ribesstefano commited on
Commit
ce90119
·
verified ·
1 Parent(s): 3dd7d77

Training in progress, step 5000, checkpoint

Browse files
last-checkpoint/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:156e2fd606378633bf307efe0df5e2bd2ab80f15220bd9fe01a693ed009aaaee
3
  size 409608164
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:daab2685c7d4d42bb5b182b38a51b023c9b2e6d344f09a2ed6796eed20ab1b48
3
  size 409608164
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:461125dda3eba6ffa10f1372d61133904450ef356926735e5dbc819f685ea332
3
  size 814647162
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d2c13dc0447209c48c887895a95ae8f34be534f2f5ac4affce2bd1fae4ed9d62
3
  size 814647162
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f05568c9474927fa3f00b14190be9bec9b14406698bef27ced0d260778a147e1
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66a8f21c743fbbd6b01ebeadc5a65d9e0ab995588cfa53a2a2d84cf86cd007b5
3
  size 14244
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:608bb834d22a938d2d06a70078475dfb0f8514808f1b5691add9ba9f97ccc014
3
  size 1192
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d65c0dc8938f45175fc83c4b1c91e571df9245368b510bc2d6e3a799c35da4f7
3
  size 1192
last-checkpoint/trainer_state.json CHANGED
@@ -1,278 +1,124 @@
1
  {
2
  "best_metric": 0.0,
3
- "best_model_checkpoint": "/mimer/NOBACKUP/groups/naiss2023-6-290/stefano/models//PROTAC-Splitter-EncoderDecoder-lr_reduce-opt25-rand-smiles/trial-number=6-learning_rate=4.9e-04-warmup_ratio=0.040-min_lr=0.000-factor=0.460/checkpoint-5000",
4
- "epoch": 0.9863878477017163,
5
  "eval_steps": 2500,
6
- "global_step": 10000,
7
  "is_hyper_param_search": true,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
  "epoch": 0.049319392385085814,
13
- "grad_norm": 0.6788614988327026,
14
- "learning_rate": 0.0004854655346216474,
15
- "loss": 3.5969,
16
  "step": 500
17
  },
18
  {
19
  "epoch": 0.09863878477017163,
20
- "grad_norm": 0.7310829758644104,
21
- "learning_rate": 0.0004854655346216474,
22
- "loss": 3.5675,
23
  "step": 1000
24
  },
25
  {
26
  "epoch": 0.14795817715525744,
27
- "grad_norm": 0.7336956858634949,
28
- "learning_rate": 0.0004854655346216474,
29
- "loss": 3.5712,
30
  "step": 1500
31
  },
32
  {
33
  "epoch": 0.19727756954034326,
34
- "grad_norm": 0.9253264665603638,
35
- "learning_rate": 0.0004854655346216474,
36
- "loss": 3.5666,
37
  "step": 2000
38
  },
39
  {
40
  "epoch": 0.24659696192542907,
41
- "grad_norm": 0.8935646414756775,
42
- "learning_rate": 0.0004854655346216474,
43
- "loss": 3.5663,
44
  "step": 2500
45
  },
46
  {
47
  "epoch": 0.2959163543105149,
48
- "grad_norm": 0.749767541885376,
49
- "learning_rate": 0.0004854655346216474,
50
- "loss": 3.5635,
51
  "step": 3000
52
  },
53
  {
54
  "epoch": 0.3452357466956007,
55
- "grad_norm": 0.9100434184074402,
56
- "learning_rate": 0.0004854655346216474,
57
- "loss": 3.5616,
58
  "step": 3500
59
  },
60
  {
61
  "epoch": 0.3945551390806865,
62
- "grad_norm": 0.9583754539489746,
63
- "learning_rate": 0.0004854655346216474,
64
- "loss": 3.5619,
65
  "step": 4000
66
  },
67
  {
68
  "epoch": 0.4438745314657723,
69
- "grad_norm": 1.0354633331298828,
70
- "learning_rate": 0.0004854655346216474,
