DiwasDiwas/t5-small-MedicoSummarizer

README.md

@@ -15,14 +15,14 @@ should probably proofread and complete it, then remove this comment. -->
 
 # t5-small-MedicoSummarizer
 
-This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on 10,000 PubMed articles.
+This model is a fine-tuned version of [t5-small](https://huggingface.co/t5-small) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.9721
-- Rouge1: 0.3222
-- Rouge2: 0.0728
-- Rougel: 0.1933
-- Rougelsum: 0.1932
-- Gen Len: 122.633
+- Loss: 2.9048
+- Rouge1: 0.3226
+- Rouge2: 0.0771
+- Rougel: 0.1964
+- Rougelsum: 0.1967
+- Gen Len: 123.675
 
 ## Model description
 
@@ -37,8 +37,6 @@ More information needed
 More information needed
 
 ## Training procedure
-The inference engine doesn't do justice to its operation as the inference engine API doesn't work good for trainer checkpoints !
-So, you should rather load it on the pipeline and just try it !
 
 ### Training hyperparameters
 
@@ -49,23 +47,18 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 10
+- num_epochs: 5
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
 |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:-------:|
-| 3.4732        | 1.0   | 625  | 3.1206          | 0.295  | 0.0629 | 0.1743 | 0.1743    | 117.058 |
-| 3.2979        | 2.0   | 1250 | 3.0677          | 0.3031 | 0.0677 | 0.1821 | 0.1821    | 116.531 |
-| 3.2329        | 3.0   | 1875 | 3.0342          | 0.31   | 0.0696 | 0.1862 | 0.1861    | 119.239 |
-| 3.1848        | 4.0   | 2500 | 3.0100          | 0.3122 | 0.0699 | 0.186  | 0.1858    | 119.044 |
-| 3.1689        | 5.0   | 3125 | 2.9970          | 0.3188 | 0.0713 | 0.1908 | 0.1907    | 122.084 |
-| 3.153         | 6.0   | 3750 | 2.9855          | 0.3187 | 0.0703 | 0.1927 | 0.1926    | 122.33  |
-| 3.1564        | 7.0   | 4375 | 2.9801          | 0.32   | 0.0706 | 0.1929 | 0.1928    | 122.479 |
-| 3.136         | 8.0   | 5000 | 2.9747          | 0.3228 | 0.0727 | 0.1936 | 0.1935    | 122.804 |
-| 3.1248        | 9.0   | 5625 | 2.9726          | 0.321  | 0.0721 | 0.1926 | 0.1925    | 122.685 |
-| 3.1309        | 10.0  | 6250 | 2.9721          | 0.3222 | 0.0728 | 0.1933 | 0.1932    | 122.633 |
+| 3.231         | 1.0   | 1875 | 2.9942          | 0.3149 | 0.0721 | 0.1871 | 0.1873    | 119.685 |
+| 3.153         | 2.0   | 3750 | 2.9426          | 0.3204 | 0.0758 | 0.1935 | 0.1936    | 122.484 |
+| 3.1201        | 3.0   | 5625 | 2.9209          | 0.3215 | 0.0753 | 0.195  | 0.1951    | 123.496 |
+| 3.1008        | 4.0   | 7500 | 2.9072          | 0.3232 | 0.0771 | 0.1969 | 0.1971    | 123.414 |
+| 3.0925        | 5.0   | 9375 | 2.9048          | 0.3226 | 0.0771 | 0.1964 | 0.1967    | 123.675 |
 
 
 ### Framework versions

all_results.json

@@ -1,12 +1,12 @@
 {
-    "epoch": 10.0,
-    "eval_gen_len": 122.633,
-    "eval_loss": 2.972090721130371,
-    "eval_rouge1": 0.3222,
-    "eval_rouge2": 0.0728,
-    "eval_rougeL": 0.1933,
-    "eval_rougeLsum": 0.1932,
-    "eval_runtime": 169.7847,
-    "eval_samples_per_second": 5.89,
-    "eval_steps_per_second": 0.371
+    "epoch": 5.0,
+    "eval_gen_len": 123.675,
+    "eval_loss": 2.904787540435791,
+    "eval_rouge1": 0.3226,
+    "eval_rouge2": 0.0771,
+    "eval_rougeL": 0.1964,
+    "eval_rougeLsum": 0.1967,
+    "eval_runtime": 164.4164,
+    "eval_samples_per_second": 6.082,
+    "eval_steps_per_second": 0.383
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:49a8c9343f09310cec45c4402da19adfe439523c11c48479060997b3c3fd3671
+oid sha256:959f21d96c94eb5fa45ea62693f65afd82444e2456585a779905bdc34e7a6192
 size 242041896

runs/Dec13_09-17-55_a0182b6e7bc8/events.out.tfevents.1702459158.a0182b6e7bc8.229.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81cbfed76d9d8992dffdb43405223cff0bf7f704e2a7610ba2ac99efd57e6281
+size 11023

runs/Dec13_09-17-55_a0182b6e7bc8/events.out.tfevents.1702468321.a0182b6e7bc8.229.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:421058a7b4a755f883a320e3f25b2a5c2650c7abefd8b0e4b82423d9cd621677
+size 613

special_tokens_map.json

@@ -101,25 +101,7 @@
     "<extra_id_98>",
     "<extra_id_99>"
   ],
-  "eos_token": {
-    "content": "</s>",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
-    "single_word": false
-  },
-  "pad_token": {
-    "content": "<pad>",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
-    "single_word": false
-  },
-  "unk_token": {
-    "content": "<unk>",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
-    "single_word": false
-  }
+  "eos_token": "</s>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
 }

tokenizer_config.json

@@ -931,12 +931,8 @@
   "clean_up_tokenization_spaces": true,
   "eos_token": "</s>",
   "extra_ids": 100,
-  "max_length": 128,
   "model_max_length": 512,
   "pad_token": "<pad>",
-  "stride": 0,
   "tokenizer_class": "T5Tokenizer",
-  "truncation_side": "right",
-  "truncation_strategy": "longest_first",
   "unk_token": "<unk>"
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f1a30541da037788c1b3cb624e1eccd2d8c54b283a414f9fc1258837dec33647
+oid sha256:459e0ba38d7350474ee4d0d03e220420a9554159f87d4056a357225a48ee2a7a
 size 4856