Commit
·
ef457fd
1
Parent(s):
9805cb3
Upload 8 files
Browse files- config.json +88 -0
- optimizer.pt +3 -0
- preprocessor_config.json +29 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +469 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,88 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "microsoft/beit-base-patch16-224-pt22k-ft22k",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BeitForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"auxiliary_channels": 256,
|
| 8 |
+
"auxiliary_concat_input": false,
|
| 9 |
+
"auxiliary_loss_weight": 0.4,
|
| 10 |
+
"auxiliary_num_convs": 1,
|
| 11 |
+
"drop_path_rate": 0.1,
|
| 12 |
+
"hidden_act": "gelu",
|
| 13 |
+
"hidden_dropout_prob": 0.0,
|
| 14 |
+
"hidden_size": 768,
|
| 15 |
+
"id2label": {
|
| 16 |
+
"0": "document_edf",
|
| 17 |
+
"1": "facture_amazon",
|
| 18 |
+
"2": "facture_edenkia",
|
| 19 |
+
"3": "facture_edf",
|
| 20 |
+
"4": "facture_engie",
|
| 21 |
+
"5": "facture_mint",
|
| 22 |
+
"6": "facture_misc",
|
| 23 |
+
"7": "facture_orange",
|
| 24 |
+
"8": "facture_primagaz",
|
| 25 |
+
"9": "facture_primeo",
|
| 26 |
+
"10": "facture_sicap",
|
| 27 |
+
"11": "facture_smeg",
|
| 28 |
+
"12": "facture_sosh",
|
| 29 |
+
"13": "facture_totalenergies",
|
| 30 |
+
"14": "ticket_carburant",
|
| 31 |
+
"15": "ticket_misc",
|
| 32 |
+
"16": "ticket_restaurant",
|
| 33 |
+
"17": "ticket_store"
|
| 34 |
+
},
|
| 35 |
+
"image_size": 224,
|
| 36 |
+
"initializer_range": 0.02,
|
| 37 |
+
"intermediate_size": 3072,
|
| 38 |
+
"label2id": {
|
| 39 |
+
"document_edf": 0,
|
| 40 |
+
"facture_amazon": 1,
|
| 41 |
+
"facture_edenkia": 2,
|
| 42 |
+
"facture_edf": 3,
|
| 43 |
+
"facture_engie": 4,
|
| 44 |
+
"facture_mint": 5,
|
| 45 |
+
"facture_misc": 6,
|
| 46 |
+
"facture_orange": 7,
|
| 47 |
+
"facture_primagaz": 8,
|
| 48 |
+
"facture_primeo": 9,
|
| 49 |
+
"facture_sicap": 10,
|
| 50 |
+
"facture_smeg": 11,
|
| 51 |
+
"facture_sosh": 12,
|
| 52 |
+
"facture_totalenergies": 13,
|
| 53 |
+
"ticket_carburant": 14,
|
| 54 |
+
"ticket_misc": 15,
|
| 55 |
+
"ticket_restaurant": 16,
|
| 56 |
+
"ticket_store": 17
|
| 57 |
+
},
|
| 58 |
+
"layer_norm_eps": 1e-12,
|
| 59 |
+
"layer_scale_init_value": 0.1,
|
| 60 |
+
"model_type": "beit",
|
| 61 |
+
"num_attention_heads": 12,
|
| 62 |
+
"num_channels": 3,
|
| 63 |
+
"num_hidden_layers": 12,
|
| 64 |
+
"out_indices": [
|
| 65 |
+
3,
|
| 66 |
+
5,
|
| 67 |
+
7,
|
| 68 |
+
11
|
| 69 |
+
],
|
| 70 |
+
"patch_size": 16,
|
| 71 |
+
"pool_scales": [
|
| 72 |
+
1,
|
| 73 |
+
2,
|
| 74 |
+
3,
|
| 75 |
+
6
|
| 76 |
+
],
|
| 77 |
+
"problem_type": "single_label_classification",
|
| 78 |
+
"semantic_loss_ignore_index": 255,
|
| 79 |
+
"torch_dtype": "float32",
|
| 80 |
+
"transformers_version": "4.33.1",
|
| 81 |
+
"use_absolute_position_embeddings": false,
|
| 82 |
+
"use_auxiliary_head": true,
|
| 83 |
+
"use_mask_token": false,
|
| 84 |
+
"use_mean_pooling": true,
|
| 85 |
+
"use_relative_position_bias": true,
|
| 86 |
+
"use_shared_relative_position_bias": false,
|
| 87 |
+
"vocab_size": 8192
|
| 88 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:43a055749e7876c3ad5198c6a3b9a25fbc6bc1d8de1f6edfc56d9117690bbc57