71
- "loss": 3.5594,
72
  "step": 4500
73
  },
74
  {
75
  "epoch": 0.49319392385085814,
76
- "grad_norm": 1.0010348558425903,
77
- "learning_rate": 0.0004854655346216474,
78
- "loss": 3.5595,
79
  "step": 5000
80
  },
81
  {
82
  "epoch": 0.49319392385085814,
83
  "eval_all_ligands_equal": 0.0,
84
- "eval_e3_equal": 0.0,
85
  "eval_e3_graph_edit_distance": 9.999999999999999e+63,
86
  "eval_e3_graph_edit_distance_norm": 1.0,
87
- "eval_e3_has_attachment_point(s)": 0.0,
88
- "eval_e3_heavy_atoms_difference": 28.086433937251435,
89
- "eval_e3_heavy_atoms_difference_norm": 1.0,
90
  "eval_e3_tanimoto_similarity": 0.0,
91
- "eval_e3_valid": 0.0,
92
- "eval_has_all_attachment_points": 0.0,
93
- "eval_has_three_substructures": 0.0,
94
- "eval_heavy_atoms_difference": 75.67370746796288,
95
- "eval_heavy_atoms_difference_norm": 1.0,
96
- "eval_linker_equal": 0.0,
97
  "eval_linker_graph_edit_distance": 9.999999999999999e+63,
98
  "eval_linker_graph_edit_distance_norm": 1.0,
99
- "eval_linker_has_attachment_point(s)": 0.0,
100
- "eval_linker_heavy_atoms_difference": 16.430402121078213,
101
- "eval_linker_heavy_atoms_difference_norm": 1.0,
102
  "eval_linker_tanimoto_similarity": 0.0,
103
- "eval_linker_valid": 0.0,
104
- "eval_loss": 3.582303047180176,
105
- "eval_num_fragments": 1.0024745912505524,
106
- "eval_poi_equal": 0.0,
107
  "eval_poi_graph_edit_distance": 9.999999999999999e+63,
108
  "eval_poi_graph_edit_distance_norm": 1.0,
109
- "eval_poi_has_attachment_point(s)": 0.0,
110
- "eval_poi_heavy_atoms_difference": 31.15687140963323,
111
- "eval_poi_heavy_atoms_difference_norm": 1.0,
112
  "eval_poi_tanimoto_similarity": 0.0,
113
- "eval_poi_valid": 0.0,
114
  "eval_reassembly": 0.0,
115
  "eval_reassembly_nostereo": 0.0,
116
- "eval_runtime": 5688.6938,
117
- "eval_samples_per_second": 1.989,
118
- "eval_steps_per_second": 0.031,
119
  "eval_tanimoto_similarity": 0.0,
120
- "eval_valid": 0.0,
121
  "step": 5000
122
- },
123
- {
124
- "epoch": 0.542513316235944,
125
- "grad_norm": 1.046298623085022,
126
- "learning_rate": 0.0004854655346216474,
127
- "loss": 3.5577,
128
- "step": 5500
129
- },
130
- {
131
- "epoch": 0.5918327086210298,
132
- "grad_norm": 1.0882964134216309,
133
- "learning_rate": 0.0004854655346216474,
134
- "loss": 3.558,
135
- "step": 6000
136
- },
137
- {
138
- "epoch": 0.6411521010061156,
139
- "grad_norm": 1.111539602279663,
140
- "learning_rate": 0.0004854655346216474,
141
- "loss": 3.5567,
142
- "step": 6500
143
- },
144
- {
145
- "epoch": 0.6904714933912014,
146
- "grad_norm": 1.0729397535324097,
147
- "learning_rate": 0.0004854655346216474,
148
- "loss": 3.5565,
149
- "step": 7000
150
- },
151
- {
152
- "epoch": 0.7397908857762873,
153
- "grad_norm": 1.0593640804290771,
154
- "learning_rate": 0.0004854655346216474,
155
- "loss": 3.5555,
156
- "step": 7500
157
- },
158
- {
159
- "epoch": 0.7397908857762873,
160
- "eval_all_ligands_equal": 0.0,
161
- "eval_e3_equal": 0.0,
162
- "eval_e3_graph_edit_distance": 9.999999999999999e+63,
163
- "eval_e3_graph_edit_distance_norm": 1.0,
164
- "eval_e3_has_attachment_point(s)": 0.0,
165
- "eval_e3_heavy_atoms_difference": 28.086433937251435,
166
