gnodevae 0.0.3__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- GNODEVAE/CLEAR.py +172 -0
- GNODEVAE/GODEVAE.py +413 -0
- GNODEVAE/GODEVAE_agent.py +385 -0
- GNODEVAE/GODEVAE_env.py +758 -0
- GNODEVAE/GODEVAE_trainer.py +309 -0
- GNODEVAE/SC3.py +198 -0
- GNODEVAE/__init__.py +6 -0
- GNODEVAE/agent.py +472 -0
- GNODEVAE/env.py +474 -0
- GNODEVAE/mixin.py +418 -0
- GNODEVAE/module.py +939 -0
- GNODEVAE/module_refined.py +561 -0
- GNODEVAE/partition_time.py +185 -0
- GNODEVAE/scGCC.py +184 -0
- GNODEVAE/scGNN.py +146 -0
- GNODEVAE/trainer.py +397 -0
- GNODEVAE/utils.py +460 -0
- GNODEVAE/utils_ODE.py +140 -0
- gnodevae-0.0.3.dist-info/METADATA +330 -0
- gnodevae-0.0.3.dist-info/RECORD +23 -0
- gnodevae-0.0.3.dist-info/WHEEL +5 -0
- gnodevae-0.0.3.dist-info/licenses/LICENSE +21 -0
- gnodevae-0.0.3.dist-info/top_level.txt +1 -0
|
@@ -0,0 +1,330 @@
|
|
|
1
|
+
Metadata-Version: 2.4
|
|
2
|
+
Name: gnodevae
|
|
3
|
+
Version: 0.0.3
|
|
4
|
+
Summary: A Graph-Based ODE-VAE Enhances Clustering for Single-Cell Data
|
|
5
|
+
Author-email: Zeyu Fu <fuzeyu99@126.com>
|
|
6
|
+
License: MIT License
|
|
7
|
+
|
|
8
|
+
Copyright (c) 2025 Zeyu Fu
|
|
9
|
+
|
|
10
|
+
Permission is hereby granted, free of charge, to any person obtaining a copy
|
|
11
|
+
of this software and associated documentation files (the "Software"), to deal
|
|
12
|
+
in the Software without restriction, including without limitation the rights
|
|
13
|
+
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
|
14
|
+
copies of the Software, and to permit persons to whom the Software is
|
|
15
|
+
furnished to do so, subject to the following conditions:
|
|
16
|
+
|
|
17
|
+
The above copyright notice and this permission notice shall be included in all
|
|
18
|
+
copies or substantial portions of the Software.
|
|
19
|
+
|
|
20
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
|
21
|
+
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
|
22
|
+
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
|
23
|
+
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
|
24
|
+
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
|
25
|
+
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
|
|
26
|
+
SOFTWARE.
|
|
27
|
+
|
|
28
|
+
Project-URL: Homepage, https://github.com/PeterPonyu/GNODEVAE
|
|
29
|
+
Project-URL: Source, https://github.com/PeterPonyu/GNODEVAE
|
|
30
|
+
Project-URL: Issues, https://github.com/PeterPonyu/GNODEVAE/issues
|
|
31
|
+
Keywords: single-cell,graph-neural-network,neural-ode,variational-autoencoder,bioinformatics
|
|
32
|
+
Classifier: Development Status :: 3 - Alpha
|
|
33
|
+
Classifier: Programming Language :: Python :: 3
|
|
34
|
+
Classifier: Programming Language :: Python :: 3.9
|
|
35
|
+
Classifier: Programming Language :: Python :: 3.10
|
|
36
|
+
Classifier: Programming Language :: Python :: 3.11
|
|
37
|
+
Classifier: Programming Language :: Python :: 3.12
|
|
38
|
+
Classifier: License :: OSI Approved :: MIT License
|
|
39
|
+
Classifier: Intended Audience :: Science/Research
|
|
40
|
+
Classifier: Topic :: Scientific/Engineering :: Bio-Informatics
|
|
41
|
+
Requires-Python: >=3.9
|
|
42
|
+
