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": 61.84484389782403,
+    "eval_f1": 78.98580203731828,
+    "eval_samples": 10624,
+    "train_loss": 1.2922884611618335,
+    "train_runtime": 6069.4707,
+    "train_samples": 87747,
+    "train_samples_per_second": 28.914,
+    "train_steps_per_second": 1.807
+}

config.json

@@ -0,0 +1,34 @@
+{
+  "_name_or_path": "CenIA/albert_xxlarge_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": 4096,
+  "initializer_range": 0.01,
+  "inner_group_num": 1,
+  "intermediate_size": 16384,
+  "layer_norm_eps": 1e-12,
+  "layers_to_keep": [],
+  "max_position_embeddings": 512,
+  "model_type": "albert",
+  "net_structure_type": 0,
+  "num_attention_heads": 64,
+  "num_hidden_groups": 1,
+  "num_hidden_layers": 12,
+  "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_nbest_predictions.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:acb86ff9e33141875ad3b6c069623c672a15d347deb50125e06e69c9236dad39
+size 44451503

eval_results.json

@@ -0,0 +1,6 @@
+{
+    "epoch": 2.0,
+    "eval_exact_match": 61.84484389782403,
+    "eval_f1": 78.98580203731828,
+    "eval_samples": 10624
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc41d7196a29fb44ce60c9c995653c0d5b7313f8b332e73008fd21586621bb3a
+size 823816087

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_xxlarge_spanish", "tokenizer_class": "AlbertTokenizer"}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.0,
+    "train_loss": 1.2922884611618335,
+    "train_runtime": 6069.4707,
+    "train_samples": 87747,
+    "train_samples_per_second": 28.914,
+    "train_steps_per_second": 1.807
+}

