Training in progress, step 4500, checkpoint
Browse files
checkpoint-4500/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-4500/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2f11766aad552c3bd740b01278fc7f992d535880de21ea92d17cb1d0e3b14269
|
3 |
+
size 1711848436
|
checkpoint-4500/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4960ed73d0fb6fef6ae83e87bca7915c8796619388b97118e9edb865422cff09
|
3 |
+
size 3424043887
|
checkpoint-4500/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:53d6607299e954768b3402606a5a5e5f5d35597509d6aebab8b3589af8a48151
|
3 |
+
size 14503
|
checkpoint-4500/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a4290d6d3e40ee42d897582f61177cbdcacf854e30b59b19f7a693b641a4ad17
|
3 |
+
size 623
|
checkpoint-4500/trainer_state.json
ADDED
@@ -0,0 +1,168 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 2.0000314712524414,
|
3 |
+
"best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l55-l/checkpoint-4000",
|
4 |
+
"epoch": 0.8288819303739179,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 4500,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.09209799226376865,
|
13 |
+
"grad_norm": 0.009597355499863625,
|
14 |
+
"learning_rate": 4.846503346227052e-05,
|
15 |
+
"loss": 0.7336,
|
16 |
+
"step": 500
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.09209799226376865,
|
20 |
+
"eval_loss": 2.0780556201934814,
|
21 |
+
"eval_runtime": 77.1404,
|
22 |
+
"eval_samples_per_second": 15.634,
|
23 |
+
"eval_steps_per_second": 1.957,
|
24 |
+
"step": 500
|
25 |
+
},
|
26 |
+
{
|
27 |
+
"epoch": 0.1841959845275373,
|
28 |
+
"grad_norm": 0.10049372166395187,
|
29 |
+
"learning_rate": 4.693006692454105e-05,
|
30 |
+
"loss": 0.6953,
|
31 |
+
"step": 1000
|
32 |
+
},
|
33 |
+
{
|
34 |
+
"epoch": 0.1841959845275373,
|
35 |
+
"eval_loss": 2.0782744884490967,
|
36 |
+
"eval_runtime": 77.8965,
|
37 |
+
"eval_samples_per_second": 15.482,
|
38 |
+
"eval_steps_per_second": 1.938,
|
39 |
+
"step": 1000
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.27629397679130596,
|
43 |
+
"grad_norm": 0.0041180383414030075,
|
44 |
+
"learning_rate": 4.539510038681157e-05,
|
45 |
+
"loss": 0.6953,
|
46 |
+
"step": 1500
|
47 |
+
},
|
48 |
+
{
|
49 |
+
"epoch": 0.27629397679130596,
|
50 |
+
"eval_loss": 2.077699899673462,
|
51 |
+
"eval_runtime": 78.9586,
|
52 |
+
"eval_samples_per_second": 15.274,
|
53 |
+
"eval_steps_per_second": 1.912,
|
54 |
+
"step": 1500
|
55 |
+
},
|
56 |
+
{
|
57 |
+
"epoch": 0.3683919690550746,
|
58 |
+
"grad_norm": 0.29397013783454895,
|
59 |
+
"learning_rate": 4.386013384908209e-05,
|
60 |
+
"loss": 0.6932,
|
61 |
+
"step": 2000
|
62 |
+
},
|
63 |
+
{
|
64 |
+
"epoch": 0.3683919690550746,
|
65 |
+
"eval_loss": 2.0721774101257324,
|
66 |
+
"eval_runtime": 77.349,
|
67 |
+
"eval_samples_per_second": 15.592,
|
68 |
+
"eval_steps_per_second": 1.952,
|
69 |
+
"step": 2000
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.46048996131884323,
|
73 |
+
"grad_norm": 0.9136261343955994,
|
74 |
+