Description-Content-Type: text/markdown
|
|
43
|
+
License-File: LICENSE
|
|
44
|
+
Requires-Dist: torch>=1.13.0
|
|
45
|
+
Requires-Dist: torch-geometric>=2.0.0
|
|
46
|
+
Requires-Dist: scanpy>=1.8.0
|
|
47
|
+
Requires-Dist: anndata>=0.8.0
|
|
48
|
+
Requires-Dist: numpy>=1.20.0
|
|
49
|
+
Requires-Dist: pandas>=1.3.0
|
|
50
|
+
Requires-Dist: scikit-learn>=1.0.0
|
|
51
|
+
Requires-Dist: tqdm>=4.60.0
|
|
52
|
+
Requires-Dist: psutil>=5.8.0
|
|
53
|
+
Requires-Dist: torchdiffeq>=0.2.0
|
|
54
|
+
Dynamic: license-file
|
|
55
|
+
|
|
56
|
+
# GNODEVAE: A Graph-Based ODE-VAE Enhances Clustering of Single-Cell Data
|
|
57
|
+
|
|
58
|
+
## Graphical Abstract
|
|
59
|
+
|
|
60
|
+

|
|
61
|
+
|
|
62
|
+
## Introduction
|
|
63
|
+
|
|
64
|
+
GNODEVAE is an innovative computational framework that integrates Graph Attention Networks (GAT), Neural Ordinary Differential Equations (NODE), and Variational Autoencoders (VAE). It addresses three critical challenges in single-cell RNA sequencing data analysis:
|
|
65
|
+
|
|
66
|
+
1. Capturing complex topological relationships between cells
|
|
67
|
+
2. Modeling continuous dynamic processes of cell differentiation
|
|
68
|
+
3. Handling high levels of technical noise and biological variation
|
|
69
|
+
|
|
70
|
+
This novel integration significantly improves the accurate identification of cell subpopulations, reconstruction of developmental trajectories, and understanding of cellular heterogeneity.
|
|
71
|
+
|
|
72
|
+
## Key Contributions
|
|
73
|
+
|
|
74
|
+
### 1. Dynamic Attention Weighting for Biological Significance
|
|
75
|
+
|
|
76
|
+
The GAT's attention mechanism adaptively weights gene expression profiles, prioritizing meaningful biological relationships while minimizing technical noise - particularly valuable for heterogeneous cell populations.
|
|
77
|
+
|
|
78
|
+
### 2. Continuous-Time Developmental Modeling via Neural ODEs
|
|
79
|
+
|
|
80
|
+
Integration of neural ordinary differential equations transforms static representations into dynamic systems, with time variables providing natural parameterization of developmental processes and enabling predictions at any point in cellular development.
|
|
81
|
+
|
|
82
|
+
### 3. Biologically Consistent Latent Space Representations
|
|
83
|
+
|
|
84
|
+
The model's latent space effectively captures biological phenomena like varying rates of cell differentiation, while attention weights align with established developmental relationships between cell types.
|
|
85
|
+
|
|
86
|
+
### 4. Comprehensive Benchmark Leadership
|
|
87
|
+
|
|
88
|
+
When compared with six advanced single-cell analysis methods (scVI, DIP-VAE, TC-VAE, β-VAE, Info-VAE, and scTour), GNODEVAE ranked first across all 13 test datasets, demonstrating exceptional versatility across diverse biological contexts.
|
|
89
|
+
|
|
90
|
+
### 5. Superior Gene Trend Analysis Performance
|
|
91
|
+
|
|
92
|
+
Quantitative evaluation shows GNODEVAE significantly outperforms existing methods (69.97% improvement over Palantir, 63.58% over Diffmap) in Calinski-Harabasz index, demonstrating clearer clustering and stronger category discrimination.
