adding model finetuned on POS

all_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 4.0,
+    "eval_accuracy": 0.9747039189441047,
+    "eval_f1": 0.9728139460666986,
+    "eval_loss": 0.08749764412641525,
+    "eval_precision": 0.9711674399155732,
+    "eval_recall": 0.9744660446344728,
+    "eval_runtime": 1.5052,
+    "eval_samples": 1654,
+    "eval_samples_per_second": 1098.84,
+    "eval_steps_per_second": 69.093,
+    "train_loss": 0.09274842659188383,
+    "train_runtime": 230.8308,
+    "train_samples": 14305,
+    "train_samples_per_second": 247.887,
+    "train_steps_per_second": 15.509
+}

config.json

@@ -0,0 +1,74 @@
+{
+  "_name_or_path": "CenIA/albert_tiny_spanish",
+  "architectures": [
+    "AlbertForTokenClassification"
+  ],
+  "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,
+  "finetuning_task": "pos",
+  "gap_size": 0,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0,
+  "hidden_size": 312,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4",
+    "5": "LABEL_5",
+    "6": "LABEL_6",
+    "7": "LABEL_7",
+    "8": "LABEL_8",
+    "9": "LABEL_9",
+    "10": "LABEL_10",
+    "11": "LABEL_11",
+    "12": "LABEL_12",
+    "13": "LABEL_13",
+    "14": "LABEL_14",
+    "15": "LABEL_15",
+    "16": "LABEL_16",
+    "17": "LABEL_17"
+  },
+  "initializer_range": 0.02,
+  "inner_group_num": 1,
+  "intermediate_size": 1248,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_10": 10,
+    "LABEL_11": 11,
+    "LABEL_12": 12,
+    "LABEL_13": 13,
+    "LABEL_14": 14,
+    "LABEL_15": 15,
+    "LABEL_16": 16,
+    "LABEL_17": 17,
+    "LABEL_2": 2,
+    "LABEL_3": 3,
+    "LABEL_4": 4,
+    "LABEL_5": 5,
+    "LABEL_6": 6,
+    "LABEL_7": 7,
+    "LABEL_8": 8,
+    "LABEL_9": 9
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "albert",
+  "net_structure_type": 0,
+  "num_attention_heads": 12,
+  "num_hidden_groups": 1,
+  "num_hidden_layers": 4,
+  "num_memory_blocks": 0,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.12.5",
+  "type_vocab_size": 2,
+  "vocab_size": 31000
+}

eval_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 4.0,
+    "eval_accuracy": 0.9747039189441047,
+    "eval_f1": 0.9728139460666986,
+    "eval_loss": 0.08749764412641525,
+    "eval_precision": 0.9711674399155732,
+    "eval_recall": 0.9744660446344728,
+    "eval_runtime": 1.5052,
+    "eval_samples": 1654,
+    "eval_samples_per_second": 1098.84,
+    "eval_steps_per_second": 69.093
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:717d7386c720680fcad0d4b754a1ac5ca06b3bda824639c2bde37f8769b2c880
+size 21022551

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_tiny_spanish", "tokenizer_class": "AlbertTokenizer"}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.0,
+    "train_loss": 0.09274842659188383,
+    "train_runtime": 230.8308,
+    "train_samples": 14305,
+    "train_samples_per_second": 247.887,
+    "train_steps_per_second": 15.509
+}

