Training in progress, epoch 1
Browse files- model.safetensors +1 -1
- run-0/checkpoint-156/tokenizer.json +2 -16
- run-0/checkpoint-156/trainer_state.json +14 -14
- run-0/checkpoint-234/tokenizer.json +2 -16
- run-0/checkpoint-234/trainer_state.json +19 -19
- run-0/checkpoint-78/model.safetensors +1 -1
- run-0/checkpoint-78/optimizer.pt +1 -1
- run-0/checkpoint-78/scheduler.pt +1 -1
- run-0/checkpoint-78/tokenizer.json +2 -16
- run-0/checkpoint-78/trainer_state.json +9 -9
- run-0/checkpoint-78/training_args.bin +1 -1
- tokenizer.json +2 -16
- training_args.bin +1 -1
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 437958648
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:638ab00ccaa37bc4e7a65c38d74ec0becba8f345692e1517641e1b426eb57268
|
3 |
size 437958648
|
run-0/checkpoint-156/tokenizer.json
CHANGED
@@ -1,21 +1,7 @@
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
-
"truncation":
|
4 |
-
|
5 |
-
"max_length": 128,
|
6 |
-
"strategy": "LongestFirst",
|
7 |
-
"stride": 0
|
8 |
-
},
|
9 |
-
"padding": {
|
10 |
-
"strategy": {
|
11 |
-
"Fixed": 128
|
12 |
-
},
|
13 |
-
"direction": "Right",
|
14 |
-
"pad_to_multiple_of": null,
|
15 |
-
"pad_id": 0,
|
16 |
-
"pad_type_id": 0,
|
17 |
-
"pad_token": "[PAD]"
|
18 |
-
},
|
19 |
"added_tokens": [
|
20 |
{
|
21 |
"id": 0,
|
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
+
"truncation": null,
|
4 |
+
"padding": null,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
"added_tokens": [
|
6 |
{
|
7 |
"id": 0,
|
run-0/checkpoint-156/trainer_state.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
"best_model_checkpoint": "bert-base-uncased-finetuned-rte-run_2/run-0/checkpoint-156",
|
4 |
"epoch": 2.0,
|
5 |
"eval_steps": 500,
|
@@ -10,20 +10,20 @@
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 1.0,
|
13 |
-
"eval_accuracy": 0.
|
14 |
-
"eval_loss": 0.
|
15 |
-
"eval_runtime": 0.
|
16 |
-
"eval_samples_per_second": 383.
|
17 |
-
"eval_steps_per_second": 12.
|
18 |
"step": 78
|
19 |
},
|
20 |
{
|
21 |
"epoch": 2.0,
|
22 |
-
"eval_accuracy": 0.
|
23 |
-
"eval_loss": 0.
|
24 |
-
"eval_runtime": 0.
|
25 |
-
"eval_samples_per_second": 407.
|
26 |
-
"eval_steps_per_second": 13.
|
27 |
"step": 156
|
28 |
}
|
29 |
],
|
@@ -48,9 +48,9 @@
|
|
48 |
"train_batch_size": 32,
|
49 |
"trial_name": null,
|
50 |
"trial_params": {
|
51 |
-
"dropout_rate": 0.
|
52 |
-
"learning_rate":
|
53 |
-
"max_seq_length":
|
54 |
"num_train_epochs": 3,
|
55 |
"per_device_train_batch_size": 32
|
56 |
}
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.628158844765343,
|
3 |
"best_model_checkpoint": "bert-base-uncased-finetuned-rte-run_2/run-0/checkpoint-156",
|
4 |
"epoch": 2.0,
|
5 |
"eval_steps": 500,
|
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 1.0,
|
13 |
+
"eval_accuracy": 0.5992779783393501,
|
14 |
+
"eval_loss": 0.6782958507537842,
|
15 |
+
"eval_runtime": 0.723,
|
16 |
+
"eval_samples_per_second": 383.105,
|
17 |
+
"eval_steps_per_second": 12.447,
|
18 |
"step": 78
|
19 |
},
|
20 |
{
|
21 |
"epoch": 2.0,
|
22 |
+
"eval_accuracy": 0.628158844765343,
|
23 |
+
"eval_loss": 0.6643242835998535,
|
24 |
+
"eval_runtime": 0.6796,
|
25 |
+
"eval_samples_per_second": 407.567,
|
26 |
+
"eval_steps_per_second": 13.242,
|
27 |
"step": 156
|
28 |
}
|
29 |
],
|
|
|
48 |
"train_batch_size": 32,
|
49 |
"trial_name": null,
|
50 |
"trial_params": {
|
51 |
+
"dropout_rate": 0.1377699058361983,
|
52 |
+
"learning_rate": 3.697705960809544e-05,
|
53 |
+
"max_seq_length": 128,
|
54 |
"num_train_epochs": 3,
|
55 |
"per_device_train_batch_size": 32
|
56 |
}
|
run-0/checkpoint-234/tokenizer.json
CHANGED
@@ -1,21 +1,7 @@
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
-
"truncation":
|
4 |
-
|
5 |
-
"max_length": 128,
|
6 |
-
"strategy": "LongestFirst",
|
7 |
-
"stride": 0
|
8 |
-
},
|
9 |
-
"padding": {
|
10 |
-
"strategy": {
|
11 |
-
"Fixed": 128
|
12 |
-
},
|
13 |
-
"direction": "Right",
|
14 |
-
"pad_to_multiple_of": null,
|
15 |
-
"pad_id": 0,
|
16 |
-
"pad_type_id": 0,
|
17 |
-
"pad_token": "[PAD]"
|
18 |
-
},
|
19 |
"added_tokens": [
|
20 |
{
|
21 |
"id": 0,
|
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
+
"truncation": null,
|
4 |
+
"padding": null,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
"added_tokens": [
|
6 |
{
|
7 |
"id": 0,
|
run-0/checkpoint-234/trainer_state.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
"best_model_checkpoint": "bert-base-uncased-finetuned-rte-run_2/run-0/checkpoint-234",
|
4 |
"epoch": 3.0,
|
5 |
"eval_steps": 500,
|
@@ -10,29 +10,29 @@
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 1.0,
|
13 |
-
"eval_accuracy": 0.
