profile
viewpoint
If you are wondering where the data of this site comes from, please visit https://api.github.com/users/RMeli/events. GitMemory does not store any data, but only uses NGINX to cache data for a period of time. The idea behind GitMemory is simply to give users a better reading experience.
Rocco Meli RMeli University of Oxford Working from Home rmeli.me

RMeli/Hartree-Fock 15

Solution of Hartree-Fock equations within Pople's STO-3G basis set

RMeli/irc 11

Transfrormation between Cartesian coordinates and redundant internal coordinates

RMeli/gsoc19 4

☀️ CNN Scoring for Flexible Docking - Google Summer of Code 2019 ☀️

dkoes/GNINA-1.0 3

Paper for release

RMeli/aescore 2

Learning Protein-Ligand Affinity with Atomic Environment Vectors

RMeli/CubeTools 2

Small toolbox to manipulate Gaussian Cube files

RMeli/applied-dl 1

A collection of materials for a new course being taught in Spring 2019.

RMeli/BertelsmannAI 1

🤖 Bertelsmann Technology Scholarship - Udacity AI Track 🤖

RMeli/code-snippets 1

Code snippets for different tasks and projects

RMeli/libmolgrid 1

Comprehensive library for fast, GPU accelerated molecular gridding for deep learning workflows

push eventRMeli/gnina-torch

Rocco Meli

commit sha e8361f9065efe87daa165945c36d4ece23620f65

better handling of --no_shuffle parameter

view details

push time in 9 hours

create barnchRMeli/gnina-torch

branch : training-simple

created branch time in 9 hours

startedRunningtarrens/MagneticCorne

started time in 14 hours

starteddivelab/DIG

started time in 15 hours

delete branch RMeli/gnina-torch

delete branch : dense

delete time in a day

push eventRMeli/gnina-torch

Rocco Meli

commit sha 733697c37f9c87f1f3ec59930cfd6661436bad72

dense block

view details

Rocco Meli

commit sha 7e8c06e35123afbff029f6b05f07e6c0edbd4dd7

dense model

view details

Rocco Meli

commit sha f4b0b88ae32a00392096d51846a92985bc333e41

add missing ReLU and move global pooling to model definition

view details

Rocco Meli

commit sha 25ab4f9617b0b7e8368cafe72893e854edc20953

fix codecov badge and remove MolSSI code

view details

Rocco Meli

commit sha dbdc9893af2b17da0167459400667696b868a1fb

reduce test resources for CI

view details

Rocco Meli

commit sha 5a1ca1cf0f3be545d71143bac4aef27972805e22

Merge pull request #5 from RMeli/dense Dense

view details

push time in a day

PR merged RMeli/gnina-torch

Dense enhancement

dense model architecture.

