End of training
Browse files- README.md +31 -11
- pytorch_model.bin +1 -1
README.md
CHANGED
|
@@ -4,9 +4,24 @@ tags:
|
|
| 4 |
- generated_from_trainer
|
| 5 |
datasets:
|
| 6 |
- govreport-summarization
|
|
|
|
|
|
|
| 7 |
model-index:
|
| 8 |
- name: led-large-16384-govreport
|
| 9 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
---
|
| 11 |
|
| 12 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -16,15 +31,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [allenai/led-base-16384](https://huggingface.co/allenai/led-base-16384) on the govreport-summarization dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
-
-
|
| 20 |
-
-
|
| 21 |
-
-
|
| 22 |
-
-
|
| 23 |
-
-
|
| 24 |
-
- eval_runtime: 5561.9865
|
| 25 |
-
- eval_samples_per_second: 0.175
|
| 26 |
-
- eval_steps_per_second: 0.175
|
| 27 |
-
- step: 0
|
| 28 |
|
| 29 |
## Model description
|
| 30 |
|
|
@@ -51,7 +62,16 @@ The following hyperparameters were used during training:
|
|
| 51 |
- total_train_batch_size: 64
|
| 52 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 53 |
- lr_scheduler_type: linear
|
| 54 |
-
- num_epochs:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 55 |
|
| 56 |
### Framework versions
|
| 57 |
|
|
|
|
| 4 |
- generated_from_trainer
|
| 5 |
datasets:
|
| 6 |
- govreport-summarization
|
| 7 |
+
metrics:
|
| 8 |
+
- rouge
|
| 9 |
model-index:
|
| 10 |
- name: led-large-16384-govreport
|
| 11 |
+
results:
|
| 12 |
+
- task:
|
| 13 |
+
name: Sequence-to-sequence Language Modeling
|
| 14 |
+
type: text2text-generation
|
| 15 |
+
dataset:
|
| 16 |
+
name: govreport-summarization
|
| 17 |
+
type: govreport-summarization
|
| 18 |
+
config: document
|
| 19 |
+
split: validation
|
| 20 |
+
args: document
|
| 21 |
+
metrics:
|
| 22 |
+
- name: Rouge1
|
| 23 |
+
type: rouge
|
| 24 |
+
value: 0.5292696934712731
|
| 25 |
---
|
| 26 |
|
| 27 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 31 |
|
| 32 |
This model is a fine-tuned version of [allenai/led-base-16384](https://huggingface.co/allenai/led-base-16384) on the govreport-summarization dataset.
|
| 33 |
It achieves the following results on the evaluation set:
|
| 34 |
+
- Loss: 1.7702
|
| 35 |
+
- Rouge1: 0.5293
|
| 36 |
+
- Rouge2: 0.2192
|
| 37 |
+
- Rougel: 0.2504
|
| 38 |
+
- Rougelsum: 0.2505
|
|
|
|
|
|
|
|
|
|
|
|
|
| 39 |
|
| 40 |
## Model description
|
| 41 |
|
|
|
|
| 62 |
- total_train_batch_size: 64
|
| 63 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 64 |
- lr_scheduler_type: linear
|
| 65 |
+
- num_epochs: 6
|
| 66 |
+
|
| 67 |
+
### Training results
|
| 68 |
+
|
| 69 |
+
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|
| 70 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|
|
| 71 |
+
| 1.8152 | 3.65 | 500 | 1.7956 | 0.5095 | 0.2040 | 0.2382 | 0.2381 |
|
| 72 |
+
| 1.6981 | 3.66 | 1000 | 1.7624 | 0.5194 | 0.2107 | 0.2437 | 0.2437 |
|
| 73 |
+
| 1.7048 | 5.49 | 1500 | 1.7448 | 0.5253 | 0.2149 | 0.2467 | 0.2467 |
|
| 74 |
+
|
| 75 |
|
| 76 |
### Framework versions
|
| 77 |
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 647678513
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:41574ce7cf69b0ae43d5dc42f803155620320fe0e94accc9aee9d2f32ec7303c
|
| 3 |
size 647678513
|