adding model fine-tuned on QA (TAR)

.gitattributes

@@ -26,3 +26,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zstandard 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

all_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 2.0,
+    "eval_exact_match": 61.00283822138127,
+    "eval_f1": 77.48501965183796,
+    "eval_samples": 10619,
+    "train_loss": 1.41421453160958,
+    "train_runtime": 7497.7066,
+    "train_samples": 87749,
+    "train_samples_per_second": 23.407,
+    "train_steps_per_second": 0.732
+}

config.json

@@ -0,0 +1,27 @@
+{
+  "_name_or_path": "dccuchile/bert-base-spanish-wwm-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,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.15.0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 31002
+}

eval_nbest_predictions.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf10070a531d4c2cde929e0f33a21eb042eac7879e05115cc1d34aa60ed2c5f2
+size 44865449

eval_results.json

@@ -0,0 +1,6 @@
+{
+    "epoch": 2.0,
+    "eval_exact_match": 61.00283822138127,
+    "eval_f1": 77.48501965183796,
+    "eval_samples": 10619
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:911690f82971c18f6a03e0db1bbae422b8cb89d4b427771148cce95b2914cb32
+size 437130673

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": false, "name_or_path": "dccuchile/bert-base-spanish-wwm-cased", "do_basic_tokenize": true, "never_split": null, "model_max_length": 512, "special_tokens_map_file": "/data/jcanete/cache/9848a00af462c42dfb4ec88ef438fbab5256330f7f6f50badc48d277f9367d49.f982506b52498d4adb4bd491f593dc92b2ef6be61bfdbe9d30f53f963f9f5b66", "tokenizer_class": "BertTokenizer"}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.0,
+    "train_loss": 1.41421453160958,
+    "train_runtime": 7497.7066,
+    "train_samples": 87749,
+    "train_samples_per_second": 23.407,
+    "train_steps_per_second": 0.732
+}

trainer_state.json

@@ -0,0 +1,193 @@
+{
+  "best_metric": 77.48501965183796,
+  "best_model_checkpoint": "/data/jcanete/all_results/tar/beto_cased/epochs_2_bs_32_lr_5e-5/checkpoint-5100",
+  "epoch": 2.0,
+  "global_step": 5486,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.11,
+      "eval_exact_match": 50.879848628193,
+      "eval_f1": 69.30387024901542,
+      "step": 300
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.546117389719285e-05,
+      "loss": 2.1254,
+      "step": 500
+    },
+    {
+      "epoch": 0.22,
+      "eval_exact_match": 54.512771996215704,
+      "eval_f1": 71.89755492209031,
+      "step": 600
+    },
+    {
+      "epoch": 0.33,
+      "eval_exact_match": 55.96972563859981,
+      "eval_f1": 73.46469051491528,
+      "step": 900
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.090411957710536e-05,
+      "loss": 1.7294,
+      "step": 1000
+    },
+    {
+      "epoch": 0.44,
+      "eval_exact_match": 56.60359508041627,
+      "eval_f1": 73.90716618298387,
+      "step": 1200
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 3.6347065257017864e-05,
+      "loss": 1.6464,
+      "step": 1500
+    },
+    {
+      "epoch": 0.55,
+      "eval_exact_match": 56.773888363292336,
+      "eval_f1": 74.46721951870393,
+      "step": 1500
+    },
+    {
+      "epoch": 0.66,
+      "eval_exact_match": 57.95648060548723,
+      "eval_f1": 75.61075089328189,
+      "step": 1800
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 3.179001093693037e-05,
+      "loss": 1.5901,
+      "step": 2000
+    },
+    {
+      "epoch": 0.77,
+      "eval_exact_match": 58.240302743614,
+      "eval_f1": 75.51663308469975,
+      "step": 2100
+    },
+    {
+      "epoch": 0.87,
+      "eval_exact_match": 59.55534531693472,
+      "eval_f1": 76.61330921600383,
+      "step": 2400
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 2.7232956616842875e-05,
+      "loss": 1.5281,
+      "step": 2500
+    },
+    {
+      "epoch": 0.98,
+      "eval_exact_match": 59.94323557237465,
+      "eval_f1": 76.49598618587426,
+      "step": 2700
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 2.2675902296755377e-05,
+      "loss": 1.3384,
+      "step": 3000
+    },
+    {
+      "epoch": 1.09,
+      "eval_exact_match": 58.99716177861873,
+      "eval_f1": 76.52241925728033,
+      "step": 3000
+    },
+    {
+      "epoch": 1.2,
+      "eval_exact_match": 59.38505203405866,
+      "eval_f1": 76.35485242703709,
+      "step": 3300
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.8118847976667883e-05,
+      "loss": 1.1384,
+      "step": 3500
+    },
+    {
+      "epoch": 1.31,
+      "eval_exact_match": 59.66887417218543,
+      "eval_f1": 76.86602160252082,
+      "step": 3600
+    },
+    {
+      "epoch": 1.42,
+      "eval_exact_match": 60.3689687795648,
+      "eval_f1": 76.98844667865373,
+      "step": 3900
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.3561793656580388e-05,
+      "loss": 1.1451,
+      "step": 4000
+    },
+    {
+      "epoch": 1.53,
+      "eval_exact_match": 60.3689687795648,
+      "eval_f1": 77.01614995652554,
+      "step": 4200
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.013853445133066e-06,
+      "loss": 1.1248,
+      "step": 4500
+    },
+    {
+      "epoch": 1.64,
+      "eval_exact_match": 60.53926206244087,
+      "eval_f1": 77.22917143472323,
+      "step": 4500
+    },
+    {
+      "epoch": 1.75,
+      "eval_exact_match": 60.68117313150426,
+      "eval_f1": 77.3287559059371,
+      "step": 4800
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 4.456799125045571e-06,
+      "loss": 1.0917,
+      "step": 5000
+    },
+    {
+      "epoch": 1.86,
+      "eval_exact_match": 61.00283822138127,
+      "eval_f1": 77.48501965183796,
+      "step": 5100
+    },
+    {
+      "epoch": 1.97,
+      "eval_exact_match": 61.031220435193944,
+      "eval_f1": 77.37778335318785,
+      "step": 5400
+    },
+    {
+      "epoch": 2.0,
+      "step": 5486,
+      "total_flos": 3.4090493726506464e+16,
+      "train_loss": 1.41421453160958,
+      "train_runtime": 7497.7066,
+      "train_samples_per_second": 23.407,
+      "train_steps_per_second": 0.732
+    }
+  ],
+  "max_steps": 5486,
+  "num_train_epochs": 2,
+  "total_flos": 3.4090493726506464e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86ba0f8705a21fd24c9901e2e1cca3874d6c88fd80a69dd82c26f1fb8f539485
+size 2991