Dense(
  (features): Sequential(
    (data_enc_init_pool): MaxPool3d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
    (data_enc_init_conv): Conv3d(32, 32, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
    (data_enc_init_conv_relu): ReLU()
    (dense_block_0): DenseBlock(
      (blocks): ModuleList(
        (0): Sequential(
          (data_enc_level0_batchnorm_conv0): BatchNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level0_conv0): Conv3d(32, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level0_conv0_relu): ReLU()
        )
        (1): Sequential(
          (data_enc_level0_batchnorm_conv1): BatchNorm3d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level0_conv1): Conv3d(48, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level0_conv1_relu): ReLU()
        )
        (2): Sequential(
          (data_enc_level0_batchnorm_conv2): BatchNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level0_conv2): Conv3d(64, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level0_conv2_relu): ReLU()
        )
        (3): Sequential(
          (data_enc_level0_batchnorm_conv3): BatchNorm3d(80, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level0_conv3): Conv3d(80, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level0_conv3_relu): ReLU()
        )
      )
    )
    (data_enc_level0_bottleneck): Conv3d(96, 96, kernel_size=(1, 1, 1), stride=(1, 1, 1))
    (data_enc_level0_bottleneck_relu): ReLU()
    (data_enc_level1_pool): MaxPool3d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
    (dense_block_1): DenseBlock(
      (blocks): ModuleList(
        (0): Sequential(
          (data_enc_level1_batchnorm_conv0): BatchNorm3d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level1_conv0): Conv3d(96, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level1_conv0_relu): ReLU()
        )
        (1): Sequential(
          (data_enc_level1_batchnorm_conv1): BatchNorm3d(112, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level1_conv1): Conv3d(112, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level1_conv1_relu): ReLU()
        )
        (2): Sequential(
          (data_enc_level1_batchnorm_conv2): BatchNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level1_conv2): Conv3d(128, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level1_conv2_relu): ReLU()
        )
        (3): Sequential(
          (data_enc_level1_batchnorm_conv3): BatchNorm3d(144, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level1_conv3): Conv3d(144, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level1_conv3_relu): ReLU()
        )
      )
    )
    (data_enc_level1_bottleneck): Conv3d(160, 160, kernel_size=(1, 1, 1), stride=(1, 1, 1))
    (data_enc_level1_bottleneck_relu): ReLU()
    (data_enc_level2_pool): MaxPool3d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
    (dense_block_2): DenseBlock(
      (blocks): ModuleList(
        (0): Sequential(
          (data_enc_level2_batchnorm_conv0): BatchNorm3d(160, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level2_conv0): Conv3d(160, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level2_conv0_relu): ReLU()
        )
        (1): Sequential(
          (data_enc_level2_batchnorm_conv1): BatchNorm3d(176, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level2_conv1): Conv3d(176, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level2_conv1_relu): ReLU()
        )
        (2): Sequential(
          (data_enc_level2_batchnorm_conv2): BatchNorm3d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level2_conv2): Conv3d(192, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level2_conv2_relu): ReLU()
        )
        (3): Sequential(
          (data_enc_level2_batchnorm_conv3): BatchNorm3d(208, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level2_conv3): Conv3d(208, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level2_conv3_relu): ReLU()
        )
      )
    )
    (data_enc_level2_global_pool): MaxPool3d(kernel_size=(6, 6, 6), stride=(6, 6, 6), padding=0, dilation=1, ceil_mode=False)
  )
  (pose): Sequential(
    (pose_output): Linear(in_features=224, out_features=2, bias=True)
  )
  (affinity): Sequential(
    (affinity_output): Linear(in_features=224, out_features=1, bias=True)
  )
)
+307 -39

1 comment

4 changed files

RMeli

pr closed time in a day

push eventRMeli/gnina-torch

Rocco Meli

commit sha dbdc9893af2b17da0167459400667696b868a1fb

reduce test resources for CI

view details

push time in a day

push eventRMeli/gnina-torch

Rocco Meli

commit sha 25ab4f9617b0b7e8368cafe72893e854edc20953

fix codecov badge and remove MolSSI code

view details

push time in 2 days

PR opened RMeli/gnina-torch

Dense enhancement

dense model architecture.