trainer_state.json

@@ -0,0 +1,367 @@
+{
+  "best_metric": 78.98580203731828,
+  "best_model_checkpoint": "/home/fbravo/data/all_results/tar/albeto_xxlarge/epochs_2_bs_16_lr_5e-6/checkpoint-7800",
+  "epoch": 2.0,
+  "global_step": 10970,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.05,
+      "eval_exact_match": 50.73793755912961,
+      "eval_f1": 69.4730775689932,
+      "step": 300
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.774384685505926e-06,
+      "loss": 2.2004,
+      "step": 500
+    },
+    {
+      "epoch": 0.11,
+      "eval_exact_match": 54.370860927152314,
+      "eval_f1": 73.05853038044215,
+      "step": 600
+    },
+    {
+      "epoch": 0.16,
+      "eval_exact_match": 54.89120151371807,
+      "eval_f1": 73.65027892880356,
+      "step": 900
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.546490428441203e-06,
+      "loss": 1.6573,
+      "step": 1000
+    },
+    {
+      "epoch": 0.22,
+      "eval_exact_match": 56.9914853358562,
+      "eval_f1": 75.0513540652836,
+      "step": 1200
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.319051959890611e-06,
+      "loss": 1.5668,
+      "step": 1500
+    },
+    {
+      "epoch": 0.27,
+      "eval_exact_match": 57.8240302743614,
+      "eval_f1": 75.63669001183044,
+      "step": 1500
+    },
+    {
+      "epoch": 0.33,
+      "eval_exact_match": 58.81740775780511,
+      "eval_f1": 76.36054723510281,
+      "step": 1800
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.091157702825889e-06,
+      "loss": 1.5322,
+      "step": 2000
+    },
+    {
+      "epoch": 0.38,
+      "eval_exact_match": 59.14853358561968,
+      "eval_f1": 76.483161808874,
+      "step": 2100
+    },
+    {
+      "epoch": 0.44,
+      "eval_exact_match": 59.25260170293283,
+      "eval_f1": 76.67791535670501,
+      "step": 2400
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 3.863263445761167e-06,
+      "loss": 1.5015,
+      "step": 2500
+    },
+    {
+      "epoch": 0.49,
+      "eval_exact_match": 59.290444654683064,
+      "eval_f1": 76.94674808406094,
+      "step": 2700
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 3.6353691886964453e-06,
+      "loss": 1.4561,
+      "step": 3000
+    },
+    {
+      "epoch": 0.55,
+      "eval_exact_match": 60.13245033112583,
+      "eval_f1": 77.30126888604488,
+      "step": 3000
+    },
+    {
+      "epoch": 0.6,
+      "eval_exact_match": 59.93377483443709,
+      "eval_f1": 77.31924407449924,
+      "step": 3300
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 3.407474931631723e-06,
+      "loss": 1.4414,
+      "step": 3500
+    },
+    {
+      "epoch": 0.66,
+      "eval_exact_match": 60.44465468306528,
+      "eval_f1": 77.63204103944324,
+      "step": 3600
+    },
+    {
+      "epoch": 0.71,
+      "eval_exact_match": 61.31504257332072,
+      "eval_f1": 78.13561765566448,
+      "step": 3900
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 3.179580674567001e-06,
+      "loss": 1.4206,
+      "step": 4000
+    },
+    {
+      "epoch": 0.77,
+      "eval_exact_match": 60.52034058656575,
+      "eval_f1": 77.94067744854881,
+      "step": 4200
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 2.9516864175022793e-06,
+      "loss": 1.3835,
+      "step": 4500
+    },
+    {
+      "epoch": 0.82,
+      "eval_exact_match": 60.98391674550615,
+      "eval_f1": 78.30309425310817,
+      "step": 4500
+    },
+    {
+      "epoch": 0.88,
+      "eval_exact_match": 61.21097445600757,
+      "eval_f1": 78.3458557935229,
+      "step": 4800
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 2.7237921604375573e-06,
+      "loss": 1.3656,
+      "step": 5000
+    },
+    {
+      "epoch": 0.93,
+      "eval_exact_match": 61.30558183538316,
+      "eval_f1": 78.35466990254258,
+      "step": 5100
+    },
+    {
+      "epoch": 0.98,
+      "eval_exact_match": 61.485335856196784,
+      "eval_f1": 78.61331349695031,
+      "step": 5400
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.4958979033728353e-06,
+      "loss": 1.3461,
+      "step": 5500
+    },
+    {
+      "epoch": 1.04,
+      "eval_exact_match": 61.258278145695364,
+      "eval_f1": 78.44203108653505,
+      "step": 5700
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 2.268459434822243e-06,
+      "loss": 1.0985,
+      "step": 6000
+    },
+    {
+      "epoch": 1.09,
+      "eval_exact_match": 61.22989593188269,
+      "eval_f1": 78.6338704184745,
+      "step": 6000
+    },
+    {
+      "epoch": 1.15,
+      "eval_exact_match": 61.21097445600757,
+      "eval_f1": 78.69664962146094,
+      "step": 6300
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 2.0405651777575204e-06,
+      "loss": 1.0548,
+      "step": 6500
+    },
+    {
+      "epoch": 1.2,
+      "eval_exact_match": 61.343424787133394,
+      "eval_f1": 78.77498098426705,
+      "step": 6600
+    },
+    {
+      "epoch": 1.26,
+      "eval_exact_match": 61.65562913907285,
+      "eval_f1": 78.67261809065106,
+      "step": 6900
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.8126709206927986e-06,
+      "loss": 1.0615,
+      "step": 7000
+    },
+    {
+      "epoch": 1.31,
+      "eval_exact_match": 61.53263954588458,
+      "eval_f1": 78.5005776920404,
+      "step": 7200
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.5847766636280768e-06,
+      "loss": 1.0828,
+      "step": 7500
+    },
+    {
+      "epoch": 1.37,
+      "eval_exact_match": 61.551561021759696,
+      "eval_f1": 78.55843313017921,
+      "step": 7500
+    },
+    {
+      "epoch": 1.42,
+      "eval_exact_match": 61.84484389782403,
+      "eval_f1": 78.98580203731828,
+      "step": 7800
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.3573381950774841e-06,
+      "loss": 1.0492,
+      "step": 8000
+    },
+    {
+      "epoch": 1.48,
+      "eval_exact_match": 61.73131504257332,
+      "eval_f1": 78.56521589229312,
+      "step": 8100
+    },
+    {
+      "epoch": 1.53,
+      "eval_exact_match": 61.81646168401135,
+      "eval_f1": 78.62740589986592,
+      "step": 8400
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 1.1294439380127621e-06,
+      "loss": 1.0551,
+      "step": 8500
+    },
+    {
+      "epoch": 1.59,
+      "eval_exact_match": 61.57994323557237,
+      "eval_f1": 78.64237467540597,
+      "step": 8700
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 9.015496809480402e-07,
+      "loss": 1.0353,
+      "step": 9000
+    },
+    {
+      "epoch": 1.64,
+      "eval_exact_match": 61.30558183538316,
+      "eval_f1": 78.50591438549111,
+      "step": 9000
+    },
+    {
+      "epoch": 1.7,
+      "eval_exact_match": 61.63670766319773,
+      "eval_f1": 78.6172319874888,
+      "step": 9300
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 6.736554238833181e-07,
+      "loss": 1.0373,
+      "step": 9500
+    },
+    {
+      "epoch": 1.75,
+      "eval_exact_match": 61.66508987701041,
+      "eval_f1": 78.81109427991775,
+      "step": 9600
+    },
+    {
+      "epoch": 1.8,
+      "eval_exact_match": 61.59886471144749,
+      "eval_f1": 78.57587678017954,
+      "step": 9900
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 4.457611668185962e-07,
+      "loss": 1.0469,
+      "step": 10000
+    },
+    {
+      "epoch": 1.86,
+      "eval_exact_match": 61.81646168401135,
+      "eval_f1": 78.77340428037925,
+      "step": 10200
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 2.1832269826800366e-07,
+      "loss": 1.0179,
+      "step": 10500
+    },
+    {
+      "epoch": 1.91,
+      "eval_exact_match": 61.84484389782403,
+      "eval_f1": 78.74115858178497,
+      "step": 10500
+    },
+    {
+      "epoch": 1.97,
+      "eval_exact_match": 61.74077578051088,
+      "eval_f1": 78.74584305622463,
+      "step": 10800
+    },
+    {
+      "epoch": 2.0,
+      "step": 10970,
+      "total_flos": 6.330675040878592e+16,
+      "train_loss": 1.2922884611618335,
+      "train_runtime": 6069.4707,
+      "train_samples_per_second": 28.914,
+      "train_steps_per_second": 1.807
+    }
+  ],
+  "max_steps": 10970,
+  "num_train_epochs": 2,
+  "total_flos": 6.330675040878592e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d6a51dc6ce854bf59380934937c244c0b6afbacb72192e862911df25c3d3f53
+size 2991