|
| 3 |
+
size 686392645
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"crop_size": {
|
| 3 |
+
"height": 224,
|
| 4 |
+
"width": 224
|
| 5 |
+
},
|
| 6 |
+
"do_center_crop": false,
|
| 7 |
+
"do_normalize": true,
|
| 8 |
+
"do_reduce_labels": false,
|
| 9 |
+
"do_rescale": true,
|
| 10 |
+
"do_resize": true,
|
| 11 |
+
"feature_extractor_type": "BeitFeatureExtractor",
|
| 12 |
+
"image_mean": [
|
| 13 |
+
0.5,
|
| 14 |
+
0.5,
|
| 15 |
+
0.5
|
| 16 |
+
],
|
| 17 |
+
"image_processor_type": "BeitImageProcessor",
|
| 18 |
+
"image_std": [
|
| 19 |
+
0.5,
|
| 20 |
+
0.5,
|
| 21 |
+
0.5
|
| 22 |
+
],
|
| 23 |
+
"resample": 2,
|
| 24 |
+
"rescale_factor": 0.00392156862745098,
|
| 25 |
+
"size": {
|
| 26 |
+
"height": 224,
|
| 27 |
+
"width": 224
|
| 28 |
+
}
|
| 29 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0ccb8eff7ac20c117d82eebd69a252d335d1dc9b7c33d76112de2bd5f762cb98
|
| 3 |
+
size 343179465
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f022cc454603cf1933f0b2f55419b47e74c16e5eccd8a74e6c810757e132ec9c
|
| 3 |
+
size 14575
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bb979105f0e674466d7fc74a0fb0f37d6272fc2bea59c80ffe700f6553534d38
|
| 3 |
+
size 627
|
trainer_state.json
ADDED
|
@@ -0,0 +1,469 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.14660954475402832,
|
| 3 |
+
"best_model_checkpoint": "./Beit-doc-classification/checkpoint-240",
|
| 4 |
+
"epoch": 18.75,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
+
"global_step": 300,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.62,
|
| 13 |
+
"learning_rate": 4.8906250000000006e-05,
|
| 14 |
+
"loss": 2.174,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.62,
|
| 19 |
+
"eval_accuracy": 0.6964285714285714,
|
| 20 |
+
"eval_loss": 0.7704058885574341,
|
| 21 |
+
"eval_runtime": 0.2746,
|
| 22 |
+
"eval_samples_per_second": 203.898,
|
| 23 |
+
"eval_steps_per_second": 25.487,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.25,
|
| 28 |
+
"learning_rate": 4.734375e-05,
|
| 29 |
+
"loss": 0.8459,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 1.25,
|
| 34 |
+
"eval_accuracy": 0.75,
|
| 35 |
+
"eval_loss": 0.6054777503013611,
|
| 36 |
+
"eval_runtime": 0.2697,
|
| 37 |
+
"eval_samples_per_second": 207.624,
|
| 38 |
+
"eval_steps_per_second": 25.953,
|
| 39 |
+
"step": 20
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.88,
|
| 43 |
+
"learning_rate": 4.5781250000000005e-05,
|
| 44 |
+
"loss": 0.5794,
|
| 45 |
+
"step": 30
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.88,
|
| 49 |
+
"eval_accuracy": 0.75,
|
| 50 |
+
"eval_loss": 0.4210772216320038,
|
| 51 |
+
"eval_runtime": 0.2721,
|
| 52 |
+
"eval_samples_per_second": 205.818,
|
| 53 |
+
"eval_steps_per_second": 25.727,
|
| 54 |
+
"step": 30
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 2.5,
|
| 58 |
+
"learning_rate": 4.421875e-05,
|
| 59 |
+
"loss": 0.356,
|
| 60 |
+
"step": 40
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 2.5,
|
| 64 |
+
"eval_accuracy": 0.9285714285714286,
|
| 65 |
+
"eval_loss": 0.23619188368320465,
|
| 66 |
+
"eval_runtime": 0.276,
|
| 67 |
+
"eval_samples_per_second": 202.895,
