Best checkpoint from 5-epoch training run on relevance-filtered ctrs

.gitignore

@@ -0,0 +1 @@
+checkpoint-*/

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 5.0,
+    "eval_accuracy": 0.615,
+    "eval_f1": 0.641860465116279,
+    "eval_loss": 0.6473702788352966,
+    "eval_precision": 0.6,
+    "eval_recall": 0.69,
+    "eval_runtime": 0.6911,
+    "eval_samples_per_second": 289.404,
+    "eval_steps_per_second": 10.129,
+    "train_loss": 0.623549058702257,
+    "train_runtime": 216.6633,
+    "train_samples": 1700,
+    "train_samples_per_second": 39.231,
+    "train_steps_per_second": 1.246
+}

config.json

@@ -0,0 +1,35 @@
+{
+  "_name_or_path": "/home/taw2/semeval/gatortron_da_breast_cancer",
+  "architectures": [
+    "MegatronBertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "Contradiction",
+    "1": "Entailment"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "Contradiction": 0,
+    "Entailment": 1
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "megatron-bert",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "num_layers": 24,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "tokenizer_type": "BertWordPieceCase",
+  "torch_dtype": "float32",
+  "transformers_version": "4.24.0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 50176
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 5.0,
+    "eval_accuracy": 0.615,
+    "eval_f1": 0.641860465116279,
+    "eval_loss": 0.6473702788352966,
+    "eval_precision": 0.6,
+    "eval_recall": 0.69,
+    "eval_runtime": 0.6911,
+    "eval_samples_per_second": 289.404,
+    "eval_steps_per_second": 10.129
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d00ec978614fed3b45388384185e2f9cc06ddff55abadc51d7be53610808f987
+size 1421216053

runs/Mar23_09-50-19_node0293.palmetto.clemson.edu/1679579473.3417587/events.out.tfevents.1679579473.node0293.palmetto.clemson.edu

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:278a419e17a4ed0d02746de2f052e141789db0b249790d75ae0563ae52b9a181
+size 5332

runs/Mar23_09-50-19_node0293.palmetto.clemson.edu/events.out.tfevents.1679579472.node0293.palmetto.clemson.edu

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9730846330b38012e686e416882cd10cef0ebc74a01cdd51c269bc923d0b178d
+size 6703

runs/Mar23_09-50-19_node0293.palmetto.clemson.edu/events.out.tfevents.1679579706.node0293.palmetto.clemson.edu

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f9dcb2414f1ca55d6d0551b7a27c17b4a369ad3a0170459e15cae8f8bf5b5ff
+size 512

runs/Mar23_10-09-36_node0382.palmetto.clemson.edu/1679580609.5511744/events.out.tfevents.1679580609.node0382.palmetto.clemson.edu

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87ff36cffbfce4cc8f331e9dedd7d9f18bba1030899a5f5af6823b73bb2e25aa
+size 5332

runs/Mar23_10-09-36_node0382.palmetto.clemson.edu/events.out.tfevents.1679580609.node0382.palmetto.clemson.edu

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7504d11e1955a3edabdd8ab1d99b9e593856eeab0e843dffeb30b632859aaef
+size 6703

runs/Mar23_10-09-36_node0382.palmetto.clemson.edu/events.out.tfevents.1679580827.node0382.palmetto.clemson.edu

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46231865a11f1166bfadf7019c0c7c55f1adb2c6f4534ead713a30e7eb063375
+size 40

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 +1,18 @@
+{
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": false,
+  "mask_token": "[MASK]",
+  "max_length": 512,
+  "name_or_path": "/home/taw2/gatortron",
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "special_tokens_map_file": "/home/taw2/gatortron/special_tokens_map.json",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "truncation": true,
+  "unk_token": "[UNK]",
+  "use_fast": true
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 5.0,
+    "train_loss": 0.623549058702257,
+    "train_runtime": 216.6633,
+    "train_samples": 1700,
+    "train_samples_per_second": 39.231,
+    "train_steps_per_second": 1.246
+}

trainer_state.json

@@ -0,0 +1,85 @@
+{
+  "best_metric": 0.6473702788352966,
+  "best_model_checkpoint": "/scratch1/taw2/semeval/nli_upload/checkpoint-162",
+  "epoch": 5.0,
+  "global_step": 270,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5,
+      "eval_f1": 0.0,
+      "eval_loss": 0.6974926590919495,
+      "eval_precision": 0.0,
+      "eval_recall": 0.0,
+      "eval_runtime": 0.739,
+      "eval_samples_per_second": 270.621,
+      "eval_steps_per_second": 9.472,
+      "step": 54
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.59,
+      "eval_f1": 0.5060240963855421,
+      "eval_loss": 0.6564074754714966,
+      "eval_precision": 0.6363636363636364,
+      "eval_recall": 0.42,
+      "eval_runtime": 0.6921,
+      "eval_samples_per_second": 288.987,
+      "eval_steps_per_second": 10.115,
+      "step": 108
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.615,
+      "eval_f1": 0.641860465116279,
+      "eval_loss": 0.6473702788352966,
+      "eval_precision": 0.6,
+      "eval_recall": 0.69,
+      "eval_runtime": 0.6899,
+      "eval_samples_per_second": 289.893,
+      "eval_steps_per_second": 10.146,
+      "step": 162
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6,
+      "eval_f1": 0.6638655462184875,
+      "eval_loss": 0.6594653129577637,
+      "eval_precision": 0.572463768115942,
+      "eval_recall": 0.79,
+      "eval_runtime": 0.6906,
+      "eval_samples_per_second": 289.595,
+      "eval_steps_per_second": 10.136,
+      "step": 216
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.61,
+      "eval_f1": 0.6355140186915889,
+      "eval_loss": 0.7758804559707642,
+      "eval_precision": 0.5964912280701754,
+      "eval_recall": 0.68,
+      "eval_runtime": 0.6911,
+      "eval_samples_per_second": 289.401,
+      "eval_steps_per_second": 10.129,
+      "step": 270
+    },
+    {
+      "epoch": 5.0,
+      "step": 270,
+      "total_flos": 837689804211456.0,
+      "train_loss": 0.623549058702257,
+      "train_runtime": 216.6633,
+      "train_samples_per_second": 39.231,
+      "train_steps_per_second": 1.246
+    }
+  ],
+  "max_steps": 270,
+  "num_train_epochs": 5,
+  "total_flos": 837689804211456.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:498fd7eab865b11df116117d9f78fbea94b9ef978de6a88f2e17ddfab7c0edb5
+size 3515