|
14 |
-
"eval_loss": 0.
|
15 |
-
"eval_runtime": 0.
|
16 |
-
"eval_samples_per_second": 383.
|
17 |
-
"eval_steps_per_second": 12.
|
18 |
"step": 78
|
19 |
},
|
20 |
{
|
21 |
"epoch": 2.0,
|
22 |
-
"eval_accuracy": 0.
|
23 |
-
"eval_loss": 0.
|
24 |
-
"eval_runtime": 0.
|
25 |
-
"eval_samples_per_second": 407.
|
26 |
-
"eval_steps_per_second": 13.
|
27 |
"step": 156
|
28 |
},
|
29 |
{
|
30 |
"epoch": 3.0,
|
31 |
-
"eval_accuracy": 0.
|
32 |
-
"eval_loss": 0.
|
33 |
-
"eval_runtime": 0.
|
34 |
-
"eval_samples_per_second":
|
35 |
-
"eval_steps_per_second": 12.
|
36 |
"step": 234
|
37 |
}
|
38 |
],
|
@@ -57,9 +57,9 @@
|
|
57 |
"train_batch_size": 32,
|
58 |
"trial_name": null,
|
59 |
"trial_params": {
|
60 |
-
"dropout_rate": 0.
|
61 |
-
"learning_rate":
|
62 |
-
"max_seq_length":
|
63 |
"num_train_epochs": 3,
|
64 |
"per_device_train_batch_size": 32
|
65 |
}
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.6570397111913358,
|
3 |
"best_model_checkpoint": "bert-base-uncased-finetuned-rte-run_2/run-0/checkpoint-234",
|
4 |
"epoch": 3.0,
|
5 |
"eval_steps": 500,
|
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 1.0,
|
13 |
+
"eval_accuracy": 0.5992779783393501,
|
14 |
+
"eval_loss": 0.6782958507537842,
|
15 |
+
"eval_runtime": 0.723,
|
16 |
+
"eval_samples_per_second": 383.105,
|
17 |
+
"eval_steps_per_second": 12.447,
|
18 |
"step": 78
|
19 |
},
|
20 |
{
|
21 |
"epoch": 2.0,
|
22 |
+
"eval_accuracy": 0.628158844765343,
|
23 |
+
"eval_loss": 0.6643242835998535,
|
24 |
+
"eval_runtime": 0.6796,
|
25 |
+
"eval_samples_per_second": 407.567,
|
26 |
+
"eval_steps_per_second": 13.242,
|
27 |
"step": 156
|
28 |
},
|
29 |
{
|
30 |
"epoch": 3.0,
|
31 |
+
"eval_accuracy": 0.6570397111913358,
|
32 |
+
"eval_loss": 0.7470307946205139,
|
33 |
+
"eval_runtime": 0.7182,
|
34 |
+
"eval_samples_per_second": 385.702,
|
35 |
+
"eval_steps_per_second": 12.532,
|
36 |
"step": 234
|
37 |
}
|
38 |
],
|
|
|
57 |
"train_batch_size": 32,
|
58 |
"trial_name": null,
|
59 |
"trial_params": {
|
60 |
+
"dropout_rate": 0.1377699058361983,
|
61 |
+
"learning_rate": 3.697705960809544e-05,
|
62 |
+
"max_seq_length": 128,
|
63 |
"num_train_epochs": 3,
|
64 |
"per_device_train_batch_size": 32
|
65 |
}
|
run-0/checkpoint-78/model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 437958648
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:638ab00ccaa37bc4e7a65c38d74ec0becba8f345692e1517641e1b426eb57268
|
3 |
size 437958648
|
run-0/checkpoint-78/optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 876038394
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6f600160c6b46e820ff56ca4c73a3606d9d600c471a5beb1fe1f625eef85ff7f
|
3 |
size 876038394
|
run-0/checkpoint-78/scheduler.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1064
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7d7b0f45906b70b36610ba990d8cc62c6375947fdb497170b4aca0a7218da74f
|
3 |
size 1064
|
run-0/checkpoint-78/tokenizer.json
CHANGED
@@ -1,21 +1,7 @@
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
-
"truncation":
|
4 |
-
|
5 |
-
"max_length": 128,
|
6 |
-
"strategy": "LongestFirst",
|
7 |
-
"stride": 0
|
8 |
-
},
|
9 |
-
"padding": {
