adding model finetuned on PAWS-X

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 2.0,
+    "eval_accuracy": 0.9039999842643738,
+    "eval_loss": 0.25371718406677246,
+    "eval_runtime": 15.8096,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 126.506,
+    "eval_steps_per_second": 7.907,
+    "train_loss": 0.2236284066976043,
+    "train_runtime": 8006.2239,
+    "train_samples": 49401,
+    "train_samples_per_second": 12.341,
+    "train_steps_per_second": 0.771
+}

config.json

@@ -0,0 +1,35 @@
+{
+  "_name_or_path": "CenIA/albert_xxlarge_spanish",
+  "architectures": [
+    "AlbertForSequenceClassification"
+  ],
+  "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",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.12.5",
+  "type_vocab_size": 2,
+  "vocab_size": 31000
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 2.0,
+    "eval_accuracy": 0.9039999842643738,
+    "eval_loss": 0.25371718406677246,
+    "eval_runtime": 15.8096,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 126.506,
+    "eval_steps_per_second": 7.907
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9dd0a4132c38c056bcd145eeadb3289ea28c10aa44af00a16b9bb906c4bb1ed8
+size 890942161

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": 0.2236284066976043,
+    "train_runtime": 8006.2239,
+    "train_samples": 49401,
+    "train_samples_per_second": 12.341,
+    "train_steps_per_second": 0.771
+}

