Hugging Face's logo

adams-story
/
vq_resnet

Model card Files Community
vq_resnet
File size: 3,905 Bytes 
3b35452
 
 
ca01c55
7708713
 
 
 
75b75fd
7708713
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
75b75fd
5457ef6
9d33806
 
 
 
 
5457ef6
 
 
 
 
9d33806
 
 
 
 
 
5457ef6
 
 
 
 
 
 
 
9d33806
5457ef6
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9d33806
5457ef6
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9d33806
 
 
5457ef6
9d33806
5457ef6
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
 
---
license: apache-2.0
---
epoch2-val-63.ckpt
* final val acc .63
* trained for 2 epochs
* More compressed embedding space, with more dropout
* git commit dc54a9bdbfcfbc83c736ac5c06ab09c5acf2d5e8
```
# lightning.pytorch==2.0.0
seed_everything: true
trainer:
  callbacks:
    - class_path: lightning.pytorch.callbacks.ModelCheckpoint
      init_args:
        save_last: true
        save_top_k: 1
        monitor: v_c_loss
  accelerator: auto
  strategy: auto
  devices: auto
  num_nodes: 1
  precision: 16-mixed
  logger: null
  callbacks: null
  fast_dev_run: false
  max_epochs: 10
  min_epochs: null
  max_steps: -1
  min_steps: null
  max_time: null
  limit_train_batches: null
  limit_val_batches: null
  limit_test_batches: null
  limit_predict_batches: null
  overfit_batches: 0.0
  val_check_interval: 0.1
  check_val_every_n_epoch: 1
  num_sanity_val_steps: null
  log_every_n_steps: 5
  enable_checkpointing: true
  enable_progress_bar: null
  enable_model_summary: null
  accumulate_grad_batches: 1
  gradient_clip_val: 0.5
  gradient_clip_algorithm: null
  deterministic: null
  benchmark: null
  inference_mode: true
  use_distributed_sampler: true
  profiler: null
  detect_anomaly: false
  barebones: false
  plugins: null
  sync_batchnorm: false
  reload_dataloaders_every_n_epochs: 0
  default_root_dir: ckpt/insert_at_4
model:
  resnet_type: 34
  is_rq: true
  quantizer_args:
    num_quantizers: 8
    shared_codebook: true
    quantize_dropout: false
    accept_image_fmap: true
    codebook_dim: 128
    codebook_size: 64
    decay: 0.8
    eps: 1.0e-05
    commitment_weight: 5.0
    threshold_ema_dead_code: 1
    sample_codebook_temp: 0.1
  resnet_insertion_index: 4
  unfreeze_resnet_block_indeces:
    - 3
  unfreeze_fc: true
  lr: 0.0002
data:
  data_dir: "/home/figes/Downloads/ILSVRC2012_CLS-LOC/"
  image_size: 224
  num_workers: 6
  batch_size: 512
  shuffle: true
  pin_memory: true
  drop_last: false
```

### epoch=5-step=14765.ckpt
* trained for 5 1/2 epochs on imagenet, on top of resnet 34
* final validation accuracy: .66
* final training accuracy: 0.64
* git hash: `c4852331f9a40393b8ffd8b7b9a689d1ff6e1021`
* config:
```
# lightning.pytorch==2.0.0
seed_everything: true
trainer:
  callbacks:
    - class_path: lightning.pytorch.callbacks.ModelCheckpoint
      init_args:
        save_last: true
        save_top_k: 1
        monitor: v_c_loss
  accelerator: auto
  strategy: auto
  devices: auto
  num_nodes: 1
  precision: 16-mixed
  logger: null
  callbacks: null
  fast_dev_run: false
  max_epochs: 10
  min_epochs: null
  max_steps: -1
  min_steps: null
  max_time: null
  limit_train_batches: null
  limit_val_batches: null
  limit_test_batches: null
  limit_predict_batches: null
  overfit_batches: 0.0
  val_check_interval: 0.1
  check_val_every_n_epoch: 1
  num_sanity_val_steps: null
  log_every_n_steps: 5
  enable_checkpointing: true
  enable_progress_bar: null
  enable_model_summary: null
  accumulate_grad_batches: 1
  gradient_clip_val: 0.5
  gradient_clip_algorithm: null
  deterministic: null
  benchmark: null
  inference_mode: true
  use_distributed_sampler: true
  profiler: null
  detect_anomaly: false
  barebones: false
  plugins: null
  sync_batchnorm: false
  reload_dataloaders_every_n_epochs: 0
  default_root_dir: ckpt/test_insert_at_4
model:
  resnet_type: 34
  is_rq: true
  quantizer_args:
    num_quantizers: 4
    shared_codebook: true
    quantize_dropout: false
    accept_image_fmap: true
    codebook_dim: 128
    codebook_size: 256
    decay: 0.8
    eps: 1.0e-05
    commitment_weight: 5.0
    threshold_ema_dead_code: 1
    sample_codebook_temp: 0.0
  resnet_insertion_index: 4
  unfreeze_resnet_block_indeces:
    - 3
  unfreeze_fc: true
  lr: 0.0002
data:
  data_dir: "/home/figes/Downloads/ILSVRC2012_CLS-LOC/"
  image_size: 224
  num_workers: 8
  batch_size: 512
  shuffle: true
  pin_memory: true
  drop_last: false
```