Spaces:
Runtime error
Runtime error
entry
Browse files- configs/dataset_configuration.json +9 -0
- configs/equations_ops_3_5000.json +0 -0
- configs/metadata.h5 +3 -0
- configs/nsr_network_config.yaml +103 -0
- configs/nsrwh_network_config.yaml +103 -0
configs/dataset_configuration.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"max_len": 20,
|
3 |
+
"operators": "add:10,mul:10,sub:5,div:5,sqrt:4,pow2:4,pow3:2,pow4:1,pow5:1,ln:1,exp:4,sin:4,cos:4,tan:4,asin:2",
|
4 |
+
"max_ops": 6,
|
5 |
+
"rewrite_functions": "simplify",
|
6 |
+
"variables": ["x_1","x_2","x_3","x_4","x_5"],
|
7 |
+
"eos_index": 1,
|
8 |
+
"pad_index": 0
|
9 |
+
}
|
configs/equations_ops_3_5000.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
configs/metadata.h5
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:45b86914f1a300d989db9e26050dba773c1293d3b990caf1140d097c61d8a25d
|
3 |
+
size 3956
|
configs/nsr_network_config.yaml
ADDED
@@ -0,0 +1,103 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
train_path: /scratch/tbe/nesymres/raw_datasets/200000000
|
2 |
+
val_path: data/validation
|
3 |
+
raw_test_path: ???
|
4 |
+
max_description_seq_len: 500
|
5 |
+
checkpointing: true
|
6 |
+
number_of_test: 20
|
7 |
+
val_check_interval: 5
|
8 |
+
name: results
|
9 |
+
benchmark_path: test_set/nc.csv
|
10 |
+
model_path: run/False/2022-11-07/13-46-03/Exp_weights/1000000_log_-epoch=104-val_loss=0.00.ckpt
|
11 |
+
test_path: data/validation/nc
|
12 |
+
resume_from_checkpoint: ''
|
13 |
+
wandb: false
|
14 |
+
num_of_workers: 5
|
15 |
+
batch_size: 400
|
16 |
+
num_sanity_val_steps: 0
|
17 |
+
epochs: 10000
|
18 |
+
precision: 16
|
19 |
+
gpu: 3
|
20 |
+
path_to_candidate: data/conditioning/equations_ops_3_5000.json
|
21 |
+
dataset:
|
22 |
+
len: 1000000
|
23 |
+
total_variables: null
|
24 |
+
total_coefficients: null
|
25 |
+
max_number_of_points: 1000
|
26 |
+
type_of_sampling_points: uniform
|
27 |
+
noise: false
|
28 |
+
fun_support:
|
29 |
+
max: 10
|
30 |
+
min: -10
|
31 |
+
min_len: 1
|
32 |
+
constants:
|
33 |
+
enabled: true
|
34 |
+
num_constants: 6
|
35 |
+
additive:
|
36 |
+
max: 10
|
37 |
+
min: -10
|
38 |
+
multiplicative:
|
39 |
+
max: 10
|
40 |
+
min: 0.05
|
41 |
+
number_of_complexity_classes: 30
|
42 |
+
conditioning:
|
43 |
+
mode: false
|
44 |
+
name: train
|
45 |
+
prob_symmetry: 0.2
|
46 |
+
prob_complexity: 0.3
|
47 |
+
prob_noise: 0
|
48 |
+
positive:
|
49 |
+
prob: 0.3
|
50 |
+
min_percent: 0
|
51 |
+
max_percent: 1
|
52 |
+
prob_pointers: 0.15
|
53 |
+
negative:
|
54 |
+
prob: 0.3
|
55 |
+
min_percent: 0
|
56 |
+
max_percent: 1
|
57 |
+
architecture:
|
58 |
+
sinuisodal_embeddings: false
|
59 |
+
dec_pf_dim: 512
|
60 |
+
dec_layers: 5
|
61 |
+
dim_hidden: 512
|
62 |
+
lr: 0.0001
|
63 |
+
dropout: 0
|
64 |
+
cond_num_layers: 3
|
65 |
+
num_features: 32
|
66 |
+
ln: true
|
67 |
+
N_p: 0
|
68 |
+
num_inds: 100
|
69 |
+
activation: relu
|
70 |
+
bit16: true
|
71 |
+
norm: true
|
72 |
+
linear: false
|
73 |
+
input_normalization: false
|
74 |
+
src_pad_idx: 0
|
75 |
+
trg_pad_idx: 0
|
76 |
+
length_eq: 60
|
77 |
+
n_l_enc: 5
|
78 |
+
mean: 0.5
|
79 |
+
std: 0.5
|
80 |
+
dim_input: 6
|
81 |
+
num_heads: 8
|
82 |
+
number_possible_tokens: 80
|
83 |
+
num_tokens_condition: 150
|
84 |
+
embedding_dim_condition: 512
|
85 |
+
conditioning: false
|
86 |
+
concat: true
|
87 |
+
predict_constants: c
|
88 |
+
wupsteps: 4000
|
89 |
+
inference:
|
90 |
+
beam_size: 10
|
91 |
+
word2id: ??