trainer_state.json

@@ -0,0 +1,271 @@
+{
+  "best_metric": 0.08749764412641525,
+  "best_model_checkpoint": "/data/jcanete/all_results/pos/albeto_tiny/epochs_4_bs_16_lr_5e-5/checkpoint-1600",
+  "epoch": 4.0,
+  "global_step": 3580,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.22,
+      "eval_accuracy": 0.9538808220802765,
+      "eval_f1": 0.9490697836636672,
+      "eval_loss": 0.17476704716682434,
+      "eval_precision": 0.9467019204007793,
+      "eval_recall": 0.9514495214880822,
+      "eval_runtime": 1.5406,
+      "eval_samples_per_second": 1073.605,
+      "eval_steps_per_second": 67.506,
+      "step": 200
+    },
+    {
+      "epoch": 0.45,
+      "eval_accuracy": 0.9654529743650565,
+      "eval_f1": 0.9616131828280513,
+      "eval_loss": 0.125724658370018,
+      "eval_precision": 0.9590238861730456,
+      "eval_recall": 0.9642164991708457,
+      "eval_runtime": 1.5211,
+      "eval_samples_per_second": 1087.347,
+      "eval_steps_per_second": 68.37,
+      "step": 400
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.303072625698324e-05,
+      "loss": 0.3133,
+      "step": 500
+    },
+    {
+      "epoch": 0.67,
+      "eval_accuracy": 0.9684688500703139,
+      "eval_f1": 0.9654409345999343,
+      "eval_loss": 0.1092953085899353,
+      "eval_precision": 0.9633379749353491,
+      "eval_recall": 0.9675530958422409,
+      "eval_runtime": 1.5243,
+      "eval_samples_per_second": 1085.096,
+      "eval_steps_per_second": 68.229,
+      "step": 600
+    },
+    {
+      "epoch": 0.89,
+      "eval_accuracy": 0.9705189678250115,
+      "eval_f1": 0.9675733365251304,
+      "eval_loss": 0.09988456219434738,
+      "eval_precision": 0.9660797514241326,
+      "eval_recall": 0.9690715470220376,
+      "eval_runtime": 1.5122,
+      "eval_samples_per_second": 1093.787,
+      "eval_steps_per_second": 68.775,
+      "step": 800
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 3.604748603351956e-05,
+      "loss": 0.0983,
+      "step": 1000
+    },
+    {
+      "epoch": 1.12,
+      "eval_accuracy": 0.9718066450924248,
+      "eval_f1": 0.9694776714513557,
+      "eval_loss": 0.09871890395879745,
+      "eval_precision": 0.9673755197039925,
+      "eval_recall": 0.971588979241174,
+      "eval_runtime": 1.5084,
+      "eval_samples_per_second": 1096.509,
+      "eval_steps_per_second": 68.946,
+      "step": 1000
+    },
+    {
+      "epoch": 1.34,
+      "eval_accuracy": 0.9728401755307433,
+      "eval_f1": 0.9709789372905697,
+      "eval_loss": 0.09597848355770111,
+      "eval_precision": 0.9693355369267836,
+      "eval_recall": 0.9726279195220875,
+      "eval_runtime": 1.4942,
+      "eval_samples_per_second": 1106.933,
+      "eval_steps_per_second": 69.602,
+      "step": 1200
+    },
+    {
+      "epoch": 1.56,
+      "eval_accuracy": 0.9736026160180274,
+      "eval_f1": 0.9714610733031314,
+      "eval_loss": 0.09189929813146591,
+      "eval_precision": 0.9703765723740606,
+      "eval_recall": 0.9725480010389402,
+      "eval_runtime": 1.5177,
+      "eval_samples_per_second": 1089.787,
+      "eval_steps_per_second": 68.523,
+      "step": 1400
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 2.9064245810055868e-05,
+      "loss": 0.0695,
+      "step": 1500
+    },
+    {
+      "epoch": 1.79,
+      "eval_accuracy": 0.9747039189441047,
+      "eval_f1": 0.9728139460666986,
+      "eval_loss": 0.08749764412641525,
+      "eval_precision": 0.9711674399155732,
+      "eval_recall": 0.9744660446344728,
+      "eval_runtime": 1.5082,
+      "eval_samples_per_second": 1096.655,
+      "eval_steps_per_second": 68.955,
+      "step": 1600
+    },
+    {
+      "epoch": 2.01,
+      "eval_accuracy": 0.974975008895139,
+      "eval_f1": 0.9729659656365613,
+      "eval_loss": 0.08767995983362198,
+      "eval_precision": 0.9712323312761298,
+      "eval_recall": 0.9747058000839144,
+      "eval_runtime": 1.4537,
+      "eval_samples_per_second": 1137.75,
+      "eval_steps_per_second": 71.539,
