Training in progress, step 3000, checkpoint
Browse files
checkpoint-3000/config.json
ADDED
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "openai/clip-vit-large-patch14-336",
|
3 |
+
"architectures": [
|
4 |
+
"CLIPModel"
|
5 |
+
],
|
6 |
+
"initializer_factor": 1.0,
|
7 |
+
"logit_scale_init_value": 2.6592,
|
8 |
+
"model_type": "clip",
|
9 |
+
"projection_dim": 768,
|
10 |
+
"text_config": {
|
11 |
+
"dropout": 0.0,
|
12 |
+
"hidden_size": 768,
|
13 |
+
"intermediate_size": 3072,
|
14 |
+
"model_type": "clip_text_model",
|
15 |
+
"num_attention_heads": 12,
|
16 |
+
"projection_dim": 768
|
17 |
+
},
|
18 |
+
"torch_dtype": "float32",
|
19 |
+
"transformers_version": "4.45.0.dev0",
|
20 |
+
"vision_config": {
|
21 |
+
"dropout": 0.0,
|
22 |
+
"hidden_size": 1024,
|
23 |
+
"image_size": 336,
|
24 |
+
"intermediate_size": 4096,
|
25 |
+
"model_type": "clip_vision_model",
|
26 |
+
"num_attention_heads": 16,
|
27 |
+
"num_hidden_layers": 24,
|
28 |
+
"patch_size": 14,
|
29 |
+
"projection_dim": 768
|
30 |
+
}
|
31 |
+
}
|
checkpoint-3000/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:42f324b5c5de3a9d4ebbd3345684f58ef75c41d122ad9a00b23755e7d3ca46f8
|
3 |
+
size 1711848436
|
checkpoint-3000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ab7399566e6150346704b377e9b5f7c9ae14e347dfcf67ea8e2099a497335f10
|
3 |
+
size 3424043887
|
checkpoint-3000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3f9733db299f85727654ada1e2241c2b4f64948c1fd868b50f2037c2a0349da2
|
3 |
+
size 14503
|
checkpoint-3000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6a4aba2e3883e8c44a7e42620e0e03aa0200eac0443482811c080e77230b85e2
|
3 |
+
size 623
|
checkpoint-3000/trainer_state.json
ADDED
@@ -0,0 +1,123 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.7885717749595642,
|
3 |
+
"best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l37-l/checkpoint-3000",
|
4 |
+
"epoch": 0.552689756816507,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 3000,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.09211495946941783,
|
13 |
+
"grad_norm": 8.437943458557129,
|
14 |
+
"learning_rate": 2.907885040530582e-07,
|
15 |
+
"loss": 0.3017,
|
16 |
+
"step": 500
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.09211495946941783,
|
20 |
+
"eval_loss": 1.164457082748413,
|
21 |
+
"eval_runtime": 75.2697,
|
22 |
+
"eval_samples_per_second": 16.036,
|
23 |
+
"eval_steps_per_second": 2.006,
|
24 |
+
"step": 500
|
25 |
+
},
|
26 |
+
{
|
27 |
+
"epoch": 0.18422991893883567,
|
28 |
+
"grad_norm": 14.740598678588867,
|
29 |
+
"learning_rate": 2.815770081061164e-07,
|
30 |
+
"loss": 0.2899,
|
31 |
+
"step": 1000
|
32 |
+
},
|
33 |
+
{
|
34 |
+
"epoch": 0.18422991893883567,
|
35 |
+
"eval_loss": 1.01627779006958,
|
36 |
+
"eval_runtime": 76.4934,
|
37 |
+
"eval_samples_per_second": 15.779,
|
38 |
+
"eval_steps_per_second": 1.974,
|
39 |
+
"step": 1000
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.2763448784082535,
|
43 |
+
"grad_norm": 0.17031528055667877,
|
44 |
+
"learning_rate": 2.723655121591746e-07,
|
45 |
+
"loss": 0.273,
|
46 |
+
"step": 1500
|
47 |
+
},
|
48 |
+
{
|
49 |
+
"epoch": 0.2763448784082535,
|
50 |
+
"eval_loss": 0.9395213723182678,
|
51 |
+
"eval_runtime": 76.6256,
|
52 |
+
"eval_samples_per_second": 15.752,
|
53 |
+
"eval_steps_per_second": 1.971,
|
54 |
+
"step": 1500
|
55 |
+
},
|
56 |
+
{
|
57 |
+
"epoch": 0.36845983787767134,
|
58 |
+
"grad_norm": 0.0030106802005320787,
|
59 |
+
"learning_rate": 2.6315401621223287e-07,
|
60 |
+
"loss": 0.1818,
|
61 |
+
"step": 2000
|
62 |
+
},
|
63 |
+
{
|
64 |
+
"epoch": 0.36845983787767134,
|
65 |
+
"eval_loss": 0.8942967057228088,
|
66 |
+
"eval_runtime": 76.8089,
|
67 |
+
"eval_samples_per_second": 15.714,
|
68 |
+
"eval_steps_per_second": 1.966,
|
69 |
+
"step": 2000
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.46057479734708917,
|
73 |
+
"grad_norm": 474.4410705566406,
|
74 |
+
"learning_rate": 2.539425202652911e-07,
|
75 |
+
"loss": 0.2094,
|
76 |
+
"step": 2500
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 0.46057479734708917,
|
80 |
+
"eval_loss": 0.8330363631248474,
|
81 |
+
"eval_runtime": 76.9939,
|
82 |
+
"eval_samples_per_second": 15.677,
|
83 |
+
"eval_steps_per_second": 1.961,
|
84 |
+
"step": 2500
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 0.552689756816507,
|
88 |
+
"grad_norm": 314.3533935546875,
|
89 |
+
"learning_rate": 2.447310243183493e-07,
|
90 |
+
"loss": 0.1949,
|
91 |
+
"step": 3000
|
92 |
+
},
|
93 |
+
{
|
94 |
+
"epoch": 0.552689756816507,
|
95 |
+
"eval_loss": 0.7885717749595642,
|
96 |
+
"eval_runtime": 76.7534,
|
97 |
+
"eval_samples_per_second": 15.726,
|
98 |
+
"eval_steps_per_second": 1.967,
|
99 |
+
"step": 3000
|
100 |
+
}
|
101 |
+
],
|
102 |
+
"logging_steps": 500,
|
103 |
+
"max_steps": 16284,
|
104 |
+
"num_input_tokens_seen": 0,
|
105 |
+
"num_train_epochs": 3,
|
106 |
+
"save_steps": 500,
|
107 |
+
"stateful_callbacks": {
|
108 |
+
"TrainerControl": {
|
109 |
+
"args": {
|
110 |
+
"should_epoch_stop": false,
|
111 |
+
"should_evaluate": false,
|
112 |
+
"should_log": false,
|
113 |
+
"should_save": true,
|
114 |
+
"should_training_stop": false
|
115 |
+
},
|
116 |
+
"attributes": {}
|
117 |
+
}
|
118 |
+
},
|
119 |
+
"total_flos": 1079994570482520.0,
|
120 |
+
"train_batch_size": 2,
|
121 |
+
"trial_name": null,
|
122 |
+
"trial_params": null
|
123 |
+
}
|
checkpoint-3000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:76f7dfc201ccc0039551250d06ef1282a727ca0a51016f874c2eff3c580dbf8e
|
3 |
+
size 4847
|