Joosep Pata
add v2.1.0 CMS model
67a8ea1
raw
history blame
5.51 kB
backend: pytorch
checkpoint_freq: 1
comet: false
comet_name: particleflow-pt
comet_offline: false
comet_step_freq: 10
config: parameters/pytorch/pyg-cms.yaml
conv_type: attention
data_dir: /scratch/project_465000301/tensorflow_datasets
dataset: cms
dtype: bfloat16
gpu_batch_multiplier: 5
gpus: 8
load: experiments/pyg-cms_20241101_090645_682892/checkpoints/checkpoint-06-3.089034.pth
lr: 0.0001
lr_schedule: cosinedecay
lr_schedule_config:
onecycle:
pct_start: 0.3
model:
attention:
activation: relu
attention_type: flash
conv_type: attention
dropout_conv_id_ff: 0.0
dropout_conv_id_mha: 0.0
dropout_conv_reg_ff: 0.0
dropout_conv_reg_mha: 0.0
dropout_ff: 0.0
head_dim: 32
num_convs: 6
num_heads: 32
use_pre_layernorm: true
cos_phi_mode: linear
energy_mode: direct-elemtype-split
eta_mode: linear
gnn_lsh:
activation: elu
bin_size: 320
conv_type: gnn_lsh
distance_dim: 128
dropout_ff: 0.0
embedding_dim: 512
ffn_dist_hidden_dim: 128
ffn_dist_num_layers: 2
layernorm: true
max_num_bins: 200
num_convs: 8
num_node_messages: 2
width: 512
input_encoding: split
learned_representation_mode: last
mamba:
activation: elu
conv_type: mamba
d_conv: 4
d_state: 32
dropout_ff: 0.0
embedding_dim: 1024
expand: 2
num_convs: 4
width: 1024
pt_mode: direct-elemtype-split
sin_phi_mode: linear
trainable: all
ntest: null
ntrain: null
num_epochs: 100
num_workers: 8
nvalid: null
patience: 20
prefetch_factor: 50
ray_train: false
raytune:
asha:
brackets: 1
grace_period: 10
max_t: 200
reduction_factor: 4
default_metric: val_loss
default_mode: min
hyperband:
max_t: 200
reduction_factor: 4
hyperopt:
n_random_steps: 10
local_dir: null
nevergrad:
n_random_steps: 10
sched: asha
search_alg: hyperopt
save_attention: false
sort_data: true
test_dataset:
cms_pf_qcd:
splits:
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
version: 2.5.0
cms_pf_qcd_nopu:
splits:
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
version: 2.5.0
cms_pf_ttbar:
splits:
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
version: 2.5.0
cms_pf_ttbar_nopu:
splits:
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
version: 2.5.0
cms_pf_ztt:
splits:
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
version: 2.5.0
cms_pf_ztt_nopu:
splits:
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
version: 2.5.0
test_datasets: []
train: true
train_dataset:
cms:
physical_nopu:
batch_size: 8
samples:
cms_pf_qcd_nopu:
splits:
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
version: 2.5.0
cms_pf_ttbar_nopu:
splits:
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
version: 2.5.0
cms_pf_ztt_nopu:
splits:
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
version: 2.5.0
physical_pu:
batch_size: 1
samples:
cms_pf_qcd:
splits:
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
version: 2.5.0
cms_pf_ttbar:
splits:
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
version: 2.5.0
cms_pf_ztt:
splits:
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
version: 2.5.0
val_freq: null
valid_dataset:
cms:
physical_nopu:
batch_size: 8
samples:
cms_pf_qcd_nopu:
splits:
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
version: 2.5.0
cms_pf_ttbar_nopu:
splits:
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
version: 2.5.0
cms_pf_ztt_nopu:
splits:
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
version: 2.5.0
physical_pu:
batch_size: 1
samples:
cms_pf_qcd:
splits:
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
version: 2.5.0
cms_pf_ttbar:
splits:
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
version: 2.5.0
cms_pf_ztt:
splits:
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
version: 2.5.0