Hugging Face's logo

morenolq
/
m2l_2022_nlp

Model card Files Community
4

Add models config, data folder

#1
by g8a9 - opened
base: refs/heads/main
←
from: refs/pr/1
Discussion Files changed
+39668
-0
Files changed (10) hide show
  1. .gitattributes +3 -0
  2. data/europarl_test.csv +3 -0
  3. data/europarl_train.csv +3 -0
  4. data/europarl_valid.csv +3 -0
  5. models/europarl/config.json +13 -0
  6. models/europarl/tokenizer.json +0 -0
  7. models/europarl/train_state_437000.pkl +3 -0
  8. models/tatoeba/config.json +13 -0
  9. mt_tokenizer.json → models/tatoeba/tokenizer.json +0 -0
  10. best_mt_train_state_468500.pkl → models/tatoeba/train_state_468500.pkl +0 -0
.gitattributes CHANGED
@@ -30,3 +30,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
30
  *.zst filter=lfs diff=lfs merge=lfs -text
31
  *tfevents* filter=lfs diff=lfs merge=lfs -text
32
  it-en.tsv filter=lfs diff=lfs merge=lfs -text
 
 
 
 
30
  *.zst filter=lfs diff=lfs merge=lfs -text
31
  *tfevents* filter=lfs diff=lfs merge=lfs -text
32
  it-en.tsv filter=lfs diff=lfs merge=lfs -text
33
+ data/europarl_test.csv filter=lfs diff=lfs merge=lfs -text
34
+ data/europarl_train.csv filter=lfs diff=lfs merge=lfs -text
35
+ data/europarl_valid.csv filter=lfs diff=lfs merge=lfs -text
data/europarl_test.csv ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84f64f0128451020ed962d4a4d580849d8da7c8f4ddbf7c00eafd9eab90b7b6c
3
+ size 331420
data/europarl_train.csv ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:646002ef8fe7d578beb8690675234c556cbae20abe3c36de46bfe96626f0b9c0
3
+ size 565391071
data/europarl_valid.csv ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e03c79cde3999b638fc40e972b63ed3ca88c2c816d2bd0aa7640cf34a9f88dec
3
+ size 62842505
models/europarl/config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "batch_size": 64,
3
+ "mask_probability": 0.15,
4
+ "num_layers": 6,
5
+ "num_heads": 8,
6
+ "d_model": 512,
7
+ "d_ff": 1024,
8
+ "p_dropout": 0.1,
9
+ "max_seq_len": 256,
10
+ "vocab_size": 20000,
11
+ "learning_rate": 1e-4,
12
+ "grad_clip_value": 1
13
+ }
models/europarl/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
models/europarl/train_state_437000.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eed8ce60e80f20b7f3bc6586e5bbc2fcb54c531d43584a7197155455194af582
3
+ size 624575923
models/tatoeba/config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "batch_size": 64,
3
+ "mask_probability": 0.15,
4
+ "num_layers": 6,
5
+ "num_heads": 8,
6
+ "d_model": 256,
7
+ "d_ff": 512,
8
+ "p_dropout": 0.1,
9
+ "max_seq_len": 128,
10
+ "vocab_size": 20000,
11
+ "learning_rate": 3e-4,
12
+ "grad_clip_value": 1
13
+ }
mt_tokenizer.json → models/tatoeba/tokenizer.json RENAMED
File without changes
best_mt_train_state_468500.pkl → models/tatoeba/train_state_468500.pkl RENAMED
File without changes