End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- id
 license: apache-2.0
 base_model: LazarusNLP/IndoNanoT5-base
 tags:
@@ -17,12 +19,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [LazarusNLP/IndoNanoT5-base](https://huggingface.co/LazarusNLP/IndoNanoT5-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6917
-- Rouge1: 70.7161
-- Rouge2: 63.2661
-- Rougel: 67.7431
-- Rougelsum: 69.8153
-- Gen Len: 98.4471
+- Loss: 0.6288
+- Rouge1: 72.0825
+- Rouge2: 64.9891
+- Rougel: 69.2273
+- Rougelsum: 71.1983
+- Gen Len: 98.2637
 
 ## Model description
 

all_results.json

@@ -0,0 +1,29 @@
+{
+    "epoch": 5.0,
+    "eval_gen_len": 98.26372155287818,
+    "eval_loss": 0.6288004517555237,
+    "eval_rouge1": 72.0825,
+    "eval_rouge2": 64.9891,
+    "eval_rougeL": 69.2273,
+    "eval_rougeLsum": 71.1983,
+    "eval_runtime": 869.4118,
+    "eval_samples": 747,
+    "eval_samples_per_second": 0.859,
+    "eval_steps_per_second": 0.028,
+    "predict_gen_len": 98.38314278117522,
+    "predict_loss": 0.6574633121490479,
+    "predict_rouge1": 71.4346,
+    "predict_rouge2": 64.3132,
+    "predict_rougeL": 68.3766,
+    "predict_rougeLsum": 70.5567,
+    "predict_runtime": 4360.7047,
+    "predict_samples": 3761,
+    "predict_samples_per_second": 0.862,
+    "predict_steps_per_second": 0.027,
+    "total_flos": 4.884374195011584e+16,
+    "train_loss": 0.5840840121555756,
+    "train_runtime": 5026.06,
+    "train_samples": 14266,
+    "train_samples_per_second": 14.192,
+    "train_steps_per_second": 0.887
+}

eval_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 5.0,
+    "eval_gen_len": 98.26372155287818,
+    "eval_loss": 0.6288004517555237,
+    "eval_rouge1": 72.0825,
+    "eval_rouge2": 64.9891,
+    "eval_rougeL": 69.2273,
+    "eval_rougeLsum": 71.1983,
+    "eval_runtime": 869.4118,
+    "eval_samples": 747,
+    "eval_samples_per_second": 0.859,
+    "eval_steps_per_second": 0.028
+}

predict_results.json

@@ -0,0 +1,12 @@
+{
+    "predict_gen_len": 98.38314278117522,
+    "predict_loss": 0.6574633121490479,
+    "predict_rouge1": 71.4346,
+    "predict_rouge2": 64.3132,
+    "predict_rougeL": 68.3766,
+    "predict_rougeLsum": 70.5567,
+    "predict_runtime": 4360.7047,
+    "predict_samples": 3761,
+    "predict_samples_per_second": 0.862,
+    "predict_steps_per_second": 0.027
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 5.0,
+    "total_flos": 4.884374195011584e+16,
+    "train_loss": 0.5840840121555756,
+    "train_runtime": 5026.06,
+    "train_samples": 14266,
+    "train_samples_per_second": 14.192,
+    "train_steps_per_second": 0.887
+}

trainer_state.json

@@ -0,0 +1,130 @@
+{
+  "best_metric": 71.5505,
+  "best_model_checkpoint": "bin/indosum-base-4/checkpoint-3568",
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 4460,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.2654496431350708,
+      "learning_rate": 0.0008,
+      "loss": 1.1965,
+      "step": 892
+    },
+    {
+      "epoch": 1.0,
+      "eval_gen_len": 101.15930388219545,
+      "eval_loss": 0.7376554608345032,
+      "eval_rouge1": 66.7755,
+      "eval_rouge2": 58.7089,
+      "eval_rougeL": 63.5735,
+      "eval_rougeLsum": 65.9275,
+      "eval_runtime": 191.1961,
+      "eval_samples_per_second": 3.907,
+      "eval_steps_per_second": 0.126,
+      "step": 892
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.673693835735321,
+      "learning_rate": 0.0006,
+      "loss": 0.691,
+      "step": 1784
+    },
+    {
+      "epoch": 2.0,
+      "eval_gen_len": 100.570281124498,
+      "eval_loss": 0.6472041606903076,
+      "eval_rouge1": 70.9085,
+      "eval_rouge2": 63.6467,
+      "eval_rougeL": 67.9945,
+      "eval_rougeLsum": 70.0311,
+      "eval_runtime": 192.2757,
+      "eval_samples_per_second": 3.885,
+      "eval_steps_per_second": 0.125,
+      "step": 1784
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.7003195285797119,
+      "learning_rate": 0.0004,
+      "loss": 0.4907,
+      "step": 2676
+    },
+    {
+      "epoch": 3.0,
+      "eval_gen_len": 96.95448460508702,
+      "eval_loss": 0.6097700595855713,
+      "eval_rouge1": 70.4573,
+      "eval_rouge2": 63.2612,
+      "eval_rougeL": 67.6711,
+      "eval_rougeLsum": 69.6221,
+      "eval_runtime": 191.6275,
+      "eval_samples_per_second": 3.898,
+      "eval_steps_per_second": 0.125,
+      "step": 2676
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.5256069898605347,
+      "learning_rate": 0.0002,
+      "loss": 0.3404,
+      "step": 3568
+    },
+    {
+      "epoch": 4.0,
+      "eval_gen_len": 97.57429718875503,
+      "eval_loss": 0.6288004517555237,
+      "eval_rouge1": 71.5505,
+      "eval_rouge2": 64.3094,
+      "eval_rougeL": 68.674,
+      "eval_rougeLsum": 70.6452,
+      "eval_runtime": 191.9313,
+      "eval_samples_per_second": 3.892,
+      "eval_steps_per_second": 0.125,
+      "step": 3568
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.6674110293388367,
+      "learning_rate": 0.0,
+      "loss": 0.2018,
+      "step": 4460
+    },
+    {
+      "epoch": 5.0,
+      "eval_gen_len": 98.44712182061579,
+      "eval_loss": 0.6917137503623962,
+      "eval_rouge1": 70.7161,
+      "eval_rouge2": 63.2661,
+      "eval_rougeL": 67.7431,
+      "eval_rougeLsum": 69.8153,
+      "eval_runtime": 193.8117,
+      "eval_samples_per_second": 3.854,
+      "eval_steps_per_second": 0.124,
+      "step": 4460
+    },
+    {
+      "epoch": 5.0,
+      "step": 4460,
+      "total_flos": 4.884374195011584e+16,
+      "train_loss": 0.5840840121555756,
+      "train_runtime": 5026.06,
+      "train_samples_per_second": 14.192,
+      "train_steps_per_second": 0.887
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 4460,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "total_flos": 4.884374195011584e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}