|
10 |
-
"strategy": {
|
11 |
-
"Fixed": 128
|
12 |
-
},
|
13 |
-
"direction": "Right",
|
14 |
-
"pad_to_multiple_of": null,
|
15 |
-
"pad_id": 0,
|
16 |
-
"pad_type_id": 0,
|
17 |
-
"pad_token": "[PAD]"
|
18 |
-
},
|
19 |
"added_tokens": [
|
20 |
{
|
21 |
"id": 0,
|
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
+
"truncation": null,
|
4 |
+
"padding": null,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
"added_tokens": [
|
6 |
{
|
7 |
"id": 0,
|
run-0/checkpoint-78/trainer_state.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
"best_model_checkpoint": "bert-base-uncased-finetuned-rte-run_2/run-0/checkpoint-78",
|
4 |
"epoch": 1.0,
|
5 |
"eval_steps": 500,
|
@@ -10,11 +10,11 @@
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 1.0,
|
13 |
-
"eval_accuracy": 0.
|
14 |
-
"eval_loss": 0.
|
15 |
-
"eval_runtime": 0.
|
16 |
-
"eval_samples_per_second": 383.
|
17 |
-
"eval_steps_per_second": 12.
|
18 |
"step": 78
|
19 |
}
|
20 |
],
|
@@ -39,9 +39,9 @@
|
|
39 |
"train_batch_size": 32,
|
40 |
"trial_name": null,
|
41 |
"trial_params": {
|
42 |
-
"dropout_rate": 0.
|
43 |
-
"learning_rate":
|
44 |
-
"max_seq_length":
|
45 |
"num_train_epochs": 3,
|
46 |
"per_device_train_batch_size": 32
|
47 |
}
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.5992779783393501,
|
3 |
"best_model_checkpoint": "bert-base-uncased-finetuned-rte-run_2/run-0/checkpoint-78",
|
4 |
"epoch": 1.0,
|
5 |
"eval_steps": 500,
|
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 1.0,
|
13 |
+
"eval_accuracy": 0.5992779783393501,
|
14 |
+
"eval_loss": 0.6782958507537842,
|
15 |
+
"eval_runtime": 0.723,
|
16 |
+
"eval_samples_per_second": 383.105,
|
17 |
+
"eval_steps_per_second": 12.447,
|
18 |
"step": 78
|
19 |
}
|
20 |
],
|
|
|
39 |
"train_batch_size": 32,
|
40 |
"trial_name": null,
|
41 |
"trial_params": {
|
42 |
+
"dropout_rate": 0.1377699058361983,
|
43 |
+
"learning_rate": 3.697705960809544e-05,
|
44 |
+
"max_seq_length": 128,
|
45 |
"num_train_epochs": 3,
|
46 |
"per_device_train_batch_size": 32
|
47 |
}
|
run-0/checkpoint-78/training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 5240
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:95028e7deee272cbc151f7568f076355b10d7dd155b317c34385edbac9e0a889
|
3 |
size 5240
|
tokenizer.json
CHANGED
@@ -1,21 +1,7 @@
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
-
"truncation":
|
4 |
-
|
5 |
-
"max_length": 64,
|
6 |
-
"strategy": "LongestFirst",
|
7 |
-
"stride": 0
|
8 |
-
},
|
9 |
-
"padding": {
|
10 |
-
"strategy": {
|
11 |
-
"Fixed": 64
|
12 |
-
},
|
13 |
-
"direction": "Right",
|
14 |
-
"pad_to_multiple_of": null,
|
15 |
-
"pad_id": 0,
|
16 |
-
"pad_type_id": 0,
|
17 |
-
"pad_token": "[PAD]"
|
18 |
-
},
|
19 |
"added_tokens": [
|
20 |
{
|
21 |
"id": 0,
|
|
|
1 |
{
|
2 |
"version": "1.0",
|
3 |
+
"truncation": null,
|
4 |
+
"padding": null,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
"added_tokens": [
|
6 |
{
|
7 |
"id": 0,
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 5240
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:95028e7deee272cbc151f7568f076355b10d7dd155b317c34385edbac9e0a889
|
3 |
size 5240
|