|
| 68 |
+
"eval_steps_per_second": 25.362,
|
| 69 |
+
"step": 40
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 3.12,
|
| 73 |
+
"learning_rate": 4.2656250000000003e-05,
|
| 74 |
+
"loss": 0.2762,
|
| 75 |
+
"step": 50
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 3.12,
|
| 79 |
+
"eval_accuracy": 0.9464285714285714,
|
| 80 |
+
"eval_loss": 0.2510121166706085,
|
| 81 |
+
"eval_runtime": 0.2788,
|
| 82 |
+
"eval_samples_per_second": 200.857,
|
| 83 |
+
"eval_steps_per_second": 25.107,
|
| 84 |
+
"step": 50
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 3.75,
|
| 88 |
+
"learning_rate": 4.1093750000000006e-05,
|
| 89 |
+
"loss": 0.2331,
|
| 90 |
+
"step": 60
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 3.75,
|
| 94 |
+
"eval_accuracy": 0.9642857142857143,
|
| 95 |
+
"eval_loss": 0.18232552707195282,
|
| 96 |
+
"eval_runtime": 0.2722,
|
| 97 |
+
"eval_samples_per_second": 205.715,
|
| 98 |
+
"eval_steps_per_second": 25.714,
|
| 99 |
+
"step": 60
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 4.38,
|
| 103 |
+
"learning_rate": 3.953125e-05,
|
| 104 |
+
"loss": 0.183,
|
| 105 |
+
"step": 70
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 4.38,
|
| 109 |
+
"eval_accuracy": 0.9642857142857143,
|
| 110 |
+
"eval_loss": 0.1663789004087448,
|
| 111 |
+
"eval_runtime": 0.2703,
|
| 112 |
+
"eval_samples_per_second": 207.189,
|
| 113 |
+
"eval_steps_per_second": 25.899,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 5.0,
|
| 118 |
+
"learning_rate": 3.7968750000000005e-05,
|
| 119 |
+
"loss": 0.1171,
|
| 120 |
+
"step": 80
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 5.0,
|
| 124 |
+
"eval_accuracy": 0.9642857142857143,
|
| 125 |
+
"eval_loss": 0.182296484708786,
|
| 126 |
+
"eval_runtime": 0.2718,
|
| 127 |
+
"eval_samples_per_second": 206.04,
|
| 128 |
+
"eval_steps_per_second": 25.755,
|
| 129 |
+
"step": 80
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 5.62,
|
| 133 |
+
"learning_rate": 3.640625e-05,
|
| 134 |
+
"loss": 0.0875,
|
| 135 |
+
"step": 90
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 5.62,
|
| 139 |
+
"eval_accuracy": 0.9642857142857143,
|
| 140 |
+
"eval_loss": 0.18081991374492645,
|
| 141 |
+
"eval_runtime": 0.2717,
|
| 142 |
+
"eval_samples_per_second": 206.097,
|
| 143 |
+
"eval_steps_per_second": 25.762,
|
| 144 |
+
"step": 90
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 6.25,
|
| 148 |
+
"learning_rate": 3.484375e-05,
|
| 149 |
+
"loss": 0.1186,
|
| 150 |
+
"step": 100
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 6.25,
|
| 154 |
+
"eval_accuracy": 0.9642857142857143,
|
| 155 |
+
"eval_loss": 0.1665651649236679,
|
| 156 |
+
"eval_runtime": 0.2734,
|
| 157 |
+
"eval_samples_per_second": 204.829,
|
| 158 |
+
"eval_steps_per_second": 25.604,
|
| 159 |
+
"step": 100
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 6.88,
|
| 163 |
+
"learning_rate": 3.3281250000000006e-05,
|
| 164 |
+
"loss": 0.1174,
|
| 165 |
+
"step": 110
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 6.88,
|
| 169 |
+
"eval_accuracy": 0.9642857142857143,
|
| 170 |
+
"eval_loss": 0.14654958248138428,
|