"learning_rate": 4.232516731135262e-05,
|
75 |
+
"loss": 0.6907,
|
76 |
+
"step": 2500
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 0.46048996131884323,
|
80 |
+
"eval_loss": 2.079211711883545,
|
81 |
+
"eval_runtime": 79.3204,
|
82 |
+
"eval_samples_per_second": 15.204,
|
83 |
+
"eval_steps_per_second": 1.904,
|
84 |
+
"step": 2500
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 0.5525879535826119,
|
88 |
+
"grad_norm": 4.484213829040527,
|
89 |
+
"learning_rate": 4.079020077362314e-05,
|
90 |
+
"loss": 0.7043,
|
91 |
+
"step": 3000
|
92 |
+
},
|
93 |
+
{
|
94 |
+
"epoch": 0.5525879535826119,
|
95 |
+
"eval_loss": 2.0204577445983887,
|
96 |
+
"eval_runtime": 78.612,
|
97 |
+
"eval_samples_per_second": 15.341,
|
98 |
+
"eval_steps_per_second": 1.921,
|
99 |
+
"step": 3000
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 0.6446859458463805,
|
103 |
+
"grad_norm": 1.1021015644073486,
|
104 |
+
"learning_rate": 3.9255234235893664e-05,
|
105 |
+
"loss": 0.7078,
|
106 |
+
"step": 3500
|
107 |
+
},
|
108 |
+
{
|
109 |
+
"epoch": 0.6446859458463805,
|
110 |
+
"eval_loss": 2.1256344318389893,
|
111 |
+
"eval_runtime": 78.63,
|
112 |
+
"eval_samples_per_second": 15.338,
|
113 |
+
"eval_steps_per_second": 1.92,
|
114 |
+
"step": 3500
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.7367839381101492,
|
118 |
+
"grad_norm": 16.864044189453125,
|
119 |
+
"learning_rate": 3.772026769816418e-05,
|
120 |
+
"loss": 0.7034,
|
121 |
+
"step": 4000
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.7367839381101492,
|
125 |
+
"eval_loss": 2.0000314712524414,
|
126 |
+
"eval_runtime": 78.6543,
|
127 |
+
"eval_samples_per_second": 15.333,
|
128 |
+
"eval_steps_per_second": 1.92,
|
129 |
+
"step": 4000
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.8288819303739179,
|
133 |
+
"grad_norm": 20.071943283081055,
|
134 |
+
"learning_rate": 3.6185301160434705e-05,
|
135 |
+
"loss": 0.6671,
|
136 |
+
"step": 4500
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 0.8288819303739179,
|
140 |
+
"eval_loss": 2.1607539653778076,
|
141 |
+
"eval_runtime": 78.6404,
|
142 |
+
"eval_samples_per_second": 15.336,
|
143 |
+
"eval_steps_per_second": 1.92,
|
144 |
+
"step": 4500
|
145 |
+
}
|
146 |
+
],
|
147 |
+
"logging_steps": 500,
|
148 |
+
"max_steps": 16287,
|
149 |
+
"num_input_tokens_seen": 0,
|
150 |
+
"num_train_epochs": 3,
|
151 |
+
"save_steps": 500,
|
152 |
+
"stateful_callbacks": {
|
153 |
+
"TrainerControl": {
|
154 |
+
"args": {
|
155 |
+
"should_epoch_stop": false,
|
156 |
+
"should_evaluate": false,
|
157 |
+
"should_log": false,
|
158 |
+
"should_save": true,
|
159 |
+
"should_training_stop": false
|
160 |
+
},
|
161 |
+
"attributes": {}
|
162 |
+
}
|
163 |
+
},
|
164 |
+
"total_flos": 1619632097372520.0,
|
165 |
+
"train_batch_size": 2,
|
166 |
+
"trial_name": null,
|
167 |
+
"trial_params": null
|
168 |
+
}
|
checkpoint-4500/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d9885fd590ddfbffc4f8db1738c1993002dfa8c1603f7992a1cda9b93bbe7d34
|
3 |
+
size 4847
|