|
92 |
+
id2word: ??
|
93 |
+
total_variables: ??
|
94 |
+
n_jobs: 1
|
95 |
+
bfgs:
|
96 |
+
activated: false
|
97 |
+
not_activated_no_fit: true
|
98 |
+
n_restarts: 10
|
99 |
+
add_coefficients_if_not_existing: false
|
100 |
+
normalization_o: false
|
101 |
+
idx_remove: true
|
102 |
+
normalization_type: MSE
|
103 |
+
stop_time: 1000000000.0
|
configs/nsrwh_network_config.yaml
ADDED
@@ -0,0 +1,103 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
train_path: /scratch/tbe/nesymres/raw_datasets/200000000
|
2 |
+
val_path: data/validation
|
3 |
+
raw_test_path: ???
|
4 |
+
max_description_seq_len: 500
|
5 |
+
checkpointing: true
|
6 |
+
number_of_test: 20
|
7 |
+
val_check_interval: 5
|
8 |
+
name: results
|
9 |
+
benchmark_path: test_set/nc.csv
|
10 |
+
model_path: run/False/2022-11-07/13-46-03/Exp_weights/1000000_log_-epoch=104-val_loss=0.00.ckpt
|
11 |
+
test_path: data/validation/nc
|
12 |
+
resume_from_checkpoint: ''
|
13 |
+
wandb: false
|
14 |
+
num_of_workers: 4
|
15 |
+
batch_size: 400
|
16 |
+
num_sanity_val_steps: 0
|
17 |
+
epochs: 10000
|
18 |
+
precision: 16
|
19 |
+
gpu: 3
|
20 |
+
path_to_candidate: configs/equations_ops_3_5000.json
|
21 |
+
dataset:
|
22 |
+
len: 1000000
|
23 |
+
total_variables: null
|
24 |
+
total_coefficients: null
|
25 |
+
max_number_of_points: 1000
|
26 |
+
type_of_sampling_points: uniform
|
27 |
+
noise: false
|
28 |
+
fun_support:
|
29 |
+
max: 10
|
30 |
+
min: -10
|
31 |
+
min_len: 1
|
32 |
+
constants:
|
33 |
+
enabled: true
|
34 |
+
num_constants: 6
|
35 |
+
additive:
|
36 |
+
max: 10
|
37 |
+
min: -10
|
38 |
+
multiplicative:
|
39 |
+
max: 10
|
40 |
+
min: 0.05
|
41 |
+
number_of_complexity_classes: 30
|
42 |
+
conditioning:
|
43 |
+
mode: true
|
44 |
+
name: train
|
45 |
+
prob_symmetry: 0.2
|
46 |
+
prob_complexity: 0.3
|
47 |
+
prob_noise: 0
|
48 |
+
positive:
|
49 |
+
prob: 0.3
|
50 |
+
min_percent: 0
|
51 |
+
max_percent: 1
|
52 |
+
prob_pointers: 0.15
|
53 |
+
negative:
|
54 |
+
prob: 0.3
|
55 |
+
min_percent: 0
|
56 |
+
max_percent: 1
|
57 |
+
architecture:
|
58 |
+
sinuisodal_embeddings: false
|
59 |
+
dec_pf_dim: 512
|
60 |
+
dec_layers: 5
|
61 |
+
dim_hidden: 512
|
62 |
+
lr: 0.0001
|
63 |
+
dropout: 0
|
64 |
+
cond_num_layers: 3
|
65 |
+
num_features: 32
|
66 |
+
ln: true
|
67 |
+
N_p: 0
|
68 |
+
num_inds: 100
|
69 |
+
activation: relu
|
70 |
+
bit16: true
|
71 |
+
norm: true
|
72 |
+
linear: false
|
73 |
+
input_normalization: false
|
74 |
+
src_pad_idx: 0
|
75 |
+
trg_pad_idx: 0
|
76 |
+
length_eq: 60
|
77 |
+
n_l_enc: 5
|
78 |
+
mean: 0.5
|
79 |
+
std: 0.5
|
80 |
+
dim_input: 6
|
81 |
+
num_heads: 8
|
82 |
+
number_possible_tokens: 80
|
83 |
+
num_tokens_condition: 150
|
84 |
+
embedding_dim_condition: 512
|
85 |
+
conditioning: v3
|
86 |
+
concat: true
|
87 |
+
predict_constants: c
|
88 |
+
wupsteps: 4000
|
89 |
+
inference:
|
90 |
+
beam_size: 10
|
91 |
+
word2id: ??
|
92 |
+
id2word: ??
|
93 |
+
total_variables: ??
|
94 |
+
n_jobs: 1
|
95 |
+
bfgs:
|
96 |
+
activated: false
|
97 |
+
not_activated_no_fit: true
|
98 |
+
n_restarts: 10
|
99 |
+
add_coefficients_if_not_existing: false
|
100 |
+
normalization_o: false
|
101 |
+
idx_remove: true
|
102 |
+
normalization_type: MSE
|
103 |
+
stop_time: 1000000000.0
|