End of training

Browse files

Files changed (5) hide show

README.md +2 -2
all_results.json +16 -0
eval_results.json +10 -0
train_results.json +9 -0
trainer_state.json +197 -0

README.md CHANGED Viewed

@@ -6,14 +6,14 @@ tags:
 metrics:
 - bleu
 model-index:
-- name: adversarial_qa_dbert_based_on
   results: []
 ---
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
-# adversarial_qa_dbert_based_on
 This model is a fine-tuned version of [google-t5/t5-large](https://huggingface.co/google-t5/t5-large) on an unknown dataset.
 It achieves the following results on the evaluation set:

 metrics:
 - bleu
 model-index:
+- name: fft-t5-large/adversarial_qa_dbert_based_on
   results: []
 ---
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
+# fft-t5-large/adversarial_qa_dbert_based_on
 This model is a fine-tuned version of [google-t5/t5-large](https://huggingface.co/google-t5/t5-large) on an unknown dataset.
 It achieves the following results on the evaluation set:

all_results.json ADDED Viewed

	@@ -0,0 +1,16 @@

+{
+    "epoch": 5.0,
+    "eval_bleu": 0.30828732530539343,
+    "eval_exact_match": 0.3466796875,
+    "eval_loss": 1.138079285621643,
+    "eval_runtime": 20.4992,
+    "eval_samples": 1000,
+    "eval_samples_per_second": 48.782,
+    "eval_steps_per_second": 0.781,
+    "total_flos": 2586783250382848.0,
+    "train_loss": 0.3393354445222824,
+    "train_runtime": 835.7154,
+    "train_samples": 500,
+    "train_samples_per_second": 2.991,
+    "train_steps_per_second": 0.377
+}

eval_results.json ADDED Viewed

	@@ -0,0 +1,10 @@

+{
+    "epoch": 5.0,
+    "eval_bleu": 0.30828732530539343,
+    "eval_exact_match": 0.3466796875,
+    "eval_loss": 1.138079285621643,
+    "eval_runtime": 20.4992,
+    "eval_samples": 1000,
+    "eval_samples_per_second": 48.782,
+    "eval_steps_per_second": 0.781
+}

train_results.json ADDED Viewed

	@@ -0,0 +1,9 @@

+{
+    "epoch": 5.0,
+    "total_flos": 2586783250382848.0,
+    "train_loss": 0.3393354445222824,
+    "train_runtime": 835.7154,
+    "train_samples": 500,
+    "train_samples_per_second": 2.991,
+    "train_steps_per_second": 0.377
+}

