End of training

README.md

@@ -1,11 +1,24 @@
 ---
+base_model: /exports/eddie/scratch/s1970716/models/summarization/longt5_xl_summ_screen_bp_only/checkpoint-210
 tags:
 - generated_from_trainer
+datasets:
+- learn3r/summ_screen_fd_bp
 metrics:
 - rouge
 model-index:
 - name: longt5_xl_summ_screen_bp_only_30
-  results: []
+  results:
+  - task:
+      name: Summarization
+      type: summarization
+    dataset:
+      name: learn3r/summ_screen_fd_bp
+      type: learn3r/summ_screen_fd_bp
+    metrics:
+    - name: Rouge1
+      type: rouge
+      value: 40.4943
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -13,14 +26,14 @@ should probably proofread and complete it, then remove this comment. -->
 
 # longt5_xl_summ_screen_bp_only_30
 
-This model was trained from scratch on an unknown dataset.
+This model is a fine-tuned version of [/exports/eddie/scratch/s1970716/models/summarization/longt5_xl_summ_screen_bp_only/checkpoint-210](https://huggingface.co//exports/eddie/scratch/s1970716/models/summarization/longt5_xl_summ_screen_bp_only/checkpoint-210) on the learn3r/summ_screen_fd_bp dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.3973
-- Rouge1: 31.9833
-- Rouge2: 14.0141
-- Rougel: 22.6823
-- Rougelsum: 30.6424
-- Gen Len: 484.3964
+- Loss: 2.2397
+- Rouge1: 40.4943
+- Rouge2: 16.4695
+- Rougel: 28.0964
+- Rougelsum: 38.3693
+- Gen Len: 246.3491
 
 ## Model description
 

all_results.json

@@ -0,0 +1,18 @@
+{
+    "epoch": 13.18,
+    "eval_gen_len": 246.3491124260355,
+    "eval_loss": 2.2397100925445557,
+    "eval_rouge1": 40.4943,
+    "eval_rouge2": 16.4695,
+    "eval_rougeL": 28.0964,
+    "eval_rougeLsum": 38.3693,
+    "eval_runtime": 467.1505,
+    "eval_samples": 338,
+    "eval_samples_per_second": 0.724,
+    "eval_steps_per_second": 0.024,
+    "train_loss": 0.16495737830797833,
+    "train_runtime": 12728.0808,
+    "train_samples": 3673,
+    "train_samples_per_second": 4.329,
+    "train_steps_per_second": 0.004
+}

eval_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 13.18,
+    "eval_gen_len": 246.3491124260355,
+    "eval_loss": 2.2397100925445557,
+    "eval_rouge1": 40.4943,
+    "eval_rouge2": 16.4695,
+    "eval_rougeL": 28.0964,
+    "eval_rougeLsum": 38.3693,
+    "eval_runtime": 467.1505,
+    "eval_samples": 338,
+    "eval_samples_per_second": 0.724,
+    "eval_steps_per_second": 0.024
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 13.18,
+    "train_loss": 0.16495737830797833,
+    "train_runtime": 12728.0808,
+    "train_samples": 3673,
+    "train_samples_per_second": 4.329,
+    "train_steps_per_second": 0.004
+}