Dense(
  (features): Sequential(
    (data_enc_init_pool): MaxPool3d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
    (data_enc_init_conv): Conv3d(32, 32, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
    (data_enc_init_conv_relu): ReLU()
    (dense_block_0): DenseBlock(
      (blocks): ModuleList(
        (0): Sequential(
          (data_enc_level0_batchnorm_conv0): BatchNorm3d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level0_conv0): Conv3d(32, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level0_conv0_relu): ReLU()
        )
        (1): Sequential(
          (data_enc_level0_batchnorm_conv1): BatchNorm3d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level0_conv1): Conv3d(48, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level0_conv1_relu): ReLU()
        )
        (2): Sequential(
          (data_enc_level0_batchnorm_conv2): BatchNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level0_conv2): Conv3d(64, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level0_conv2_relu): ReLU()
        )
        (3): Sequential(
          (data_enc_level0_batchnorm_conv3): BatchNorm3d(80, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level0_conv3): Conv3d(80, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level0_conv3_relu): ReLU()
        )
      )
    )
    (data_enc_level0_bottleneck): Conv3d(96, 96, kernel_size=(1, 1, 1), stride=(1, 1, 1))
    (data_enc_level0_bottleneck_relu): ReLU()
    (data_enc_level1_pool): MaxPool3d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
    (dense_block_1): DenseBlock(
      (blocks): ModuleList(
        (0): Sequential(
          (data_enc_level1_batchnorm_conv0): BatchNorm3d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level1_conv0): Conv3d(96, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level1_conv0_relu): ReLU()
        )
        (1): Sequential(
          (data_enc_level1_batchnorm_conv1): BatchNorm3d(112, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level1_conv1): Conv3d(112, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level1_conv1_relu): ReLU()
        )
        (2): Sequential(
          (data_enc_level1_batchnorm_conv2): BatchNorm3d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level1_conv2): Conv3d(128, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level1_conv2_relu): ReLU()
        )
        (3): Sequential(
          (data_enc_level1_batchnorm_conv3): BatchNorm3d(144, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level1_conv3): Conv3d(144, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level1_conv3_relu): ReLU()
        )
      )
    )
    (data_enc_level1_bottleneck): Conv3d(160, 160, kernel_size=(1, 1, 1), stride=(1, 1, 1))
    (data_enc_level1_bottleneck_relu): ReLU()
    (data_enc_level2_pool): MaxPool3d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
    (dense_block_2): DenseBlock(
      (blocks): ModuleList(
        (0): Sequential(
          (data_enc_level2_batchnorm_conv0): BatchNorm3d(160, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level2_conv0): Conv3d(160, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level2_conv0_relu): ReLU()
        )
        (1): Sequential(
          (data_enc_level2_batchnorm_conv1): BatchNorm3d(176, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level2_conv1): Conv3d(176, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level2_conv1_relu): ReLU()
        )
        (2): Sequential(
          (data_enc_level2_batchnorm_conv2): BatchNorm3d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level2_conv2): Conv3d(192, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level2_conv2_relu): ReLU()
        )
        (3): Sequential(
          (data_enc_level2_batchnorm_conv3): BatchNorm3d(208, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (data_enc_level2_conv3): Conv3d(208, 16, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1))
          (data_enc_level2_conv3_relu): ReLU()
        )
      )
    )
    (data_enc_level2_global_pool): MaxPool3d(kernel_size=(6, 6, 6), stride=(6, 6, 6), padding=0, dilation=1, ceil_mode=False)
  )
  (pose): Sequential(
    (pose_output): Linear(in_features=224, out_features=2, bias=True)
  )
  (affinity): Sequential(
    (affinity_output): Linear(in_features=224, out_features=1, bias=True)
  )
)
+286 -7

0 comment

2 changed files

pr created time in 2 days

push eventRMeli/gnina-torch

Rocco Meli

commit sha f4b0b88ae32a00392096d51846a92985bc333e41

add missing ReLU and move global pooling to model definition

view details

push time in 2 days

push eventRMeli/gnina-torch

Rocco Meli

commit sha ff3a424c92d6bd25f37858e46b957805a0900c62

add pre-commit configuration file

view details

Rocco Meli

commit sha e2dc06b6f69b8de8ea711fe83cbb37cdf9e99f2d

Merge pull request #4 from RMeli/pre-commit Add pre-commit

view details

Rocco Meli

commit sha 733697c37f9c87f1f3ec59930cfd6661436bad72

dense block

view details

Rocco Meli

commit sha 7e8c06e35123afbff029f6b05f07e6c0edbd4dd7

dense model

view details

push time in 2 days

startedMDAnalysis/mdacli

started time in 2 days

create barnchRMeli/gnina-torch

branch : dense

created branch time in 2 days

delete branch RMeli/gnina-torch

delete branch : default2017

delete time in 2 days

delete branch RMeli/gnina-torch

delete branch : pre-commit

delete time in 2 days

push eventRMeli/gnina-torch

Rocco Meli

commit sha ff3a424c92d6bd25f37858e46b957805a0900c62

add pre-commit configuration file

view details

Rocco Meli

commit sha e2dc06b6f69b8de8ea711fe83cbb37cdf9e99f2d

Merge pull request #4 from RMeli/pre-commit Add pre-commit

view details

push time in 2 days

PR merged RMeli/gnina-torch

Add pre-commit enhancement

Add pre-commit configuration file.

