adding model fine-tuned on QA (SQAC)

all_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 4.0,
+    "eval_exact_match": 46.24463519313305,
+    "eval_f1": 63.88217177699186,
+    "eval_samples": 2088,
+    "train_loss": 1.5509035589605484,
+    "train_runtime": 2076.6526,
+    "train_samples": 16624,
+    "train_samples_per_second": 32.021,
+    "train_steps_per_second": 2.001
+}

config.json

@@ -0,0 +1,24 @@
+{
+  "_name_or_path": "CenIA/distillbert-base-spanish-uncased",
+  "activation": "gelu",
+  "architectures": [
+    "DistilBertForQuestionAnswering"
+  ],
+  "attention_dropout": 0.1,
+  "dim": 768,
+  "dropout": 0.1,
+  "hidden_dim": 3072,
+  "initializer_range": 0.02,
+  "max_position_embeddings": 512,
+  "model_type": "distilbert",
+  "n_heads": 12,
+  "n_layers": 6,
+  "pad_token_id": 0,
+  "qa_dropout": 0.1,
+  "seq_classif_dropout": 0.2,
+  "sinusoidal_pos_embds": true,
+  "tie_weights_": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.15.0",
+  "vocab_size": 31002
+}

eval_results.json

@@ -0,0 +1,6 @@
+{
+    "epoch": 4.0,
+    "eval_exact_match": 46.24463519313305,
+    "eval_f1": 63.88217177699186,
+    "eval_samples": 2088
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c718d0fcc506239ed23bfb4d27e91c738e9e4b49f5661fc1f67f0bf36bc3e7e5
+size 266971637

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": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": false, "do_basic_tokenize": true, "never_split": null, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "CenIA/distillbert-base-spanish-uncased", "tokenizer_class": "DistilBertTokenizer"}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.0,
+    "train_loss": 1.5509035589605484,
+    "train_runtime": 2076.6526,
+    "train_samples": 16624,
+    "train_samples_per_second": 32.021,
+    "train_steps_per_second": 2.001
+}

trainer_state.json

@@ -0,0 +1,193 @@
+{
+  "best_metric": 63.88217177699186,
+  "best_model_checkpoint": "/data/jcanete/all_results/sqac/distillbeto/epochs_4_bs_16_lr_5e-5/checkpoint-3600",
+  "epoch": 4.0,
+  "global_step": 4156,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.19,
+      "eval_exact_match": 8.959227467811159,
+      "eval_f1": 18.27584505760202,
+      "step": 200
+    },
+    {
+      "epoch": 0.38,
+      "eval_exact_match": 19.15236051502146,
+      "eval_f1": 35.446786517907775,
+      "step": 400
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.402069297401348e-05,
+      "loss": 3.5504,
+      "step": 500
+    },
+    {
+      "epoch": 0.58,
+      "eval_exact_match": 28.916309012875537,
+      "eval_f1": 46.039519440199285,
+      "step": 600
+    },
+    {
+      "epoch": 0.77,
+      "eval_exact_match": 34.17381974248927,
+      "eval_f1": 51.92505825555892,
+      "step": 800
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 3.800529355149182e-05,
+      "loss": 2.427,
+      "step": 1000
+    },
+    {
+      "epoch": 0.96,
+      "eval_exact_match": 37.71459227467811,
+      "eval_f1": 55.374436147788664,
+      "step": 1000
+    },
+    {
+      "epoch": 1.15,
+      "eval_exact_match": 39.216738197424895,
+      "eval_f1": 57.333325590107926,
+      "step": 1200
+    },
+    {
+      "epoch": 1.35,
+      "eval_exact_match": 42.27467811158798,
+      "eval_f1": 59.56679980116175,
+      "step": 1400
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 3.1989894128970165e-05,
+      "loss": 1.6405,
+      "step": 1500
+    },
+    {
+      "epoch": 1.54,
+      "eval_exact_match": 42.7038626609442,
+      "eval_f1": 59.74522889845748,
+      "step": 1600
+    },
+    {
+      "epoch": 1.73,
+      "eval_exact_match": 44.47424892703863,
+      "eval_f1": 61.88744236282341,
+      "step": 1800
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 2.597449470644851e-05,
+      "loss": 1.5245,
+      "step": 2000
+    },
+    {
+      "epoch": 1.92,
+      "eval_exact_match": 43.562231759656655,
+      "eval_f1": 61.742252102489076,
+      "step": 2000
+    },
+    {
+      "epoch": 2.12,
+      "eval_exact_match": 44.527896995708154,
+      "eval_f1": 62.510713884003145,
+      "step": 2200
+    },
+    {
+      "epoch": 2.31,
+      "eval_exact_match": 44.09871244635193,
+      "eval_f1": 61.82732823382964,
+      "step": 2400
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 1.9971126082771897e-05,
+      "loss": 1.1014,
+      "step": 2500
+    },
+    {
+      "epoch": 2.5,
+      "eval_exact_match": 44.4206008583691,
+      "eval_f1": 62.43808399893021,
+      "step": 2600
+    },
+    {
+      "epoch": 2.69,
+      "eval_exact_match": 44.68884120171674,
+      "eval_f1": 61.28700070537817,
+      "step": 2800
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 1.395572666025024e-05,
+      "loss": 0.9906,
+      "step": 3000
+    },
+    {
+      "epoch": 2.89,
+      "eval_exact_match": 45.01072961373391,
+      "eval_f1": 62.82370988061508,
+      "step": 3000
+    },
+    {
+      "epoch": 3.08,
+      "eval_exact_match": 45.11802575107296,
+      "eval_f1": 63.06687333231593,
+      "step": 3200
+    },
+    {
+      "epoch": 3.27,
+      "eval_exact_match": 45.708154506437765,
+      "eval_f1": 63.02210675456867,
+      "step": 3400
+    },
+    {
+      "epoch": 3.37,
+      "learning_rate": 7.95235803657363e-06,
+      "loss": 0.7625,
+      "step": 3500
+    },
+    {
+      "epoch": 3.46,
+      "eval_exact_match": 46.24463519313305,
+      "eval_f1": 63.88217177699186,
+      "step": 3600
+    },
+    {
+      "epoch": 3.66,
+      "eval_exact_match": 46.24463519313305,
+      "eval_f1": 63.6386746913192,
+      "step": 3800
+    },
+    {
+      "epoch": 3.85,
+      "learning_rate": 1.936958614051973e-06,
+      "loss": 0.6729,
+      "step": 4000
+    },
+    {
+      "epoch": 3.85,
+      "eval_exact_match": 46.08369098712446,
+      "eval_f1": 63.82552212938861,
+      "step": 4000
+    },
+    {
+      "epoch": 4.0,
+      "step": 4156,
+      "total_flos": 8497742952631296.0,
+      "train_loss": 1.5509035589605484,
+      "train_runtime": 2076.6526,
+      "train_samples_per_second": 32.021,
+      "train_steps_per_second": 2.001
+    }
+  ],
+  "max_steps": 4156,
+  "num_train_epochs": 4,
+  "total_flos": 8497742952631296.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b71db473aa9d045f98ec7746d8d2124d1e8b5be85f8e0cb5a03de3e71410469
+size 2991