End of training

README.md

@@ -2,11 +2,24 @@
 base_model: demdecuong/vihealthbert-base-word
 tags:
 - generated_from_trainer
+datasets:
+- tmnam20/pretrained-vn-med-nli
 metrics:
 - accuracy
 model-index:
 - name: vihealthbert-w_unsup-SynPD
-  results: []
+  results:
+  - task:
+      name: Masked Language Modeling
+      type: fill-mask
+    dataset:
+      name: tmnam20/pretrained-vn-med-nli all
+      type: tmnam20/pretrained-vn-med-nli
+      args: all
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.686153705209395
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -14,10 +27,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # vihealthbert-w_unsup-SynPD
 
-This model is a fine-tuned version of [demdecuong/vihealthbert-base-word](https://huggingface.co/demdecuong/vihealthbert-base-word) on an unknown dataset.
+This model is a fine-tuned version of [demdecuong/vihealthbert-base-word](https://huggingface.co/demdecuong/vihealthbert-base-word) on the tmnam20/pretrained-vn-med-nli all dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.5540
-- Accuracy: 0.6880
+- Loss: 1.5768
+- Accuracy: 0.6862
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.686153705209395,
+    "eval_loss": 1.5768413543701172,
+    "eval_runtime": 36.0583,
+    "eval_samples": 9774,
+    "eval_samples_per_second": 271.061,
+    "eval_steps_per_second": 16.945,
+    "perplexity": 4.83964491985086,
+    "total_flos": 0.0,
+    "train_loss": 4.85471958516693,
+    "train_runtime": 21406.4835,
+    "train_samples": 185684,
+    "train_samples_per_second": 86.742,
+    "train_steps_per_second": 2.711
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.686153705209395,
+    "eval_loss": 1.5768413543701172,
+    "eval_runtime": 36.0583,
+    "eval_samples": 9774,
+    "eval_samples_per_second": 271.061,
+    "eval_steps_per_second": 16.945,
+    "perplexity": 4.83964491985086
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 10.0,
+    "total_flos": 0.0,
+    "train_loss": 4.85471958516693,
+    "train_runtime": 21406.4835,
+    "train_samples": 185684,
+    "train_samples_per_second": 86.742,
+    "train_steps_per_second": 2.711
+}