|
|
93
|
+
|
|
94
|
+
|
|
95
|
+
## Installation
|
|
96
|
+
|
|
97
|
+
### Prerequisites
|
|
98
|
+
|
|
99
|
+
- Python 3.8 or higher
|
|
100
|
+
- PyTorch 1.12 or higher (with CUDA support recommended for GPU acceleration)
|
|
101
|
+
- PyTorch Geometric
|
|
102
|
+
|
|
103
|
+
### Install from Source
|
|
104
|
+
|
|
105
|
+
```bash
|
|
106
|
+
# Clone the repository
|
|
107
|
+
git clone https://github.com/PeterPonyu/GNODEVAE.git
|
|
108
|
+
cd GNODEVAE
|
|
109
|
+
|
|
110
|
+
# Install dependencies
|
|
111
|
+
pip install torch torch-geometric scanpy anndata numpy pandas scikit-learn tqdm psutil torchdiffeq
|
|
112
|
+
```
|
|
113
|
+
|
|
114
|
+
### Dependencies
|
|
115
|
+
|
|
116
|
+
The main dependencies include:
|
|
117
|
+
- `torch` - PyTorch deep learning framework
|
|
118
|
+
- `torch-geometric` - Geometric deep learning extension for PyTorch
|
|
119
|
+
- `scanpy` - Single-cell analysis toolkit
|
|
120
|
+
- `anndata` - Annotated data structures for single-cell data
|
|
121
|
+
- `torchdiffeq` - Differentiable ODE solvers for PyTorch
|
|
122
|
+
- `numpy`, `pandas` - Data manipulation
|
|
123
|
+
- `scikit-learn` - Machine learning utilities
|
|
124
|
+
- `tqdm` - Progress bars
|
|
125
|
+
- `psutil` - System resource monitoring
|
|
126
|
+
|
|
127
|
+
## Quick Start
|
|
128
|
+
|
|
129
|
+
### Basic Usage
|
|
130
|
+
|
|
131
|
+
```python
|
|
132
|
+
import scanpy as sc
|
|
133
|
+
from GNODEVAE import agent_r # GraphVAE with refined architecture
|
|
134
|
+
# OR
|
|
135
|
+
from GNODEVAE import agent # Standard GraphVAE
|
|
136
|
+
# For full GNODEVAE with ODE support, use:
|
|
137
|
+
# from GNODEVAE.GODEVAE_agent import GNODEVAE_agent_r
|
|
138
|
+
|
|
139
|
+
# Load your single-cell data
|
|
140
|
+
adata = sc.read_h5ad('your_data.h5ad')
|
|
141
|
+
|
|
142
|
+
# Initialize the GNODEVAE agent
|
|
143
|
+
model = agent_r(
|
|
144
|
+
adata=adata,
|
|
145
|
+
layer='counts', # Layer containing count data
|
|
146
|
+
n_var=2000, # Number of highly variable genes
|
|
147
|
+
tech='PCA', # Dimensionality reduction technique
|
|
148
|
+
n_neighbors=15, # Number of neighbors for graph construction
|
|
149
|
+
latent_dim=10, # Latent space dimension
|
|
150
|
+
hidden_dim=128, # Hidden layer dimension
|
|
151
|
+
encoder_type='graph', # Use graph encoder
|
|
152
|
+
graph_type='GAT', # Graph Attention Network
|
|
153
|
+
lr=1e-4, # Learning rate
|
|
154
|
+
device='cuda' # Use GPU if available
|
|
155
|
+
)
|
|
156
|
+
|
|
157
|
+
# Train the model
|
|
158
|
+
model.fit(epochs=300, update_steps=10, silent=False)
|
|
159
|
+
|
|
160
|
+
# Extract latent representations
|
|
161
|
+
latent = model.get_latent()
|
|
162
|
+
|
|
163
|
+
# Store latent representation in AnnData object
|
|
164
|
+
adata.obsm['X_gnodevae'] = latent
|
|
165
|
+
|
|
166
|
+
# Perform downstream analysis (e.g., clustering)
|
|
167
|
+
import scanpy as sc
|
|
168
|
+
sc.pp.neighbors(adata, use_rep='X_gnodevae')
|
|
169
|
+
sc.tl.leiden(adata)
|
|
170
|
+
sc.tl.umap(adata)
|
|
171
|
+
```
|
|
172
|
+
|
|
173
|
+
### Using Standard GraphVAE (without ODE)
|
|
174
|
+
|
|
175
|
+
```python
|
|
176
|
+
from GNODEVAE import agent
|
|
177
|
+
|
|
178
|
