best model for qa

config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "/projects/SolarWindowsADSP/shu/models/bert-base-cased/",
+  "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",
+  "transformers_version": "4.17.0.dev0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 28996
+}

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73b89345bde715153dd3aabb21ac41112de25fe10796a29357f84c61adf1b357
+size 861894637

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28d889c4a1e2018c2400fbb4facef2470becbfa230163aa2f03eb4a337f01853
+size 430973431

rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:65e52c4dc9f0b6027861d9f82533bfa75552e49459a52af747f965b9a3e173d1
+size 14657

rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f0590c7cc50662a1d524c6e2e05d15f088f80aa6f638eebf6d05611e53f7241
+size 14657

rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72f978760997b87af8e24f87ce24f24353a6454ad7b2b368ef601d3adb13ddcf
+size 14657

rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1fdd13f7e6c0bdc70fe28b406d558e13924f29022fe182c6dc451c3227e60bbb
+size 14657

rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4589a3500937edd1bf434bd7c9059e931792c96922065277779a10a295ec5b8
+size 14657

rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5146c979f432c7fdcff56f197d333de5fea21e202dd9f47d1d92d84f5ed944c
+size 14657

rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8659629f68899fff8ea73907f316f7c58a7426b3593eac457f160208d9e64a23
+size 14657

rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:78fd75e0aa8e02bf3cbef338768a2e39cc1feb146e4d9497b1d76a5294789e40
+size 14657

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f069092755d43f4443a85925a4bd11e00e6f0af33e6dbf8088b6343672afce0
+size 623

special_tokens_map.json

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

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": false, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "special_tokens_map_file": null, "name_or_path": "/projects/SolarWindowsADSP/shu/models/bert-base-cased/", "tokenizer_class": "BertTokenizer"}

trainer_state.json

@@ -0,0 +1,94 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "global_step": 5546,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.774612333213127e-05,
+      "loss": 1.7894,
+      "step": 500
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.549224666426253e-05,
+      "loss": 1.2681,
+      "step": 1000
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.32383699963938e-05,
+      "loss": 1.1511,
+      "step": 1500
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 4.0984493328525064e-05,
+      "loss": 1.1003,
+      "step": 2000
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 3.873061666065633e-05,
+      "loss": 1.0589,
+      "step": 2500
+    },
+    {
+      "epoch": 1.0,
+      "eval_exact_match": 79.72563859981078,
+      "eval_f1": 87.33897800522966,
+      "step": 2773
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 3.6476739992787594e-05,
+      "loss": 0.8936,
+      "step": 3000
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 3.422286332491886e-05,
+      "loss": 0.7398,
+      "step": 3500
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 3.1968986657050125e-05,
+      "loss": 0.738,
+      "step": 4000
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 2.9715109989181394e-05,
+      "loss": 0.7341,
+      "step": 4500
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 2.7461233321312656e-05,
+      "loss": 0.7264,
+      "step": 5000
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 2.5207356653443925e-05,
+      "loss": 0.728,
+      "step": 5500
+    },
+    {
+      "epoch": 2.0,
+      "eval_exact_match": 81.2961210974456,
+      "eval_f1": 88.57827007260809,
+      "step": 5546
+    }
+  ],
+  "max_steps": 11092,
+  "num_train_epochs": 4,
+  "total_flos": 3.4779642825015296e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fce5da6709f1e587806932d8b6a58f9aba97649f9be714b914efbf3c6530007a
+size 3119