End of training

README.md

@@ -3,9 +3,23 @@ license: apache-2.0
 base_model: google/mt5-small
 tags:
 - generated_from_trainer
+datasets:
+- cwchang/tw_address_large
+metrics:
+- rouge
 model-index:
 - name: formatted_address
-  results: []
+  results:
+  - task:
+      name: Summarization
+      type: summarization
+    dataset:
+      name: cwchang/tw_address_large
+      type: cwchang/tw_address_large
+    metrics:
+    - name: Rouge1
+      type: rouge
+      value: 99.396
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -13,7 +27,14 @@ should probably proofread and complete it, then remove this comment. -->
 
 # formatted_address
 
-This model is a fine-tuned version of [google/mt5-small](https://huggingface.co/google/mt5-small) on an unknown dataset.
+This model is a fine-tuned version of [google/mt5-small](https://huggingface.co/google/mt5-small) on the cwchang/tw_address_large dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.0613
+- Rouge1: 99.396
+- Rouge2: 50.6957
+- Rougel: 99.3961
+- Rougelsum: 99.4032
+- Gen Len: 13.744
 
 ## Model description
 

all_results.json

@@ -0,0 +1,18 @@
+{
+    "epoch": 3.0,
+    "eval_gen_len": 13.744,
+    "eval_loss": 0.06133730337023735,
+    "eval_rouge1": 99.396,
+    "eval_rouge2": 50.6957,
+    "eval_rougeL": 99.3961,
+    "eval_rougeLsum": 99.4032,
+    "eval_runtime": 149.1742,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 13.407,
+    "eval_steps_per_second": 3.352,
+    "train_loss": 0.2953249713897705,
+    "train_runtime": 2842.9361,
+    "train_samples": 20000,
+    "train_samples_per_second": 21.105,
+    "train_steps_per_second": 5.276
+}

eval_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 3.0,
+    "eval_gen_len": 13.744,
+    "eval_loss": 0.06133730337023735,
+    "eval_rouge1": 99.396,
+    "eval_rouge2": 50.6957,
+    "eval_rougeL": 99.3961,
+    "eval_rougeLsum": 99.4032,
+    "eval_runtime": 149.1742,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 13.407,
+    "eval_steps_per_second": 3.352
+}

generation_config.json

@@ -1,5 +1,4 @@
 {
-  "_from_model_config": true,
   "decoder_start_token_id": 0,
   "eos_token_id": 1,
   "pad_token_id": 0,

runs/Dec14_18-03-50_290501f97bb0/events.out.tfevents.1702580148.290501f97bb0.1579.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89c0a5277f80b80fafb0b0a66c58dac9670b5349e25aab252a779d60b65c91a7
+size 613

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "train_loss": 0.2953249713897705,
+    "train_runtime": 2842.9361,
+    "train_samples": 20000,
+    "train_samples_per_second": 21.105,
+    "train_steps_per_second": 5.276
+}

trainer_state.json

@@ -0,0 +1,210 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 15000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.8333333333333334e-05,
+      "loss": 5.4708,
+      "step": 500
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.666666666666667e-05,
+      "loss": 0.6907,
+      "step": 1000
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.5e-05,
+      "loss": 0.3683,
+      "step": 1500
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.3333333333333334e-05,
+      "loss": 0.2588,
+      "step": 2000
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 0.1798,
+      "step": 2500
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 4e-05,
+      "loss": 0.1087,
+      "step": 3000
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 3.8333333333333334e-05,
+      "loss": 0.1103,
+      "step": 3500
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 3.6666666666666666e-05,
+      "loss": 0.1145,
+      "step": 4000
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 3.5e-05,
+      "loss": 0.109,
+      "step": 4500
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 0.0909,
+      "step": 5000
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 3.1666666666666666e-05,
+      "loss": 0.0674,
+      "step": 5500
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 3e-05,
+      "loss": 0.0778,
+      "step": 6000
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 2.8333333333333335e-05,
+      "loss": 0.0895,
+      "step": 6500
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 2.6666666666666667e-05,
+      "loss": 0.0756,
+      "step": 7000
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 2.5e-05,
+      "loss": 0.0793,
+      "step": 7500
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 2.3333333333333336e-05,
+      "loss": 0.0792,
+      "step": 8000
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 2.1666666666666667e-05,
+      "loss": 0.0778,
+      "step": 8500
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 2e-05,
+      "loss": 0.0742,
+      "step": 9000
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 1.8333333333333333e-05,
+      "loss": 0.0613,
+      "step": 9500
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.0655,
+      "step": 10000
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 1.5e-05,
+      "loss": 0.0641,
+      "step": 10500
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 0.0519,
+      "step": 11000
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 1.1666666666666668e-05,
+      "loss": 0.061,
+      "step": 11500
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 1e-05,
+      "loss": 0.0695,
+      "step": 12000
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 0.0601,
+      "step": 12500
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 0.0648,
+      "step": 13000
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 5e-06,
+      "loss": 0.0642,
+      "step": 13500
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 3.3333333333333333e-06,
+      "loss": 0.057,
+      "step": 14000
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 1.6666666666666667e-06,
+      "loss": 0.0482,
+      "step": 14500
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.0,
+      "loss": 0.0698,
+      "step": 15000
+    },
+    {
+      "epoch": 3.0,
+      "step": 15000,
+      "total_flos": 1432684465152000.0,
+      "train_loss": 0.2953249713897705,
+      "train_runtime": 2842.9361,
+      "train_samples_per_second": 21.105,
+      "train_steps_per_second": 5.276
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 15000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 20000,
+  "total_flos": 1432684465152000.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}