+
# Initialize standard GraphVAE agent
|
|
179
|
+
model = agent(
|
|
180
|
+
adata=adata,
|
|
181
|
+
layer='counts',
|
|
182
|
+
n_var=2000,
|
|
183
|
+
tech='PCA',
|
|
184
|
+
n_neighbors=15,
|
|
185
|
+
latent_dim=10,
|
|
186
|
+
hidden_dim=128,
|
|
187
|
+
encoder_type='GAT',
|
|
188
|
+
lr=1e-4
|
|
189
|
+
)
|
|
190
|
+
|
|
191
|
+
# Train and extract embeddings
|
|
192
|
+
model.fit(epochs=300)
|
|
193
|
+
latent = model.get_latent()
|
|
194
|
+
```
|
|
195
|
+
|
|
196
|
+
## Key Parameters
|
|
197
|
+
|
|
198
|
+
### Data Preprocessing Parameters
|
|
199
|
+
- `layer` (str): Layer of AnnData to use (default: 'counts')
|
|
200
|
+
- `n_var` (int): Number of highly variable genes to select (default: None, uses all)
|
|
201
|
+
- `tech` (str): Dimensionality reduction method - 'PCA', 'NMF', 'FastICA', 'TruncatedSVD', 'FactorAnalysis', or 'LatentDirichletAllocation' (default: 'PCA')
|
|
202
|
+
- `n_neighbors` (int): Number of neighbors for graph construction (default: 15)
|
|
203
|
+
- `batch_tech` (str): Batch correction method - 'harmony' or 'scvi' (default: None)
|
|
204
|
+
- `all_feat` (bool): Whether to use all features or only highly variable genes (default: True)
|
|
205
|
+
|
|
206
|
+
### Model Architecture Parameters
|
|
207
|
+
- `hidden_dim` (int): Hidden layer dimension (default: 128)
|
|
208
|
+
- `latent_dim` (int): Latent space dimension for embeddings (default: 10)
|
|
209
|
+
- `encoder_type` (str): Encoder type - 'graph' or 'linear' (default: 'graph')
|
|
210
|
+
- `graph_type` (str): Graph convolution type - 'GAT', 'GCN', 'SAGE', 'Transformer', etc. (default: 'GAT')
|
|
211
|
+
- `structure_decoder_type` (str): Structure decoder type - 'mlp', 'bilinear', or 'inner_product' (default: 'mlp')
|
|
212
|
+
- `feature_decoder_type` (str): Feature decoder type - 'linear' or 'graph' (default: 'linear')
|
|
213
|
+
- `hidden_layers` (int): Number of hidden layers (default: 2)
|
|
214
|
+
- `dropout` (float): Dropout rate (default: 0.05)
|
|
215
|
+
- `use_residual` (bool): Whether to use residual connections (default: True)
|
|
216
|
+
|
|
217
|
+
### Training Parameters
|
|
218
|
+
- `lr` (float): Learning rate for optimizer (default: 1e-4)
|
|
219
|
+
- `beta` (float): Weight for KL divergence loss term (default: 1.0)
|
|
220
|
+
- `graph` (float): Weight for graph reconstruction loss (default: 1.0)
|
|
221
|
+
- `epochs` (int): Number of training epochs (default: 300)
|
|
222
|
+
- `device` (str or torch.device): Computing device - 'cuda' or 'cpu' (default: auto-detect)
|
|
223
|
+
- `num_parts` (int): Number of graph partitions for mini-batch training (default: 10)
|
|
224
|
+
|
|
225
|
+
### GNODEVAE-Specific Parameters (agent_r)
|
|
226
|
+
- `n_ode_hidden` (int): Number of hidden units in ODE function (default: varies)
|
|
227
|
+
- `w_recon` (float): Weight for reconstruction loss (default: 1.0)
|
|
228
|
+
- `w_kl` (float): Weight for KL divergence loss (default: 1.0)
|
|
229
|
+
- `w_adj` (float): Weight for adjacency matrix loss (default: 1.0)
|
|
230
|
+
- `w_recon_ode` (float): Weight for ODE reconstruction loss (default: 1.0)
|
|
231
|
+
|
|
232
|
+
## Model Architecture
|
|
233
|
+
|
|
234
|
+