trainer_state.json ADDED Viewed

	@@ -0,0 +1,197 @@

+{
+  "best_metric": 0.3466796875,
+  "best_model_checkpoint": "/data/leeye/checkpoints_fft_t5_large/adversarial_qa_dbert_based_on/checkpoint-315",
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 315,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.31746031746031744,
+      "grad_norm": 18.570829391479492,
+      "learning_rate": 0.00018730158730158731,
+      "loss": 1.2416,
+      "step": 20
+    },
+    {
+      "epoch": 0.6349206349206349,
+      "grad_norm": 15.798161506652832,
+      "learning_rate": 0.00017460317460317462,
+      "loss": 1.0892,
+      "step": 40
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 7.127645015716553,
+      "learning_rate": 0.00016190476190476192,
+      "loss": 1.0162,
+      "step": 60
+    },
+    {
+      "epoch": 1.0,
+      "eval_bleu": 0.2748783583138883,
+      "eval_exact_match": 0.275390625,
+      "eval_loss": 0.7607467174530029,
+      "eval_runtime": 22.1309,
+      "eval_samples_per_second": 45.186,
+      "eval_steps_per_second": 0.723,
+      "step": 63
+    },
+    {
+      "epoch": 1.2698412698412698,
+      "grad_norm": 6.7987775802612305,
+      "learning_rate": 0.00014920634920634923,
+      "loss": 0.4064,
+      "step": 80
+    },
+    {
+      "epoch": 1.5873015873015874,
+      "grad_norm": 10.752936363220215,
+      "learning_rate": 0.0001365079365079365,
+      "loss": 0.3513,
+      "step": 100
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 7.6475348472595215,
+      "learning_rate": 0.0001238095238095238,
+      "loss": 0.3929,
+      "step": 120
+    },
+    {
+      "epoch": 2.0,
+      "eval_bleu": 0.24117803737423354,
+      "eval_exact_match": 0.2958984375,
+      "eval_loss": 0.794335126876831,
+      "eval_runtime": 24.0083,
+      "eval_samples_per_second": 41.652,
+      "eval_steps_per_second": 0.666,
+      "step": 126
+    },
+    {
+      "epoch": 2.2222222222222223,
+      "grad_norm": 2.862144947052002,
+      "learning_rate": 0.00011111111111111112,
+      "loss": 0.2161,
+      "step": 140
+    },
+    {
+      "epoch": 2.5396825396825395,
+      "grad_norm": 5.910730838775635,
+      "learning_rate": 9.841269841269841e-05,
+      "loss": 0.138,
+      "step": 160
+    },
+    {
+      "epoch": 2.857142857142857,
+      "grad_norm": 1.7475248575210571,
+      "learning_rate": 8.571428571428571e-05,
+      "loss": 0.1542,
+      "step": 180
+    },
+    {
+      "epoch": 3.0,
+      "eval_bleu": 0.27200128946842234,
+      "eval_exact_match": 0.3017578125,
+      "eval_loss": 1.0052680969238281,
+      "eval_runtime": 21.8202,
+      "eval_samples_per_second": 45.829,
+      "eval_steps_per_second": 0.733,
+      "step": 189
+    },
+    {
+      "epoch": 3.1746031746031744,
+      "grad_norm": 2.242680549621582,
+      "learning_rate": 7.301587301587302e-05,
+      "loss": 0.095,
+      "step": 200
+    },
+    {
+      "epoch": 3.492063492063492,
+      "grad_norm": 1.32904851436615,
+      "learning_rate": 6.0317460317460316e-05,
+      "loss": 0.0614,
+      "step": 220
+    },
+    {
+      "epoch": 3.8095238095238093,
+      "grad_norm": 37.23604202270508,
+      "learning_rate": 4.761904761904762e-05,
+      "loss": 0.0544,
+      "step": 240
+    },
+    {
+      "epoch": 4.0,
+      "eval_bleu": 0.31847923264557687,
+      "eval_exact_match": 0.345703125,
+      "eval_loss": 1.1005158424377441,
+      "eval_runtime": 18.54,
+      "eval_samples_per_second": 53.937,
+      "eval_steps_per_second": 0.863,
+      "step": 252
+    },
+    {
+      "epoch": 4.1269841269841265,
+      "grad_norm": 0.15852811932563782,
+      "learning_rate": 3.492063492063492e-05,
+      "loss": 0.0319,
+      "step": 260
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 0.4201483130455017,
+      "learning_rate": 2.2222222222222223e-05,
+      "loss": 0.0249,
+      "step": 280
+    },
+    {
+      "epoch": 4.761904761904762,
+      "grad_norm": 0.46498769521713257,
+      "learning_rate": 9.523809523809523e-06,
+      "loss": 0.0239,
+      "step": 300
+    },
+    {
+      "epoch": 5.0,
+      "eval_bleu": 0.30828732530539343,
+      "eval_exact_match": 0.3466796875,
+      "eval_loss": 1.138079285621643,
+      "eval_runtime": 20.547,
+      "eval_samples_per_second": 48.669,
+      "eval_steps_per_second": 0.779,
+      "step": 315
+    },
+    {
+      "epoch": 5.0,
+      "step": 315,
+      "total_flos": 2586783250382848.0,
+      "train_loss": 0.3393354445222824,
+      "train_runtime": 835.7154,
+      "train_samples_per_second": 2.991,
+      "train_steps_per_second": 0.377
+    }
+  ],
+  "logging_steps": 20,
+  "max_steps": 315,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2586783250382848.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}