| 171 |
+
"eval_runtime": 0.2716,
|
| 172 |
+
"eval_samples_per_second": 206.219,
|
| 173 |
+
"eval_steps_per_second": 25.777,
|
| 174 |
+
"step": 110
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 7.5,
|
| 178 |
+
"learning_rate": 3.171875e-05,
|
| 179 |
+
"loss": 0.0519,
|
| 180 |
+
"step": 120
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 7.5,
|
| 184 |
+
"eval_accuracy": 0.9642857142857143,
|
| 185 |
+
"eval_loss": 0.1532459259033203,
|
| 186 |
+
"eval_runtime": 0.2731,
|
| 187 |
+
"eval_samples_per_second": 205.051,
|
| 188 |
+
"eval_steps_per_second": 25.631,
|
| 189 |
+
"step": 120
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 8.12,
|
| 193 |
+
"learning_rate": 3.015625e-05,
|
| 194 |
+
"loss": 0.0338,
|
| 195 |
+
"step": 130
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 8.12,
|
| 199 |
+
"eval_accuracy": 0.9642857142857143,
|
| 200 |
+
"eval_loss": 0.18999779224395752,
|
| 201 |
+
"eval_runtime": 0.272,
|
| 202 |
+
"eval_samples_per_second": 205.856,
|
| 203 |
+
"eval_steps_per_second": 25.732,
|
| 204 |
+
"step": 130
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 8.75,
|
| 208 |
+
"learning_rate": 2.8593750000000004e-05,
|
| 209 |
+
"loss": 0.0241,
|
| 210 |
+
"step": 140
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 8.75,
|
| 214 |
+
"eval_accuracy": 0.9642857142857143,
|
| 215 |
+
"eval_loss": 0.1611316055059433,
|
| 216 |
+
"eval_runtime": 0.2718,
|
| 217 |
+
"eval_samples_per_second": 206.047,
|
| 218 |
+
"eval_steps_per_second": 25.756,
|
| 219 |
+
"step": 140
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 9.38,
|
| 223 |
+
"learning_rate": 2.7031250000000003e-05,
|
| 224 |
+
"loss": 0.0138,
|
| 225 |
+
"step": 150
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 9.38,
|
| 229 |
+
"eval_accuracy": 0.9642857142857143,
|
| 230 |
+
"eval_loss": 0.17584912478923798,
|
| 231 |
+
"eval_runtime": 0.2713,
|
| 232 |
+
"eval_samples_per_second": 206.411,
|
| 233 |
+
"eval_steps_per_second": 25.801,
|
| 234 |
+
"step": 150
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 10.0,
|
| 238 |
+
"learning_rate": 2.5468750000000002e-05,
|
| 239 |
+
"loss": 0.0155,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 10.0,
|
| 244 |
+
"eval_accuracy": 0.9642857142857143,
|
| 245 |
+
"eval_loss": 0.14303383231163025,
|
| 246 |
+
"eval_runtime": 0.2704,
|
| 247 |
+
"eval_samples_per_second": 207.101,
|
| 248 |
+
"eval_steps_per_second": 25.888,
|
| 249 |
+
"step": 160
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 10.62,
|
| 253 |
+
"learning_rate": 2.3906250000000002e-05,
|
| 254 |
+
"loss": 0.0478,
|
| 255 |
+
"step": 170
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 10.62,
|
| 259 |
+
"eval_accuracy": 0.9642857142857143,
|
| 260 |
+
"eval_loss": 0.15157583355903625,
|
| 261 |
+
"eval_runtime": 0.2795,
|
| 262 |
+
"eval_samples_per_second": 200.393,
|
| 263 |
+
"eval_steps_per_second": 25.049,
|
| 264 |
+
"step": 170
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 11.25,
|
| 268 |
+
"learning_rate": 2.234375e-05,
|
| 269 |
+
"loss": 0.0228,
|
| 270 |
+
"step": 180
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 11.25,
