Add Seq2Seq LSTM trained on Multi30K
Browse files
README.md
CHANGED
|
@@ -7,7 +7,7 @@ language:
|
|
| 7 |
library_name: pytorch
|
| 8 |
pipeline_tag: translation
|
| 9 |
tags:
|
| 10 |
-
-
|
| 11 |
- lstm
|
| 12 |
- seq2seq
|
| 13 |
- machine-translation
|
|
@@ -16,7 +16,7 @@ datasets:
|
|
| 16 |
- bentrevett/multi30k
|
| 17 |
---
|
| 18 |
|
| 19 |
-
# Seq2Seq
|
| 20 |
|
| 21 |
A PyTorch Seq2Seq Encoder-Decoder LSTM trained from scratch on the Multi30k dataset.
|
| 22 |
|
|
@@ -49,10 +49,10 @@ A PyTorch Seq2Seq Encoder-Decoder LSTM trained from scratch on the Multi30k data
|
|
| 49 |
|
| 50 |
| Metric | Value |
|
| 51 |
|------|------:|
|
| 52 |
-
| Train Loss |
|
| 53 |
-
| Validation Loss | 5.
|
| 54 |
-
| Test Loss |
|
| 55 |
-
| BLEU |
|
| 56 |
|
| 57 |

|
| 58 |
|
|
|
|
| 7 |
library_name: pytorch
|
| 8 |
pipeline_tag: translation
|
| 9 |
tags:
|
| 10 |
+
- pytorchf
|
| 11 |
- lstm
|
| 12 |
- seq2seq
|
| 13 |
- machine-translation
|
|
|
|
| 16 |
- bentrevett/multi30k
|
| 17 |
---
|
| 18 |
|
| 19 |
+
# Seq2Seq LSTM Translator (English -> German)
|
| 20 |
|
| 21 |
A PyTorch Seq2Seq Encoder-Decoder LSTM trained from scratch on the Multi30k dataset.
|
| 22 |
|
|
|
|
| 49 |
|
| 50 |
| Metric | Value |
|
| 51 |
|------|------:|
|
| 52 |
+
| Train Loss | 0.62 |
|
| 53 |
+
| Validation Loss | 5.55 |
|
| 54 |
+
| Test Loss | 3.79 |
|
| 55 |
+
| BLEU | 14.70 |
|
| 56 |
|
| 57 |

|
| 58 |
|