+27 -0

1 comment

1 changed file

RMeli

pr closed time in 2 days

PR opened RMeli/gnina-torch

Add pre-commit

Add pre-commit configuration file.

+27 -0

0 comment

1 changed file

pr created time in 2 days

create barnchRMeli/gnina-torch

branch : pre-commit

created branch time in 2 days

delete branch RMeli/gnina-torch

delete branch : default2018

delete time in 2 days

push eventRMeli/gnina-torch

Rocco Meli

commit sha 343ed40466614995bbdbf979f90eaca652c4fbc0

default2018 model

view details

Rocco Meli

commit sha 108e3c89b6107021d026ee14254b4437fc11aacd

Merge pull request #3 from RMeli/default2018 Default2018 model

view details

push time in 2 days

PR merged RMeli/gnina-torch

Default2018 model
+176 -9

1 comment

4 changed files

RMeli

pr closed time in 2 days

PR opened RMeli/gnina-torch

Default2018 model
+176 -9

0 comment

4 changed files

pr created time in 2 days

push eventRMeli/gnina-torch

Rocco Meli

commit sha 343ed40466614995bbdbf979f90eaca652c4fbc0

default2018 model

view details

push time in 2 days

create barnchRMeli/gnina-torch

branch : default2018

created branch time in 2 days

issue commentgnina/scripts

Dependencies: Caffe

I think I added the has_rmsd to the wrong place.

I think you added it to the correct place (same place where has_affinity: true is located). However, it is not the only modification you need to do (depending on if/how you want to use the RMSD); I suggest to search for "rmsd" in one of the models that uses it.

The error

Check failed: ExactNumTopBlobs() == top.size()

is likely caused by the fact that you did not add a RMSD blob to the MolgrdiDataLayer: see this line

Besides, I'm also very interested in the Monte Carlo Chain Sampling methods.

You can find thee implementation of Monte Carlo sampling in GNINA in gninasrc/lib/monte_carlo.h.

If you are just interested in the general idea, any book on molecular simulation usually has a section on Monte Carlo methods:

  • Understanding Molecular Simulation (D. Frenkel and B. Smit)
  • Computer simulation of liquids (M. P. Allen and D. J. Tildesley)
  • Statistical Mechanics: Theory and Molecular Simulation (M. Tuckerman)
CHANG-Shaole

comment created time in 2 days

push eventRMeli/gnina-torch

Rocco Meli

commit sha 32b89f956b70bafae43f24449155e28a07918946

fix readme

view details

Rocco Meli

commit sha da4370a9dd065426785ba89e54d010736a5d53ca

define conda environment for gnina

view details

Rocco Meli

commit sha d11444b1204ed54e6267fbd6be9b17487accec07

test molgrid import

view details

Rocco Meli

commit sha 0ed9e82a7c8c1f29ca719cecf5d9c45904c25e3b

apply black formatting

view details

Rocco Meli

commit sha be35d05971ebe0c074209707f0d534037a046610

flake8 and isort

view details

Rocco Meli

commit sha ae2cb1c702ac320492c13bbb5a1f9848c3791dd5

mypy

view details

Rocco Meli

commit sha 00d1d6943a6b348bc6de56ee57369f8bd629484a

sketch default2017 and test forward pass

view details

Rocco Meli

commit sha 69521fffa028fd277630c4875c76ec76959ef9a4

Merge pull request #2 from RMeli/default2017

view details

push time in 2 days

PR merged RMeli/gnina-torch

Default2017 model enhancement
+546 -278

1 comment

14 changed files

RMeli

pr closed time in 2 days

PR opened RMeli/gnina-torch

Default2017 model
+546 -278

0 comment

14 changed files

pr created time in 2 days