trainer_state.json

@@ -0,0 +1,303 @@
+{
+  "best_metric": 2.237640380859375,
+  "best_model_checkpoint": "/exports/eddie/scratch/s1970716/models/summarization/longt5_xl_summ_screen_bp_only_30/checkpoint-14",
+  "epoch": 13.182608695652174,
+  "eval_steps": 500,
+  "global_step": 45,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.0005,
+      "loss": 0.3076,
+      "step": 2
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.0005,
+      "loss": 0.3217,
+      "step": 4
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.0005,
+      "loss": 0.3519,
+      "step": 6
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.0005,
+      "loss": 0.3449,
+      "step": 8
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.0005,
+      "loss": 0.3197,
+      "step": 10
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.0005,
+      "loss": 0.3342,
+      "step": 12
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 0.0005,
+      "loss": 0.324,
+      "step": 14
+    },
+    {
+      "epoch": 0.97,
+      "eval_gen_len": 246.7396449704142,
+      "eval_loss": 2.237640380859375,
+      "eval_rouge1": 40.4388,
+      "eval_rouge2": 16.4662,
+      "eval_rougeL": 28.0771,
+      "eval_rougeLsum": 38.3405,
+      "eval_runtime": 1741.7742,
+      "eval_samples_per_second": 0.194,
+      "eval_steps_per_second": 0.025,
+      "step": 14
+    },
+    {
+      "epoch": 4.83,
+      "eval_gen_len": 503.56508875739644,
+      "eval_loss": 2.572709798812866,
+      "eval_rouge1": 30.0123,
+      "eval_rouge2": 12.3701,
+      "eval_rougeL": 21.2834,
+      "eval_rougeLsum": 28.891,
+      "eval_runtime": 472.1582,
+      "eval_samples_per_second": 0.716,
+      "eval_steps_per_second": 0.023,
+      "step": 15
+    },
+    {
+      "epoch": 5.11,
+      "learning_rate": 0.0005,
+      "loss": 0.278,
+      "step": 16
+    },
+    {
+      "epoch": 5.67,
+      "learning_rate": 0.0005,
+      "loss": 0.3036,
+      "step": 18
+    },
+    {
+      "epoch": 5.95,
+      "eval_gen_len": 506.9526627218935,
+      "eval_loss": 2.2658941745758057,
+      "eval_rouge1": 27.8421,
+      "eval_rouge2": 11.1942,
+      "eval_rougeL": 20.4713,
+      "eval_rougeLsum": 26.6097,
+      "eval_runtime": 472.9998,
+      "eval_samples_per_second": 0.715,
+      "eval_steps_per_second": 0.023,
+      "step": 19
+    },
+    {
+      "epoch": 6.23,
+      "learning_rate": 0.0005,
+      "loss": 0.2937,
+      "step": 20
+    },
+    {
+      "epoch": 6.78,
+      "learning_rate": 0.0005,
+      "loss": 0.2941,
+      "step": 22
+    },
+    {
+      "epoch": 6.78,
+      "eval_gen_len": 284.2307692307692,
+      "eval_loss": 2.2636122703552246,
+      "eval_rouge1": 40.8304,
+      "eval_rouge2": 17.3615,
+      "eval_rougeL": 28.0971,
+      "eval_rougeLsum": 39.0943,
+      "eval_runtime": 468.1667,
+      "eval_samples_per_second": 0.722,
+      "eval_steps_per_second": 0.023,
+      "step": 22
+    },
+    {
+      "epoch": 7.34,
+      "learning_rate": 0.0005,
+      "loss": 0.2508,
+      "step": 24
+    },
+    {
+      "epoch": 7.9,
+      "learning_rate": 0.0005,
+      "loss": 0.2642,
+      "step": 26
+    },
+    {
+      "epoch": 7.9,
+      "eval_gen_len": 341.25147928994085,
+      "eval_loss": 2.286357879638672,
+      "eval_rouge1": 38.3377,
+      "eval_rouge2": 15.8119,
+      "eval_rougeL": 26.4838,
+      "eval_rougeLsum": 36.5174,
+      "eval_runtime": 469.1516,
+      "eval_samples_per_second": 0.72,
+      "eval_steps_per_second": 0.023,
+      "step": 26
+    },
+    {
+      "epoch": 8.45,
+      "learning_rate": 0.0005,
+      "loss": 0.2604,