trainer_state.json

@@ -0,0 +1,277 @@
+{
+  "best_metric": 0.25371718406677246,
+  "best_model_checkpoint": "/data/jcanete/all_results/pawsx/albeto_xxlarge/epochs_2_bs_16_lr_5e-6/checkpoint-2100",
+  "epoch": 2.0,
+  "global_step": 6176,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.1,
+      "eval_accuracy": 0.609000027179718,
+      "eval_loss": 0.6674321889877319,
+      "eval_runtime": 16.3969,
+      "eval_samples_per_second": 121.974,
+      "eval_steps_per_second": 7.623,
+      "step": 300
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.599255181347151e-06,
+      "loss": 0.6462,
+      "step": 500
+    },
+    {
+      "epoch": 0.19,
+      "eval_accuracy": 0.8460000157356262,
+      "eval_loss": 0.3770250976085663,
+      "eval_runtime": 16.4044,
+      "eval_samples_per_second": 121.918,
+      "eval_steps_per_second": 7.62,
+      "step": 600
+    },
+    {
+      "epoch": 0.29,
+      "eval_accuracy": 0.8659999966621399,
+      "eval_loss": 0.34182578325271606,
+      "eval_runtime": 16.4066,
+      "eval_samples_per_second": 121.903,
+      "eval_steps_per_second": 7.619,
+      "step": 900
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.194462435233161e-06,
+      "loss": 0.3074,
+      "step": 1000
+    },
+    {
+      "epoch": 0.39,
+      "eval_accuracy": 0.890999972820282,
+      "eval_loss": 0.3314864933490753,
+      "eval_runtime": 16.4011,
+      "eval_samples_per_second": 121.943,
+      "eval_steps_per_second": 7.621,
+      "step": 1200
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 3.7896696891191716e-06,
+      "loss": 0.2366,
+      "step": 1500
+    },
+    {
+      "epoch": 0.49,
+      "eval_accuracy": 0.8884999752044678,
+      "eval_loss": 0.2938086986541748,
+      "eval_runtime": 16.3983,
+      "eval_samples_per_second": 121.964,
+      "eval_steps_per_second": 7.623,
+      "step": 1500
+    },
+    {
+      "epoch": 0.58,
+      "eval_accuracy": 0.8974999785423279,
+      "eval_loss": 0.2964068055152893,
+      "eval_runtime": 16.4019,
+      "eval_samples_per_second": 121.937,
+      "eval_steps_per_second": 7.621,
+      "step": 1800
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 3.3856865284974094e-06,
+      "loss": 0.2202,
+      "step": 2000
+    },
+    {
+      "epoch": 0.68,
+      "eval_accuracy": 0.9039999842643738,
+      "eval_loss": 0.25371718406677246,
+      "eval_runtime": 16.4082,
+      "eval_samples_per_second": 121.891,
+      "eval_steps_per_second": 7.618,
+      "step": 2100
+    },
+    {
+      "epoch": 0.78,
+      "eval_accuracy": 0.9020000100135803,
+      "eval_loss": 0.2744397819042206,
+      "eval_runtime": 16.4021,
+      "eval_samples_per_second": 121.936,
+      "eval_steps_per_second": 7.621,
+      "step": 2400
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 2.98089378238342e-06,
+      "loss": 0.2175,
+      "step": 2500
+    },
+    {
+      "epoch": 0.87,
+      "eval_accuracy": 0.9070000052452087,
+      "eval_loss": 0.31618574261665344,
+      "eval_runtime": 16.4017,
+      "eval_samples_per_second": 121.939,
+      "eval_steps_per_second": 7.621,
+      "step": 2700
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 2.5761010362694307e-06,
+      "loss": 0.2036,
+      "step": 3000
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.9049999713897705,
+      "eval_loss": 0.32718589901924133,
+      "eval_runtime": 16.408,
+      "eval_samples_per_second": 121.891,
+      "eval_steps_per_second": 7.618,
+      "step": 3000
+    },
+    {
+      "epoch": 1.07,
+      "eval_accuracy": 0.9004999995231628,
+      "eval_loss": 0.36231866478919983,
+      "eval_runtime": 16.4102,
+      "eval_samples_per_second": 121.875,
+      "eval_steps_per_second": 7.617,
+      "step": 3300
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 2.1713082901554404e-06,
+      "loss": 0.1545,
+      "step": 3500
+    },
+    {
+      "epoch": 1.17,
+      "eval_accuracy": 0.8989999890327454,
+      "eval_loss": 0.4374341368675232,
+      "eval_runtime": 16.3991,
+      "eval_samples_per_second": 121.958,
+      "eval_steps_per_second": 7.622,
+      "step": 3600
+    },
+    {
+      "epoch": 1.26,
+      "eval_accuracy": 0.906499981880188,
+      "eval_loss": 0.36568117141723633,
+      "eval_runtime": 16.4019,
+      "eval_samples_per_second": 121.937,
+      "eval_steps_per_second": 7.621,
+      "step": 3900
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.7665155440414508e-06,
+      "loss": 0.1491,
+      "step": 4000
+    },
+    {
+      "epoch": 1.36,
+      "eval_accuracy": 0.9035000205039978,
+      "eval_loss": 0.37311026453971863,
+      "eval_runtime": 16.3997,
+      "eval_samples_per_second": 121.954,
+      "eval_steps_per_second": 7.622,
+      "step": 4200
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.3617227979274613e-06,
+      "loss": 0.1509,
+      "step": 4500
+    },
+    {
+      "epoch": 1.46,
+      "eval_accuracy": 0.9020000100135803,
+      "eval_loss": 0.4289727210998535,
+      "eval_runtime": 16.4113,
+      "eval_samples_per_second": 121.867,
+      "eval_steps_per_second": 7.617,
+      "step": 4500
+    },
+    {
+      "epoch": 1.55,
+      "eval_accuracy": 0.909500002861023,
+      "eval_loss": 0.37183818221092224,
+      "eval_runtime": 16.4021,
+      "eval_samples_per_second": 121.935,
+      "eval_steps_per_second": 7.621,
+      "step": 4800
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 9.569300518134715e-07,
+      "loss": 0.1495,
+      "step": 5000
+    },
+    {
+      "epoch": 1.65,
+      "eval_accuracy": 0.9100000262260437,
+      "eval_loss": 0.42287707328796387,
+      "eval_runtime": 16.3951,
+      "eval_samples_per_second": 121.988,
+      "eval_steps_per_second": 7.624,
+      "step": 5100
+    },
+    {
+      "epoch": 1.75,
+      "eval_accuracy": 0.9054999947547913,
+      "eval_loss": 0.4088725745677948,
+      "eval_runtime": 16.4026,
+      "eval_samples_per_second": 121.932,
+      "eval_steps_per_second": 7.621,
+      "step": 5400
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 5.521373056994819e-07,
+      "loss": 0.1339,
+      "step": 5500
+    },
+    {
+      "epoch": 1.85,
+      "eval_accuracy": 0.9085000157356262,
+      "eval_loss": 0.36614999175071716,
+      "eval_runtime": 16.403,
+      "eval_samples_per_second": 121.929,
+      "eval_steps_per_second": 7.621,
+      "step": 5700
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.4734455958549224e-07,
+      "loss": 0.1488,
+      "step": 6000
+    },
+    {
+      "epoch": 1.94,
+      "eval_accuracy": 0.9104999899864197,
+      "eval_loss": 0.36664167046546936,
+      "eval_runtime": 16.4079,
+      "eval_samples_per_second": 121.892,
+      "eval_steps_per_second": 7.618,
+      "step": 6000
+    },
+    {
+      "epoch": 2.0,
+      "step": 6176,
+      "total_flos": 1.2074879173071936e+16,
+      "train_loss": 0.2236284066976043,
+      "train_runtime": 8006.2239,
+      "train_samples_per_second": 12.341,
+      "train_steps_per_second": 0.771
+    }
+  ],
+  "max_steps": 6176,
+  "num_train_epochs": 2,
+  "total_flos": 1.2074879173071936e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ee7101177dbffa5edf4e38268a119f86f1de25aa2e201a978c697319624ad9b
+size 2863