trainer_state.json

@@ -0,0 +1,213 @@
+{
+  "best_metric": 0.6879923855830649,
+  "best_model_checkpoint": "output/pretraining/vihealthbert-w_unsup-SynPD/lr3e-5_wr0.1_wd0.0/checkpoint-55000",
+  "epoch": 10.0,
+  "eval_steps": 5000,
+  "global_step": 58030,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00017232465965879716,
+      "grad_norm": 41.827266693115234,
+      "learning_rate": 5.169739789763915e-09,
+      "loss": 13.7242,
+      "step": 1
+    },
+    {
+      "epoch": 0.8616232982939859,
+      "grad_norm": 5.6151957511901855,
+      "learning_rate": 2.584869894881958e-05,
+      "loss": 7.0234,
+      "step": 5000
+    },
+    {
+      "epoch": 0.8616232982939859,
+      "eval_accuracy": 0.5576033171446898,
+      "eval_loss": 2.590928077697754,
+      "eval_runtime": 37.589,
+      "eval_samples_per_second": 260.023,
+      "eval_steps_per_second": 16.255,
+      "step": 5000
+    },
+    {
+      "epoch": 1.7232465965879717,
+      "grad_norm": 4.410743713378906,
+      "learning_rate": 2.758917801137343e-05,
+      "loss": 5.2736,
+      "step": 10000
+    },
+    {
+      "epoch": 1.7232465965879717,
+      "eval_accuracy": 0.596153624823046,
+      "eval_loss": 2.188957452774048,
+      "eval_runtime": 37.6297,
+      "eval_samples_per_second": 259.741,
+      "eval_steps_per_second": 16.237,
+      "step": 10000
+    },
+    {
+      "epoch": 2.5848698948819577,
+      "grad_norm": 4.754213809967041,
+      "learning_rate": 2.4717100350393475e-05,
+      "loss": 4.9126,
+      "step": 15000
+    },
+    {
+      "epoch": 2.5848698948819577,
+      "eval_accuracy": 0.6381037526075694,
+      "eval_loss": 1.909491777420044,
+      "eval_runtime": 36.8556,
+      "eval_samples_per_second": 265.197,
+      "eval_steps_per_second": 16.578,
+      "step": 15000
+    },
+    {
+      "epoch": 3.4464931931759435,
+      "grad_norm": 4.2047200202941895,
+      "learning_rate": 2.184502268941352e-05,
+      "loss": 4.791,
+      "step": 20000
+    },
+    {
+      "epoch": 3.4464931931759435,
+      "eval_accuracy": 0.6468589272593681,
+      "eval_loss": 1.8286069631576538,
+      "eval_runtime": 35.9342,
+      "eval_samples_per_second": 271.997,
+      "eval_steps_per_second": 17.003,
+      "step": 20000
+    },
+    {
+      "epoch": 4.30811649146993,
+      "grad_norm": 4.629519462585449,
+      "learning_rate": 1.897294502843357e-05,
+      "loss": 4.6538,
+      "step": 25000
+    },
+    {
+      "epoch": 4.30811649146993,
+      "eval_accuracy": 0.6644416567883901,
+      "eval_loss": 1.714424967765808,
+      "eval_runtime": 35.6122,
+      "eval_samples_per_second": 274.456,
+      "eval_steps_per_second": 17.157,
+      "step": 25000
+    },
+    {
+      "epoch": 5.1697397897639155,
+      "grad_norm": 3.993448257446289,
+      "learning_rate": 1.6100867367453616e-05,
+      "loss": 4.5846,
+      "step": 30000
+    },
+    {
+      "epoch": 5.1697397897639155,
+      "eval_accuracy": 0.6703549228453575,
+      "eval_loss": 1.6779303550720215,
+      "eval_runtime": 36.1572,
+      "eval_samples_per_second": 270.32,
+      "eval_steps_per_second": 16.898,
+      "step": 30000
+    },
+    {
+      "epoch": 6.031363088057901,
+      "grad_norm": 4.6959357261657715,
+      "learning_rate": 1.3228789706473663e-05,
+      "loss": 4.5568,
+      "step": 35000
+    },
+    {
+      "epoch": 6.031363088057901,
+      "eval_accuracy": 0.6765665494261385,
+      "eval_loss": 1.6362268924713135,
+      "eval_runtime": 36.2906,
+      "eval_samples_per_second": 269.326,
+      "eval_steps_per_second": 16.836,
+      "step": 35000
+    },
+    {
+      "epoch": 6.892986386351887,
+      "grad_norm": 4.4082441329956055,
+      "learning_rate": 1.035671204549371e-05,
+      "loss": 4.5079,
+      "step": 40000
+    },
+    {
+      "epoch": 6.892986386351887,
+      "eval_accuracy": 0.681425908452811,
+      "eval_loss": 1.6008453369140625,
+      "eval_runtime": 36.1284,
+      "eval_samples_per_second": 270.535,
+      "eval_steps_per_second": 16.912,
+      "step": 40000
+    },
+    {
+      "epoch": 7.754609684645873,
+      "grad_norm": 5.832913398742676,
+      "learning_rate": 7.484634384513757e-06,
+      "loss": 4.469,
+      "step": 45000
+    },
+    {
+      "epoch": 7.754609684645873,
+      "eval_accuracy": 0.6805218690729352,
+      "eval_loss": 1.6063588857650757,
+      "eval_runtime": 36.0752,
+      "eval_samples_per_second": 270.934,
+      "eval_steps_per_second": 16.937,
+      "step": 45000
+    },
+    {
+      "epoch": 8.61623298293986,
+      "grad_norm": 3.991995096206665,
+      "learning_rate": 4.612556723533804e-06,
+      "loss": 4.4514,
+      "step": 50000
+    },
+    {
+      "epoch": 8.61623298293986,
+      "eval_accuracy": 0.6852411542306852,
+      "eval_loss": 1.5799689292907715,
+      "eval_runtime": 37.5803,
+      "eval_samples_per_second": 260.083,
+      "eval_steps_per_second": 16.259,
+      "step": 50000
+    },
+    {
+      "epoch": 9.477856281233844,
+      "grad_norm": 4.424502849578857,
+      "learning_rate": 1.7404790625538515e-06,
+      "loss": 4.4317,
+      "step": 55000
+    },
+    {
+      "epoch": 9.477856281233844,
+      "eval_accuracy": 0.6879923855830649,
+      "eval_loss": 1.5539859533309937,
+      "eval_runtime": 35.7439,
+      "eval_samples_per_second": 273.445,
+      "eval_steps_per_second": 17.094,
+      "step": 55000
+    },
+    {
+      "epoch": 10.0,
+      "step": 58030,
+      "total_flos": 0.0,
+      "train_loss": 4.85471958516693,
+      "train_runtime": 21406.4835,
+      "train_samples_per_second": 86.742,
+      "train_steps_per_second": 2.711
+    }
+  ],
+  "logging_steps": 5000,
+  "max_steps": 58030,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 5000,
+  "total_flos": 0.0,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}