- "eval_e3_heavy_atoms_difference_norm": 1.0,
167
- "eval_e3_tanimoto_similarity": 0.0,
168
- "eval_e3_valid": 0.0,
169
- "eval_has_all_attachment_points": 0.0,
170
- "eval_has_three_substructures": 0.0,
171
- "eval_heavy_atoms_difference": 75.67370746796288,
172
- "eval_heavy_atoms_difference_norm": 1.0,
173
- "eval_linker_equal": 0.0,
174
- "eval_linker_graph_edit_distance": 9.999999999999999e+63,
175
- "eval_linker_graph_edit_distance_norm": 1.0,
176
- "eval_linker_has_attachment_point(s)": 0.0,
177
- "eval_linker_heavy_atoms_difference": 16.430402121078213,
178
- "eval_linker_heavy_atoms_difference_norm": 1.0,
179
- "eval_linker_tanimoto_similarity": 0.0,
180
- "eval_linker_valid": 0.0,
181
- "eval_loss": 3.5780556201934814,
182
- "eval_num_fragments": 1.0022978347326557,
183
- "eval_poi_equal": 0.0,
184
- "eval_poi_graph_edit_distance": 9.999999999999999e+63,
185
- "eval_poi_graph_edit_distance_norm": 1.0,
186
- "eval_poi_has_attachment_point(s)": 0.0,
187
- "eval_poi_heavy_atoms_difference": 31.15687140963323,
188
- "eval_poi_heavy_atoms_difference_norm": 1.0,
189
- "eval_poi_tanimoto_similarity": 0.0,
190
- "eval_poi_valid": 0.0,
191
- "eval_reassembly": 0.0,
192
- "eval_reassembly_nostereo": 0.0,
193
- "eval_runtime": 5679.3065,
194
- "eval_samples_per_second": 1.992,
195
- "eval_steps_per_second": 0.031,
196
- "eval_tanimoto_similarity": 0.0,
197
- "eval_valid": 0.0,
198
- "step": 7500
199
- },
200
- {
201
- "epoch": 0.789110278161373,
202
- "grad_norm": 1.0324289798736572,
203
- "learning_rate": 0.0004854655346216474,
204
- "loss": 3.5564,
205
- "step": 8000
206
- },
207
- {
208
- "epoch": 0.8384296705464589,
209
- "grad_norm": 1.1101008653640747,
210
- "learning_rate": 0.0004854655346216474,
211
- "loss": 3.5561,
212
- "step": 8500
213
- },
214
- {
215
- "epoch": 0.8877490629315447,
216
- "grad_norm": 0.9862679839134216,
217
- "learning_rate": 0.0004854655346216474,
218
- "loss": 3.5558,
219
- "step": 9000
220
- },
221
- {
222
- "epoch": 0.9370684553166305,
223
- "grad_norm": 1.0627096891403198,
224
- "learning_rate": 0.0004854655346216474,
225
- "loss": 3.5552,
226
- "step": 9500
227
- },
228
- {
229
- "epoch": 0.9863878477017163,
230
- "grad_norm": 0.9944122433662415,
231
- "learning_rate": 0.0004854655346216474,
232
- "loss": 3.5559,
233
- "step": 10000
234
- },
235
- {
236
- "epoch": 0.9863878477017163,
237
- "eval_all_ligands_equal": 0.0,
238
- "eval_e3_equal": 0.0,
239
- "eval_e3_graph_edit_distance": 9.999999999999999e+63,
240
- "eval_e3_graph_edit_distance_norm": 1.0,
241
- "eval_e3_has_attachment_point(s)": 0.0,
242
- "eval_e3_heavy_atoms_difference": 28.086433937251435,
243
- "eval_e3_heavy_atoms_difference_norm": 1.0,
244
- "eval_e3_tanimoto_similarity": 0.0,
245
- "eval_e3_valid": 0.0,
246
- "eval_has_all_attachment_points": 0.0,
247
- "eval_has_three_substructures": 0.0,
248
- "eval_heavy_atoms_difference": 75.67370746796288,
249
- "eval_heavy_atoms_difference_norm": 1.0,
250
- "eval_linker_equal": 0.0,
251
- "eval_linker_graph_edit_distance": 9.999999999999999e+63,
252
- "eval_linker_graph_edit_distance_norm": 1.0,
253
- "eval_linker_has_attachment_point(s)": 0.0,
254
- "eval_linker_heavy_atoms_difference": 16.430402121078213,
