End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- id
 license: apache-2.0
 base_model: LazarusNLP/IndoNanoT5-base
 tags:
@@ -18,11 +20,11 @@ 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: 0.4741
-- Rouge1: 72.7878
-- Rouge2: 66.023
-- Rougel: 70.0879
-- Rougelsum: 71.9606
-- Gen Len: 102.3829
+- Rouge1: 73.4959
+- Rouge2: 66.8131
+- Rougel: 70.7936
+- Rougelsum: 72.6487
+- Gen Len: 102.2932
 
 ## Model description
 

all_results.json

@@ -0,0 +1,29 @@
+{
+    "epoch": 5.0,
+    "eval_gen_len": 102.29317269076306,
+    "eval_loss": 0.4740569591522217,
+    "eval_rouge1": 73.4959,
+    "eval_rouge2": 66.8131,
+    "eval_rougeL": 70.7936,
+    "eval_rougeLsum": 72.6487,
+    "eval_runtime": 984.2529,
+    "eval_samples": 747,
+    "eval_samples_per_second": 0.759,
+    "eval_steps_per_second": 0.024,
+    "predict_gen_len": 102.97607019409732,
+    "predict_loss": 0.49634960293769836,
+    "predict_rouge1": 72.4383,
+    "predict_rouge2": 65.5956,
+    "predict_rougeL": 69.5276,
+    "predict_rougeLsum": 71.6049,
+    "predict_runtime": 4935.3593,
+    "predict_samples": 3761,
+    "predict_samples_per_second": 0.762,
+    "predict_steps_per_second": 0.024,
+    "total_flos": 4.90376103985152e+16,
+    "train_loss": 0.6289605795000701,
+    "train_runtime": 4825.6683,
+    "train_samples": 14266,
+    "train_samples_per_second": 14.781,
+    "train_steps_per_second": 0.924
+}

eval_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 5.0,
+    "eval_gen_len": 102.29317269076306,
+    "eval_loss": 0.4740569591522217,
+    "eval_rouge1": 73.4959,
+    "eval_rouge2": 66.8131,
+    "eval_rougeL": 70.7936,
+    "eval_rougeLsum": 72.6487,
+    "eval_runtime": 984.2529,
+    "eval_samples": 747,
+    "eval_samples_per_second": 0.759,
+    "eval_steps_per_second": 0.024
+}

predict_results.json

@@ -0,0 +1,12 @@
+{
+    "predict_gen_len": 102.97607019409732,
+    "predict_loss": 0.49634960293769836,
+    "predict_rouge1": 72.4383,
+    "predict_rouge2": 65.5956,
+    "predict_rougeL": 69.5276,
+    "predict_rougeLsum": 71.6049,
+    "predict_runtime": 4935.3593,
+    "predict_samples": 3761,
+    "predict_samples_per_second": 0.762,
+    "predict_steps_per_second": 0.024
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 5.0,
+    "total_flos": 4.90376103985152e+16,
+    "train_loss": 0.6289605795000701,
+    "train_runtime": 4825.6683,
+    "train_samples": 14266,
+    "train_samples_per_second": 14.781,
+    "train_steps_per_second": 0.924
+}

trainer_state.json

@@ -0,0 +1,130 @@
+{
+  "best_metric": 72.7878,
+  "best_model_checkpoint": "bin/indosum-lora-4/checkpoint-4460",
+  "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": 0.6500911712646484,
+      "learning_rate": 0.0008,
+      "loss": 0.8265,
+      "step": 892
+    },
+    {
+      "epoch": 1.0,
+      "eval_gen_len": 94.2182061579652,
+      "eval_loss": 0.517518937587738,
+      "eval_rouge1": 68.9705,
+      "eval_rouge2": 61.6703,
+      "eval_rougeL": 66.1048,
+      "eval_rougeLsum": 68.0943,
+      "eval_runtime": 217.0807,
+      "eval_samples_per_second": 3.441,
+      "eval_steps_per_second": 0.111,
+      "step": 892
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.749599039554596,
+      "learning_rate": 0.0006,
+      "loss": 0.6279,
+      "step": 1784
+    },
+    {
+      "epoch": 2.0,
+      "eval_gen_len": 99.76840696117804,
+      "eval_loss": 0.4892139136791229,
+      "eval_rouge1": 71.4829,
+      "eval_rouge2": 64.4236,
+      "eval_rougeL": 68.7077,
+      "eval_rougeLsum": 70.6264,
+      "eval_runtime": 216.1862,
+      "eval_samples_per_second": 3.455,
+      "eval_steps_per_second": 0.111,
+      "step": 1784
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.7917335629463196,
+      "learning_rate": 0.0004,
+      "loss": 0.5857,
+      "step": 2676
+    },
+    {
+      "epoch": 3.0,
+      "eval_gen_len": 99.37751004016064,
+      "eval_loss": 0.4836407005786896,
+      "eval_rouge1": 72.3238,
+      "eval_rouge2": 65.462,
+      "eval_rougeL": 69.6316,
+      "eval_rougeLsum": 71.4946,
+      "eval_runtime": 218.1617,
+      "eval_samples_per_second": 3.424,
+      "eval_steps_per_second": 0.11,
+      "step": 2676
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.5692319273948669,
+      "learning_rate": 0.0002,
+      "loss": 0.561,
+      "step": 3568
+    },
+    {
+      "epoch": 4.0,
+      "eval_gen_len": 100.25167336010709,
+      "eval_loss": 0.4793739318847656,
+      "eval_rouge1": 72.2216,
+      "eval_rouge2": 65.3919,
+      "eval_rougeL": 69.5752,
+      "eval_rougeLsum": 71.3864,
+      "eval_runtime": 217.306,
+      "eval_samples_per_second": 3.438,
+      "eval_steps_per_second": 0.11,
+      "step": 3568
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.6150219440460205,
+      "learning_rate": 0.0,
+      "loss": 0.5438,
+      "step": 4460
+    },
+    {
+      "epoch": 5.0,
+      "eval_gen_len": 102.38286479250334,
+      "eval_loss": 0.4740569591522217,
+      "eval_rouge1": 72.7878,
+      "eval_rouge2": 66.023,
+      "eval_rougeL": 70.0879,
+      "eval_rougeLsum": 71.9606,
+      "eval_runtime": 220.2537,
+      "eval_samples_per_second": 3.392,
+      "eval_steps_per_second": 0.109,
+      "step": 4460
+    },
+    {
+      "epoch": 5.0,
+      "step": 4460,
+      "total_flos": 4.90376103985152e+16,
+      "train_loss": 0.6289605795000701,
+      "train_runtime": 4825.6683,
+      "train_samples_per_second": 14.781,
+      "train_steps_per_second": 0.924
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 4460,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "total_flos": 4.90376103985152e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}