+      "step": 28
+    },
+    {
+      "epoch": 8.73,
+      "eval_gen_len": 435.26331360946745,
+      "eval_loss": 2.455064535140991,
+      "eval_rouge1": 33.2021,
+      "eval_rouge2": 13.6577,
+      "eval_rougeL": 23.3288,
+      "eval_rougeLsum": 31.8326,
+      "eval_runtime": 471.959,
+      "eval_samples_per_second": 0.716,
+      "eval_steps_per_second": 0.023,
+      "step": 29
+    },
+    {
+      "epoch": 9.01,
+      "learning_rate": 0.0005,
+      "loss": 0.2422,
+      "step": 30
+    },
+    {
+      "epoch": 9.57,
+      "learning_rate": 0.0005,
+      "loss": 0.2237,
+      "step": 32
+    },
+    {
+      "epoch": 9.84,
+      "eval_gen_len": 234.61242603550295,
+      "eval_loss": 2.615253210067749,
+      "eval_rouge1": 40.3297,
+      "eval_rouge2": 15.3786,
+      "eval_rougeL": 28.1208,
+      "eval_rougeLsum": 38.2426,
+      "eval_runtime": 470.1581,
+      "eval_samples_per_second": 0.719,
+      "eval_steps_per_second": 0.023,
+      "step": 33
+    },
+    {
+      "epoch": 10.12,
+      "learning_rate": 0.0005,
+      "loss": 0.2184,
+      "step": 34
+    },
+    {
+      "epoch": 10.68,
+      "learning_rate": 0.0005,
+      "loss": 0.1904,
+      "step": 36
+    },
+    {
+      "epoch": 10.96,
+      "eval_gen_len": 174.57396449704143,
+      "eval_loss": 2.666517734527588,
+      "eval_rouge1": 39.6006,
+      "eval_rouge2": 14.9586,
+      "eval_rougeL": 27.2453,
+      "eval_rougeLsum": 37.6744,
+      "eval_runtime": 450.9467,
+      "eval_samples_per_second": 0.75,
+      "eval_steps_per_second": 0.024,
+      "step": 37
+    },
+    {
+      "epoch": 11.23,
+      "learning_rate": 0.0005,
+      "loss": 0.2076,
+      "step": 38
+    },
+    {
+      "epoch": 11.79,
+      "learning_rate": 0.0005,
+      "loss": 0.2247,
+      "step": 40
+    },
+    {
+      "epoch": 11.79,
+      "eval_gen_len": 500.58284023668637,
+      "eval_loss": 2.722362518310547,
+      "eval_rouge1": 30.5957,
+      "eval_rouge2": 13.3496,
+      "eval_rougeL": 21.9712,
+      "eval_rougeLsum": 29.22,
+      "eval_runtime": 477.6586,
+      "eval_samples_per_second": 0.708,
+      "eval_steps_per_second": 0.023,
+      "step": 40
+    },
+    {
+      "epoch": 12.35,
+      "learning_rate": 0.0005,
+      "loss": 0.1866,
+      "step": 42
+    },
+    {
+      "epoch": 12.9,
+      "learning_rate": 0.0005,
+      "loss": 0.182,
+      "step": 44
+    },
+    {
+      "epoch": 12.9,
+      "eval_gen_len": 259.6568047337278,
+      "eval_loss": 3.271503448486328,
+      "eval_rouge1": 41.6828,
+      "eval_rouge2": 17.0818,
+      "eval_rougeL": 28.087,
+      "eval_rougeLsum": 39.5947,
+      "eval_runtime": 468.2637,
+      "eval_samples_per_second": 0.722,
+      "eval_steps_per_second": 0.023,
+      "step": 44
+    },
+    {
+      "epoch": 13.18,
+      "eval_gen_len": 484.396449704142,
+      "eval_loss": 2.397312641143799,
+      "eval_rouge1": 31.9833,
+      "eval_rouge2": 14.0141,
+      "eval_rougeL": 22.6823,
+      "eval_rougeLsum": 30.6424,
+      "eval_runtime": 471.2439,
+      "eval_samples_per_second": 0.717,
+      "eval_steps_per_second": 0.023,
+      "step": 45
+    },
+    {
+      "epoch": 13.18,
+      "step": 45,
+      "total_flos": 2.4232217682745754e+18,
+      "train_loss": 0.16495737830797833,
+      "train_runtime": 12728.0808,
+      "train_samples_per_second": 4.329,
+      "train_steps_per_second": 0.004
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 45,
+  "num_train_epochs": 15,
+  "save_steps": 500,
+  "total_flos": 2.4232217682745754e+18,
+  "trial_name": null,
+  "trial_params": null
+}