
+      "loss": 1.3406,
+      "step": 3500
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 2.458874458874459e-05,
+      "loss": 1.2729,
+      "step": 4000
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 2.3912337662337663e-05,
+      "loss": 1.278,
+      "step": 4500
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 2.3235930735930737e-05,
+      "loss": 1.2732,
+      "step": 5000
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 2.255952380952381e-05,
+      "loss": 1.2511,
+      "step": 5500
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 2.1883116883116884e-05,
+      "loss": 1.214,
+      "step": 6000
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 2.1206709956709957e-05,
+      "loss": 1.2218,
+      "step": 6500
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 2.053030303030303e-05,
+      "loss": 1.1724,
+      "step": 7000
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 1.9853896103896105e-05,
+      "loss": 1.1945,
+      "step": 7500
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 1.9177489177489178e-05,
+      "loss": 1.1869,
+      "step": 8000
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.8501082251082252e-05,
+      "loss": 1.1306,
+      "step": 8500
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.7824675324675322e-05,
+      "loss": 1.1042,
+      "step": 9000
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 1.71482683982684e-05,
+      "loss": 1.1363,
+      "step": 9500
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.6471861471861473e-05,
+      "loss": 1.0794,
+      "step": 10000
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.5795454545454546e-05,
+      "loss": 1.0797,
+      "step": 10500
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 1.511904761904762e-05,
+      "loss": 1.089,
+      "step": 11000
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 1.4442640692640693e-05,
+      "loss": 0.845,
+      "step": 11500
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 1.3766233766233767e-05,
+      "loss": 0.801,
+      "step": 12000
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.308982683982684e-05,
+      "loss": 0.8163,
+      "step": 12500
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.2413419913419914e-05,
+      "loss": 0.7864,
+      "step": 13000
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.1737012987012986e-05,
+      "loss": 0.8054,
+      "step": 13500
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.106060606060606e-05,
+      "loss": 0.7873,
+      "step": 14000
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.0384199134199135e-05,
+      "loss": 0.8018,
+      "step": 14500
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 9.707792207792209e-06,
+      "loss": 0.7835,
+      "step": 15000
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 9.031385281385282e-06,
+      "loss": 0.7646,
+      "step": 15500
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 8.354978354978356e-06,
+      "loss": 0.7452,
+      "step": 16000
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 7.678571428571428e-06,
+      "loss": 0.8174,
+      "step": 16500
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 7.002164502164502e-06,
+      "loss": 0.7848,
+      "step": 17000
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 6.325757575757576e-06,
+      "loss": 0.7889,
+      "step": 17500
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 5.64935064935065e-06,
+      "loss": 0.7721,
+      "step": 18000
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 4.972943722943723e-06,
+      "loss": 0.7488,
+      "step": 18500
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 4.2965367965367965e-06,
+      "loss": 0.7553,
+      "step": 19000
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 3.62012987012987e-06,
+      "loss": 0.7433,
+      "step": 19500
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 2.943722943722944e-06,
+      "loss": 0.7438,
+      "step": 20000
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 2.2673160173160173e-06,
+      "loss": 0.739,
+      "step": 20500
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 1.590909090909091e-06,
+      "loss": 0.7234,
+      "step": 21000
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 9.145021645021645e-07,
+      "loss": 0.7282,
+      "step": 21500
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 2.380952380952381e-07,
+      "loss": 0.7243,
+      "step": 22000
+    },
+    {
+      "epoch": 2.0,
+      "step": 22176,
+      "total_flos": 5.214986800612762e+16,
+      "train_loss": 1.043660425004505,
+      "train_runtime": 7588.3192,
+      "train_samples_per_second": 35.068,
+      "train_steps_per_second": 2.922
+    }
+  ],
+  "max_steps": 22176,
+  "num_train_epochs": 2,
+  "total_flos": 5.214986800612762e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1f49ca01d9678500a5173fe917cdce2c3cd6b6433f04e3bca82a5772cffca97
+size 3439