255
- "eval_linker_heavy_atoms_difference_norm": 1.0,
256
- "eval_linker_tanimoto_similarity": 0.0,
257
- "eval_linker_valid": 0.0,
258
- "eval_loss": 3.57759952545166,
259
- "eval_num_fragments": 1.0024745912505524,
260
- "eval_poi_equal": 0.0,
261
- "eval_poi_graph_edit_distance": 9.999999999999999e+63,
262
- "eval_poi_graph_edit_distance_norm": 1.0,
263
- "eval_poi_has_attachment_point(s)": 0.0,
264
- "eval_poi_heavy_atoms_difference": 31.15687140963323,
265
- "eval_poi_heavy_atoms_difference_norm": 1.0,
266
- "eval_poi_tanimoto_similarity": 0.0,
267
- "eval_poi_valid": 0.0,
268
- "eval_reassembly": 0.0,
269
- "eval_reassembly_nostereo": 0.0,
270
- "eval_runtime": 5682.366,
271
- "eval_samples_per_second": 1.991,
272
- "eval_steps_per_second": 0.031,
273
- "eval_tanimoto_similarity": 0.0,
274
- "eval_valid": 0.0,
275
- "step": 10000
276
  }
277
  ],
278
  "logging_steps": 500,
@@ -287,18 +133,18 @@
287
  "should_evaluate": false,
288
  "should_log": false,
289
  "should_save": true,
290
- "should_training_stop": true
291
  },
292
  "attributes": {}
293
  }
294
  },
295
- "total_flos": 8.729673191023411e+16,
296
  "train_batch_size": 128,
297
- "trial_name": "trial-number=6-learning_rate=4.9e-04-warmup_ratio=0.040-min_lr=0.000-factor=0.460",
298
  "trial_params": {
299
- "factor": 0.45999999999999996,
300
- "learning_rate": 0.0004854655346216474,
301
- "min_lr": 5.158242893359211e-10,
302
- "warmup_ratio": 0.04
303
  }
304
  }
 
1
  {
2
  "best_metric": 0.0,
3
+ "best_model_checkpoint": "/mimer/NOBACKUP/groups/naiss2023-6-290/stefano/models//PROTAC-Splitter-EncoderDecoder-lr_reduce-opt25-rand-smiles/trial-number=7-learning_rate=1.8e-04-warmup_ratio=0.100-min_lr=0.000-factor=0.710/checkpoint-5000",
4
+ "epoch": 0.49319392385085814,
5
  "eval_steps": 2500,
6
+ "global_step": 5000,
7
  "is_hyper_param_search": true,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
  "epoch": 0.049319392385085814,
13
+ "grad_norm": 0.49629172682762146,
14
+ "learning_rate": 0.00017553249516694252,
15
+ "loss": 0.9567,
16
  "step": 500
17
  },
18
  {
19
  "epoch": 0.09863878477017163,
20
+ "grad_norm": 0.34096312522888184,
21
+ "learning_rate": 0.00017553249516694252,
22
+ "loss": 0.1992,
23
  "step": 1000
24
  },
25
  {
26
  "epoch": 0.14795817715525744,
27
+ "grad_norm": 0.24350646138191223,
28
+ "learning_rate": 0.00017553249516694252,
29
+ "loss": 0.1742,
30
  "step": 1500
31
  },
32
  {
33
  "epoch": 0.19727756954034326,
34
+ "grad_norm": 0.2776834964752197,
35
+ "learning_rate": 0.00017553249516694252,
36
+ "loss": 0.1627,
37
  "step": 2000
38
  },
39
  {
40
  "epoch": 0.24659696192542907,
41
+ "grad_norm": 0.2741040885448456,
42
+ "learning_rate": 0.00017553249516694252,
43
+ "loss": 0.1553,
44
  "step": 2500
45
  },
46
  {
47
  "epoch": 0.2959163543105149,
48
+ "grad_norm": 0.3086952269077301,
49
+ "learning_rate": 0.00017553249516694252,
50
+ "loss": 0.1516,
51
  "step": 3000
52
  },
53
  {
54
  "epoch": 0.3452357466956007,
55
+ "grad_norm": 0.38999202847480774,
56
+ "learning_rate": 0.00017553249516694252,
57
+ "loss": 0.1474,
58
  "step": 3500
59
  },
60
  {
61
  "epoch": 0.3945551390806865,
62
+ "grad_norm": 0.2551262378692627,
