adding model fine-tuned on QA (SQAC)

all_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 4.0,
+    "eval_exact_match": 40.50429184549356,
+    "eval_f1": 59.382051635096076,
+    "eval_samples": 2082,
+    "train_loss": 1.34779805898207,
+    "train_runtime": 436.6525,
+    "train_samples": 16598,
+    "train_samples_per_second": 152.048,
+    "train_steps_per_second": 9.509
+}

config.json

@@ -0,0 +1,33 @@
+{
+  "_name_or_path": "CenIA/albert_tiny_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": 312,
+  "initializer_range": 0.02,
+  "inner_group_num": 1,
+  "intermediate_size": 1248,
+  "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": "4.15.0",
+  "type_vocab_size": 2,
+  "vocab_size": 31000
+}

eval_results.json

@@ -0,0 +1,6 @@
+{
+    "epoch": 4.0,
+    "eval_exact_match": 40.50429184549356,
+    "eval_f1": 59.382051635096076,
+    "eval_samples": 2082
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0beeb32e24c386397b5836cb41392f35c0bc59b58b5694f26fc8af1e71e309e6
+size 21002519

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": 1.34779805898207,
+    "train_runtime": 436.6525,
+    "train_samples": 16598,
+    "train_samples_per_second": 152.048,
+    "train_steps_per_second": 9.509
+}

trainer_state.json

@@ -0,0 +1,193 @@
+{
+  "best_metric": 59.382051635096076,
+  "best_model_checkpoint": "/data/jcanete/all_results/sqac/albeto_tiny/epochs_4_bs_16_lr_5e-5/checkpoint-2400",
+  "epoch": 4.0,
+  "global_step": 4152,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.19,
+      "eval_exact_match": 12.339055793991417,
+      "eval_f1": 28.93103280159431,
+      "step": 200
+    },
+    {
+      "epoch": 0.39,
+      "eval_exact_match": 23.92703862660944,
+      "eval_f1": 41.67915333843358,
+      "step": 400
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.4026974951830445e-05,
+      "loss": 3.1377,
+      "step": 500
+    },
+    {
+      "epoch": 0.58,
+      "eval_exact_match": 31.86695278969957,
+      "eval_f1": 51.18317520944259,
+      "step": 600
+    },
+    {
+      "epoch": 0.77,
+      "eval_exact_match": 34.871244635193136,
+      "eval_f1": 53.9432519365253,
+      "step": 800
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 3.800578034682081e-05,
+      "loss": 2.0078,
+      "step": 1000
+    },
+    {
+      "epoch": 0.96,
+      "eval_exact_match": 35.51502145922747,
+      "eval_f1": 55.221685158386705,
+      "step": 1000
+    },
+    {
+      "epoch": 1.16,
+      "eval_exact_match": 36.85622317596567,
+      "eval_f1": 55.20708770919938,
+      "step": 1200
+    },
+    {
+      "epoch": 1.35,
+      "eval_exact_match": 37.98283261802575,
+      "eval_f1": 56.614251341537866,
+      "step": 1400
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 3.198458574181118e-05,
+      "loss": 1.4374,
+      "step": 1500
+    },
+    {
+      "epoch": 1.54,
+      "eval_exact_match": 39.002145922746784,
+      "eval_f1": 57.702126636307476,
+      "step": 1600
+    },
+    {
+      "epoch": 1.73,
+      "eval_exact_match": 39.64592274678112,
+      "eval_f1": 58.69230555304524,
+      "step": 1800
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 2.596339113680154e-05,
+      "loss": 1.3597,
+      "step": 2000
+    },
+    {
+      "epoch": 1.93,
+      "eval_exact_match": 39.64592274678112,
+      "eval_f1": 58.05721827009323,
+      "step": 2000
+    },
+    {
+      "epoch": 2.12,
+      "eval_exact_match": 39.27038626609442,
+      "eval_f1": 57.648146496721296,
+      "step": 2200
+    },
+    {
+      "epoch": 2.31,
+      "eval_exact_match": 40.50429184549356,
+      "eval_f1": 59.382051635096076,
+      "step": 2400
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 1.995423892100193e-05,
+      "loss": 0.9536,
+      "step": 2500
+    },
+    {
+      "epoch": 2.5,
+      "eval_exact_match": 40.71888412017167,
+      "eval_f1": 59.060573449175955,
+      "step": 2600
+    },
+    {
+      "epoch": 2.7,
+      "eval_exact_match": 39.8068669527897,
+      "eval_f1": 58.80449299317162,
+      "step": 2800
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 1.3933044315992294e-05,
+      "loss": 0.8884,
+      "step": 3000
+    },
+    {
+      "epoch": 2.89,
+      "eval_exact_match": 40.772532188841204,
+      "eval_f1": 58.85814483853365,
+      "step": 3000
+    },
+    {
+      "epoch": 3.08,
+      "eval_exact_match": 40.02145922746781,
+      "eval_f1": 59.12527092459649,
+      "step": 3200
+    },
+    {
+      "epoch": 3.28,
+      "eval_exact_match": 40.933476394849784,
+      "eval_f1": 59.19992976262888,
+      "step": 3400
+    },
+    {
+      "epoch": 3.37,
+      "learning_rate": 7.91184971098266e-06,
+      "loss": 0.6567,
+      "step": 3500
+    },
+    {
+      "epoch": 3.47,
+      "eval_exact_match": 40.933476394849784,
+      "eval_f1": 59.280000618875235,
+      "step": 3600
+    },
+    {
+      "epoch": 3.66,
+      "eval_exact_match": 40.50429184549356,
+      "eval_f1": 58.81358724565813,
+      "step": 3800
+    },
+    {
+      "epoch": 3.85,
+      "learning_rate": 1.890655105973025e-06,
+      "loss": 0.5809,
+      "step": 4000
+    },
+    {
+      "epoch": 3.85,
+      "eval_exact_match": 40.50429184549356,
+      "eval_f1": 58.87104735095327,
+      "step": 4000
+    },
+    {
+      "epoch": 4.0,
+      "step": 4152,
+      "total_flos": 241925105088000.0,
+      "train_loss": 1.34779805898207,
+      "train_runtime": 436.6525,
+      "train_samples_per_second": 152.048,
+      "train_steps_per_second": 9.509
+    }
+  ],
+  "max_steps": 4152,
+  "num_train_epochs": 4,
+  "total_flos": 241925105088000.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27f20796b688bec5325b88f516fb0f76d5418b718d7e7da9868af8d8a47db176
+size 2991