adding model fine-tuned on QA (SQAC)

all_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 3.0,
+    "eval_exact_match": 61.69527896995708,
+    "eval_f1": 80.21953385689976,
+    "eval_samples": 2082,
+    "train_loss": 1.1593981150564048,
+    "train_runtime": 2287.6753,
+    "train_samples": 16598,
+    "train_samples_per_second": 21.766,
+    "train_steps_per_second": 1.361
+}

config.json

@@ -0,0 +1,33 @@
+{
+  "_name_or_path": "CenIA/albert_xlarge_spanish",
+  "architectures": [
+    "AlbertForQuestionAnswering"
+  ],
+  "attention_probs_dropout_prob": 0,
+  "bos_token_id": 2,
+  "classifier_dropout_prob": 0.1,
+  "down_scale_factor": 1,
+  "embedding_size": 128,
+  "eos_token_id": 3,
+  "gap_size": 0,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0,
+  "hidden_size": 2048,
+  "initializer_range": 0.01,
+  "inner_group_num": 1,
+  "intermediate_size": 8192,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "albert",
+  "net_structure_type": 0,
+  "num_attention_heads": 32,
+  "num_hidden_groups": 1,
+  "num_hidden_layers": 24,
+  "num_memory_blocks": 0,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.15.0",
+  "type_vocab_size": 2,
+  "vocab_size": 31000
+}

eval_results.json

@@ -0,0 +1,6 @@
+{
+    "epoch": 3.0,
+    "eval_exact_match": 61.69527896995708,
+    "eval_f1": 80.21953385689976,
+    "eval_samples": 2082
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:58c32033af89efd9722453357a28725052419c4289fa536de149a3147ed67a8b
+size 218656663

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "[CLS]", "eos_token": "[SEP]", "unk_token": "<unk>", "sep_token": "[SEP]", "pad_token": "<pad>", "cls_token": "[CLS]", "mask_token": "[MASK]"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"do_lower_case": true, "remove_space": true, "keep_accents": true, "bos_token": "[CLS]", "eos_token": "[SEP]", "unk_token": "<unk>", "sep_token": "[SEP]", "pad_token": "<pad>", "cls_token": "[CLS]", "mask_token": {"content": "[MASK]", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "CenIA/albert_xlarge_spanish", "tokenizer_class": "AlbertTokenizer"}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "train_loss": 1.1593981150564048,
+    "train_runtime": 2287.6753,
+    "train_samples": 16598,
+    "train_samples_per_second": 21.766,
+    "train_steps_per_second": 1.361
+}

trainer_state.json

@@ -0,0 +1,151 @@
+{
+  "best_metric": 80.21953385689976,
+  "best_model_checkpoint": "/home/fbravo/data/all_results/sqac/albeto_xlarge/epochs_3_bs_16_lr_5e-6/checkpoint-2600",
+  "epoch": 3.0,
+  "global_step": 3114,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.19,
+      "eval_exact_match": 36.427038626609445,
+      "eval_f1": 58.55860208350086,
+      "step": 200
+    },
+    {
+      "epoch": 0.39,
+      "eval_exact_match": 49.73175965665236,
+      "eval_f1": 70.53361014532948,
+      "step": 400
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.210019267822737e-06,
+      "loss": 2.5712,
+      "step": 500
+    },
+    {
+      "epoch": 0.58,
+      "eval_exact_match": 52.73605150214592,
+      "eval_f1": 73.46972587500662,
+      "step": 600
+    },
+    {
+      "epoch": 0.77,
+      "eval_exact_match": 55.79399141630901,
+      "eval_f1": 75.66446519793949,
+      "step": 800
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 3.4071933204881185e-06,
+      "loss": 1.2782,
+      "step": 1000
+    },
+    {
+      "epoch": 0.96,
+      "eval_exact_match": 57.081545064377686,
+      "eval_f1": 76.89960784606122,
+      "step": 1000
+    },
+    {
+      "epoch": 1.16,
+      "eval_exact_match": 59.17381974248927,
+      "eval_f1": 78.01947007379184,
+      "step": 1200
+    },
+    {
+      "epoch": 1.35,
+      "eval_exact_match": 58.90557939914163,
+      "eval_f1": 78.1836005785906,
+      "step": 1400
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 2.6043673731535007e-06,
+      "loss": 0.9553,
+      "step": 1500
+    },
+    {
+      "epoch": 1.54,
+      "eval_exact_match": 60.08583690987125,
+      "eval_f1": 78.6550065433202,
+      "step": 1600
+    },
+    {
+      "epoch": 1.73,
+      "eval_exact_match": 60.407725321888414,
+      "eval_f1": 79.3455012767423,
+      "step": 1800
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.8031470777135517e-06,
+      "loss": 0.8937,
+      "step": 2000
+    },
+    {
+      "epoch": 1.93,
+      "eval_exact_match": 60.35407725321888,
+      "eval_f1": 78.98705198670261,
+      "step": 2000
+    },
+    {
+      "epoch": 2.12,
+      "eval_exact_match": 60.67596566523605,
+      "eval_f1": 79.36928943419082,
+      "step": 2200
+    },
+    {
+      "epoch": 2.31,
+      "eval_exact_match": 60.94420600858369,
+      "eval_f1": 79.48649391462303,
+      "step": 2400
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 1.000321130378934e-06,
+      "loss": 0.6922,
+      "step": 2500
+    },
+    {
+      "epoch": 2.5,
+      "eval_exact_match": 61.69527896995708,
+      "eval_f1": 80.21953385689976,
+      "step": 2600
+    },
+    {
+      "epoch": 2.7,
+      "eval_exact_match": 61.31974248927039,
+      "eval_f1": 79.68483897616439,
+      "step": 2800
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 1.97495183044316e-07,
+      "loss": 0.67,
+      "step": 3000
+    },
+    {
+      "epoch": 2.89,
+      "eval_exact_match": 61.37339055793991,
+      "eval_f1": 79.75248785860451,
+      "step": 3000
+    },
+    {
+      "epoch": 3.0,
+      "step": 3114,
+      "total_flos": 7572958746771456.0,
+      "train_loss": 1.1593981150564048,
+      "train_runtime": 2287.6753,
+      "train_samples_per_second": 21.766,
+      "train_steps_per_second": 1.361
+    }
+  ],
+  "max_steps": 3114,
+  "num_train_epochs": 3,
+  "total_flos": 7572958746771456.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8473d93c39f92151ce1b5c245bd10e2179cf69a65c1256da4ddfc5f17c75c0e7
+size 2991