|
| 274 |
+
"eval_accuracy": 0.9642857142857143,
|
| 275 |
+
"eval_loss": 0.17853426933288574,
|
| 276 |
+
"eval_runtime": 0.2946,
|
| 277 |
+
"eval_samples_per_second": 190.103,
|
| 278 |
+
"eval_steps_per_second": 23.763,
|
| 279 |
+
"step": 180
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 11.88,
|
| 283 |
+
"learning_rate": 2.0781250000000004e-05,
|
| 284 |
+
"loss": 0.0295,
|
| 285 |
+
"step": 190
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 11.88,
|
| 289 |
+
"eval_accuracy": 0.9642857142857143,
|
| 290 |
+
"eval_loss": 0.16022925078868866,
|
| 291 |
+
"eval_runtime": 0.2758,
|
| 292 |
+
"eval_samples_per_second": 203.034,
|
| 293 |
+
"eval_steps_per_second": 25.379,
|
| 294 |
+
"step": 190
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 12.5,
|
| 298 |
+
"learning_rate": 1.9218750000000003e-05,
|
| 299 |
+
"loss": 0.0084,
|
| 300 |
+
"step": 200
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 12.5,
|
| 304 |
+
"eval_accuracy": 0.9642857142857143,
|
| 305 |
+
"eval_loss": 0.19061991572380066,
|
| 306 |
+
"eval_runtime": 0.281,
|
| 307 |
+
"eval_samples_per_second": 199.27,
|
| 308 |
+
"eval_steps_per_second": 24.909,
|
| 309 |
+
"step": 200
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 13.12,
|
| 313 |
+
"learning_rate": 1.7656250000000002e-05,
|
| 314 |
+
"loss": 0.0064,
|
| 315 |
+
"step": 210
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 13.12,
|
| 319 |
+
"eval_accuracy": 0.9642857142857143,
|
| 320 |
+
"eval_loss": 0.1927066147327423,
|
| 321 |
+
"eval_runtime": 0.272,
|
| 322 |
+
"eval_samples_per_second": 205.909,
|
| 323 |
+
"eval_steps_per_second": 25.739,
|
| 324 |
+
"step": 210
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 13.75,
|
| 328 |
+
"learning_rate": 1.609375e-05,
|
| 329 |
+
"loss": 0.0131,
|
| 330 |
+
"step": 220
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 13.75,
|
| 334 |
+
"eval_accuracy": 0.9642857142857143,
|
| 335 |
+
"eval_loss": 0.17037805914878845,
|
| 336 |
+
"eval_runtime": 0.2756,
|
| 337 |
+
"eval_samples_per_second": 203.19,
|
| 338 |
+
"eval_steps_per_second": 25.399,
|
| 339 |
+
"step": 220
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 14.38,
|
| 343 |
+
"learning_rate": 1.4531250000000003e-05,
|
| 344 |
+
"loss": 0.0134,
|
| 345 |
+
"step": 230
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 14.38,
|
| 349 |
+
"eval_accuracy": 0.9642857142857143,
|
| 350 |
+
"eval_loss": 0.14340999722480774,
|
| 351 |
+
"eval_runtime": 0.2733,
|
| 352 |
+
"eval_samples_per_second": 204.909,
|
| 353 |
+
"eval_steps_per_second": 25.614,
|
| 354 |
+
"step": 230
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 15.0,
|
| 358 |
+
"learning_rate": 1.2968750000000002e-05,
|
| 359 |
+
"loss": 0.0082,
|
| 360 |
+
"step": 240
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 15.0,
|
| 364 |
+
"eval_accuracy": 0.9821428571428571,
|
| 365 |
+
"eval_loss": 0.14660954475402832,
|
| 366 |
+
"eval_runtime": 0.2748,
|
| 367 |
+
"eval_samples_per_second": 203.761,
|
| 368 |
+
"eval_steps_per_second": 25.47,
|
| 369 |
+
"step": 240
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 15.62,
|