+      "step": 1800
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 2.2094972067039108e-05,
+      "loss": 0.0597,
+      "step": 2000
+    },
+    {
+      "epoch": 2.23,
+      "eval_accuracy": 0.974636146456346,
+      "eval_f1": 0.9728004468426774,
+      "eval_loss": 0.08931880444288254,
+      "eval_precision": 0.9712794773741237,
+      "eval_recall": 0.9743261872889653,
+      "eval_runtime": 1.4451,
+      "eval_samples_per_second": 1144.553,
+      "eval_steps_per_second": 71.967,
+      "step": 2000
+    },
+    {
+      "epoch": 2.46,
+      "eval_accuracy": 0.975025838260958,
+      "eval_f1": 0.9732900972811175,
+      "eval_loss": 0.0905543640255928,
+      "eval_precision": 0.9720971020847451,
+      "eval_recall": 0.9744860242552597,
+      "eval_runtime": 1.4435,
+      "eval_samples_per_second": 1145.863,
+      "eval_steps_per_second": 72.049,
+      "step": 2200
+    },
+    {
+      "epoch": 2.68,
+      "eval_accuracy": 0.9755510750410871,
+      "eval_f1": 0.9741129056392965,
+      "eval_loss": 0.09149234741926193,
+      "eval_precision": 0.9727446605036659,
+      "eval_recall": 0.9754850052945995,
+      "eval_runtime": 1.4454,
+      "eval_samples_per_second": 1144.307,
+      "eval_steps_per_second": 71.952,
+      "step": 2400
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 1.511173184357542e-05,
+      "loss": 0.0458,
+      "step": 2500
+    },
+    {
+      "epoch": 2.91,
+      "eval_accuracy": 0.9755171887972077,
+      "eval_f1": 0.9740555211846648,
+      "eval_loss": 0.08821560442447662,
+      "eval_precision": 0.9729681239160337,
+      "eval_recall": 0.975145351741224,
+      "eval_runtime": 1.4692,
+      "eval_samples_per_second": 1125.788,
+      "eval_steps_per_second": 70.787,
+      "step": 2600
+    },
+    {
+      "epoch": 3.13,
+      "eval_accuracy": 0.9757374493824232,
+      "eval_f1": 0.9740832369019593,
+      "eval_loss": 0.09264585375785828,
+      "eval_precision": 0.9727053573207419,
+      "eval_recall": 0.9754650256738127,
+      "eval_runtime": 1.4438,
+      "eval_samples_per_second": 1145.593,
+      "eval_steps_per_second": 72.032,
+      "step": 2800
+    },
+    {
+      "epoch": 3.35,
+      "learning_rate": 8.128491620111732e-06,
+      "loss": 0.0383,
+      "step": 3000
+    },
+    {
+      "epoch": 3.35,
+      "eval_accuracy": 0.9753985869436302,
+      "eval_f1": 0.973785013167345,
+      "eval_loss": 0.093608058989048,
+      "eval_precision": 0.9723688666653386,
+      "eval_recall": 0.9752052906035843,
+      "eval_runtime": 1.444,
+      "eval_samples_per_second": 1145.395,
+      "eval_steps_per_second": 72.02,
+      "step": 3000
+    },
+    {
+      "epoch": 3.58,
+      "eval_accuracy": 0.9756357906507853,
+      "eval_f1": 0.974050980196538,
+      "eval_loss": 0.09429396688938141,
+      "eval_precision": 0.9727602423083055,
+      "eval_recall": 0.9753451479490919,
+      "eval_runtime": 1.4392,
+      "eval_samples_per_second": 1149.213,
+      "eval_steps_per_second": 72.26,
+      "step": 3200
+    },
+    {
+      "epoch": 3.8,
+      "eval_accuracy": 0.9750936107487166,
+      "eval_f1": 0.9735209019255712,
+      "eval_loss": 0.09423112124204636,
+      "eval_precision": 0.9722792403196556,
+      "eval_recall": 0.9747657389462748,
+      "eval_runtime": 1.4459,
+      "eval_samples_per_second": 1143.936,
+      "eval_steps_per_second": 71.928,
+      "step": 3400
+    },
+    {
+      "epoch": 3.91,
+      "learning_rate": 1.1452513966480447e-06,
+      "loss": 0.0344,
+      "step": 3500
+    },
+    {
+      "epoch": 4.0,
+      "step": 3580,
+      "total_flos": 32358086511744.0,
+      "train_loss": 0.09274842659188383,
+      "train_runtime": 230.8308,
+      "train_samples_per_second": 247.887,
+      "train_steps_per_second": 15.509
+    }
+  ],
+  "max_steps": 3580,
+  "num_train_epochs": 4,
+  "total_flos": 32358086511744.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af36acbe1b11273adc4b0a2f9e6702459837f2fdc63a75485358b2911aa68b20
+size 2863