Saving weights and logs of step 10000 - epoch 25
Browse files- README.md +5 -4
- flax_model.msgpack +1 -1
- training_state.bin +2 -2
README.md
CHANGED
@@ -48,7 +48,7 @@ The following hyperparameters were used during training:
|
|
48 |
- num_train_dataset_workers: 32
|
49 |
- num_hosts: 1
|
50 |
- total_num_training_examples: 4,800,000
|
51 |
-
- steps_per_epoch:
|
52 |
- num_beams: None
|
53 |
- weight_decay: 0.01
|
54 |
- adam_beta1: 0.9
|
@@ -60,9 +60,10 @@ The following hyperparameters were used during training:
|
|
60 |
|
61 |
### Training results
|
62 |
|
63 |
-
| step
|
64 |
-
|
65 |
-
| 0
|
|
|
66 |
|
67 |
|
68 |
### Framework versions
|
|
|
48 |
- num_train_dataset_workers: 32
|
49 |
- num_hosts: 1
|
50 |
- total_num_training_examples: 4,800,000
|
51 |
+
- steps_per_epoch: 385
|
52 |
- num_beams: None
|
53 |
- weight_decay: 0.01
|
54 |
- adam_beta1: 0.9
|
|
|
60 |
|
61 |
### Training results
|
62 |
|
63 |
+
| step | validation_loss | train_loss | validation_wer | validation_cer | validation_exact_wer | validation_exact_cer |
|
64 |
+
|:-----:|:---------------:|:----------:|:--------------:|:--------------:|:--------------------:|:--------------------:|
|
65 |
+
| 0 | 4.2254 | 4.7016 | 112.7660 | 59.8700 | 108.1117 | 62.0594 |
|
66 |
+
| 10000 | 0.8352 | 0.3801 | 18.0851 | 5.5037 | 22.2074 | 6.0546 |
|
67 |
|
68 |
|
69 |
### Framework versions
|
flax_model.msgpack
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 3086611993
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:02dba61b8c525620026b89fce9d5d78a2b2a3c6e7d95b9106a2819a643cf6d1a
|
3 |
size 3086611993
|
training_state.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:62d4fe8c25a7574fca42e3121674a1af47929b9b4d5b80f69b652d652aec99e1
|
3 |
+
size 526
|