| 373 |
+
"learning_rate": 1.140625e-05,
|
| 374 |
+
"loss": 0.0082,
|
| 375 |
+
"step": 250
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 15.62,
|
| 379 |
+
"eval_accuracy": 0.9642857142857143,
|
| 380 |
+
"eval_loss": 0.15767399966716766,
|
| 381 |
+
"eval_runtime": 0.2783,
|
| 382 |
+
"eval_samples_per_second": 201.226,
|
| 383 |
+
"eval_steps_per_second": 25.153,
|
| 384 |
+
"step": 250
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 16.25,
|
| 388 |
+
"learning_rate": 9.84375e-06,
|
| 389 |
+
"loss": 0.0104,
|
| 390 |
+
"step": 260
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 16.25,
|
| 394 |
+
"eval_accuracy": 0.9642857142857143,
|
| 395 |
+
"eval_loss": 0.17043821513652802,
|
| 396 |
+
"eval_runtime": 0.2775,
|
| 397 |
+
"eval_samples_per_second": 201.786,
|
| 398 |
+
"eval_steps_per_second": 25.223,
|
| 399 |
+
"step": 260
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 16.88,
|
| 403 |
+
"learning_rate": 8.28125e-06,
|
| 404 |
+
"loss": 0.0099,
|
| 405 |
+
"step": 270
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 16.88,
|
| 409 |
+
"eval_accuracy": 0.9642857142857143,
|
| 410 |
+
"eval_loss": 0.1689167469739914,
|
| 411 |
+
"eval_runtime": 0.2808,
|
| 412 |
+
"eval_samples_per_second": 199.425,
|
| 413 |
+
"eval_steps_per_second": 24.928,
|
| 414 |
+
"step": 270
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 17.5,
|
| 418 |
+
"learning_rate": 6.71875e-06,
|
| 419 |
+
"loss": 0.0033,
|
| 420 |
+
"step": 280
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 17.5,
|
| 424 |
+
"eval_accuracy": 0.9642857142857143,
|
| 425 |
+
"eval_loss": 0.1613139808177948,
|
| 426 |
+
"eval_runtime": 0.2796,
|
| 427 |
+
"eval_samples_per_second": 200.293,
|
| 428 |
+
"eval_steps_per_second": 25.037,
|
| 429 |
+
"step": 280
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 18.12,
|
| 433 |
+
"learning_rate": 5.15625e-06,
|
| 434 |
+
"loss": 0.0046,
|
| 435 |
+
"step": 290
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 18.12,
|
| 439 |
+
"eval_accuracy": 0.9642857142857143,
|
| 440 |
+
"eval_loss": 0.15769068896770477,
|
| 441 |
+
"eval_runtime": 0.2801,
|
| 442 |
+
"eval_samples_per_second": 199.909,
|
| 443 |
+
"eval_steps_per_second": 24.989,
|
| 444 |
+
"step": 290
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 18.75,
|
| 448 |
+
"learning_rate": 3.5937499999999997e-06,
|
| 449 |
+
"loss": 0.0017,
|
| 450 |
+
"step": 300
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 18.75,
|
| 454 |
+
"eval_accuracy": 0.9642857142857143,
|
| 455 |
+
"eval_loss": 0.15903140604496002,
|
| 456 |
+
"eval_runtime": 0.2817,
|
| 457 |
+
"eval_samples_per_second": 198.793,
|
| 458 |
+
"eval_steps_per_second": 24.849,
|
| 459 |
+
"step": 300
|
| 460 |
+
}
|
| 461 |
+
],
|
| 462 |
+
"logging_steps": 10,
|
| 463 |
+
"max_steps": 320,
|
| 464 |
+
"num_train_epochs": 20,
|
| 465 |
+
"save_steps": 30,
|
| 466 |
+
"total_flos": 7.227949247524454e+17,
|
| 467 |
+
"trial_name": null,
|
| 468 |
+
"trial_params": null
|
| 469 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:631ee13b1e3bdaa49056a14c291287ffe97a6a46ff5b14386113f7f57f4a67e7
|
| 3 |
+
size 4027
|