63
+ "learning_rate": 0.00017553249516694252,
64
+ "loss": 0.1454,
65
  "step": 4000
66
  },
67
  {
68
  "epoch": 0.4438745314657723,
69
+ "grad_norm": 0.2856215238571167,
70
+ "learning_rate": 0.00017553249516694252,
71
+ "loss": 0.1424,
72
  "step": 4500
73
  },
74
  {
75
  "epoch": 0.49319392385085814,
76
+ "grad_norm": 0.5204154253005981,
77
+ "learning_rate": 0.00017553249516694252,
78
+ "loss": 0.1431,
79
  "step": 5000
80
  },
81
  {
82
  "epoch": 0.49319392385085814,
83
  "eval_all_ligands_equal": 0.0,
84
+ "eval_e3_equal": 0.0346442775077331,
85
  "eval_e3_graph_edit_distance": 9.999999999999999e+63,
86
  "eval_e3_graph_edit_distance_norm": 1.0,
87
+ "eval_e3_has_attachment_point(s)": 1.0,
88
+ "eval_e3_heavy_atoms_difference": 1.5730446310207689,
89
+ "eval_e3_heavy_atoms_difference_norm": -0.0005120283470902502,
90
  "eval_e3_tanimoto_similarity": 0.0,
91
+ "eval_e3_valid": 1.0,
92
+ "eval_has_all_attachment_points": 1.0,
93
+ "eval_has_three_substructures": 1.0,
94
+ "eval_heavy_atoms_difference": 3.9468846663720725,
95
+ "eval_heavy_atoms_difference_norm": 0.023292187761076128,
96
+ "eval_linker_equal": 0.0023862129916040653,
97
  "eval_linker_graph_edit_distance": 9.999999999999999e+63,
98
  "eval_linker_graph_edit_distance_norm": 1.0,
99
+ "eval_linker_has_attachment_point(s)": 1.0,
100
+ "eval_linker_heavy_atoms_difference": 0.4876712328767123,
101
+ "eval_linker_heavy_atoms_difference_norm": -0.19550334215272724,
102
  "eval_linker_tanimoto_similarity": 0.0,
103
+ "eval_linker_valid": 1.0,
104
+ "eval_loss": 0.4921937882900238,
105
+ "eval_num_fragments": 3.0,
106
+ "eval_poi_equal": 0.003535130357931949,
107
  "eval_poi_graph_edit_distance": 9.999999999999999e+63,
108
  "eval_poi_graph_edit_distance_norm": 1.0,
109
+ "eval_poi_has_attachment_point(s)": 1.0,
110
+ "eval_poi_heavy_atoms_difference": 1.8861688024745913,
111
+ "eval_poi_heavy_atoms_difference_norm": -0.07049150153739635,
112
  "eval_poi_tanimoto_similarity": 0.0,
113
+ "eval_poi_valid": 1.0,
114
  "eval_reassembly": 0.0,
115
  "eval_reassembly_nostereo": 0.0,
116
+ "eval_runtime": 1323.3177,
117
+ "eval_samples_per_second": 8.55,
118
+ "eval_steps_per_second": 0.134,
119
  "eval_tanimoto_similarity": 0.0,
120
+ "eval_valid": 1.0,
121
  "step": 5000
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
122
  }
123
  ],
124
  "logging_steps": 500,
 
133
  "should_evaluate": false,
134
  "should_log": false,
135
  "should_save": true,
136
+ "should_training_stop": false
137
  },
138
  "attributes": {}
139
  }
140
  },
141
+ "total_flos": 4.364959997072179e+16,
142
  "train_batch_size": 128,
143
+ "trial_name": "trial-number=7-learning_rate=1.8e-04-warmup_ratio=0.100-min_lr=0.000-factor=0.710",
144
  "trial_params": {
145
+ "factor": 0.71,
146
+ "learning_rate": 0.00017553249516694252,
147
+ "min_lr": 1.836881026598859e-11,
148
+ "warmup_ratio": 0.09999999999999999
149
  }
150
  }
last-checkpoint/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b95041f6a010f2cf869dc8bc44f5902cbc00a43c6932510edd799e444f572a41
3
  size 7544
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e535304a50d6c7bf62265d8313d99249f4fa14351f55f318cc66b31cfb03d0c
3
  size 7544