GNODEVAE consists of three main components:
|
|
235
|
+
|
|
236
|
+
1. **Graph Encoder**: Encodes cell-cell relationships and gene expression using Graph Attention Networks (GAT) or other graph convolution layers
|
|
237
|
+
2. **Neural ODE**: Models continuous developmental trajectories in the latent space
|
|
238
|
+
3. **Decoder**: Reconstructs both graph structure and gene expression from latent representations
|
|
239
|
+
|
|
240
|
+
The model learns a low-dimensional latent representation that captures:
|
|
241
|
+
- Cell type identity
|
|
242
|
+
- Developmental state
|
|
243
|
+
- Cell-cell relationships
|
|
244
|
+
- Temporal dynamics (with ODE component)
|
|
245
|
+
|
|
246
|
+
## Output
|
|
247
|
+
|
|
248
|
+
After training, GNODEVAE produces:
|
|
249
|
+
- **Latent representations**: Low-dimensional embeddings for each cell
|
|
250
|
+
- **Clustering metrics**: ARI, NMI, Silhouette score, Calinski-Harabasz index, Davies-Bouldin index
|
|
251
|
+
- **Pseudo-time**: Developmental trajectory information (for agent_r with ODE)
|
|
252
|
+
- **Graph structure**: Learned cell-cell similarity graph
|
|
253
|
+
|
|
254
|
+
## Advanced Usage
|
|
255
|
+
|
|
256
|
+
### Custom Graph Construction
|
|
257
|
+
|
|
258
|
+
```python
|
|
259
|
+
# Use custom graph construction parameters
|
|
260
|
+
model = agent_r(
|
|
261
|
+
adata=adata,
|
|
262
|
+
n_neighbors=30, # Increase neighbors for denser graph
|
|
263
|
+
graph_type='Transformer', # Use Transformer convolution
|
|
264
|
+
alpha=0.5 # Set alpha for specific layers
|
|
265
|
+
)
|
|
266
|
+
```
|
|
267
|
+
|
|
268
|
+
### Interpretable Mode
|
|
269
|
+
|
|
270
|
+
```python
|
|
271
|
+
# Use interpretable GraphVAE
|
|
272
|
+
model = agent(
|
|
273
|
+
adata=adata,
|
|
274
|
+
interpretable=True, # Enable interpretable mode
|
|
275
|
+
idim=2 # Interpretable dimension
|
|
276
|
+
)
|
|
277
|
+
```
|
|
278
|
+
|
|
279
|
+
### Extract Pseudo-time
|
|
280
|
+
|
|
281
|
+
```python
|
|
282
|
+
# For GNODEVAE models with ODE component
|
|
283
|
+
# Note: Use GNODEVAE_agent_r from GODEVAE_agent module for pseudo-time functionality
|
|
284
|
+
from GNODEVAE.GODEVAE_agent import GNODEVAE_agent_r
|
|
285
|
+
|
|
286
|
+
model = GNODEVAE_agent_r(adata=adata, ...)
|
|
287
|
+
model.fit(epochs=300)
|
|
288
|
+
|
|
289
|
+
# Get pseudo-time for cells
|
|
290
|
+
pseudotime_df = model.partition_time()
|
|
291
|
+
```
|
|
292
|
+
|
|
293
|
+
## Evaluation Metrics
|
|
294
|
+
|
|
295
|
+
GNODEVAE automatically computes several clustering quality metrics during training:
|
|
296
|
+
- **ARI** (Adjusted Rand Index): Measures clustering agreement with ground truth
|
|
297
|
+
- **NMI** (Normalized Mutual Information): Information-theoretic clustering metric
|
|
298
|
+
- **ASW** (Average Silhouette Width): Measures cluster separation
|
|
299
|
+
- **C_H** (Calinski-Harabasz Index): Ratio of between-cluster to within-cluster variance
|
|
300
|
+
- **D_B** (Davies-Bouldin Index): Average similarity between clusters
|
|
301
|
+
- **P_C** (Pearson Correlation): Correlation between latent dimensions
|
|
302
|
+
|
|
303
|
+
## Citation
|
|
304
|
+
|
|
305
|
+
If you use GNODEVAE in your research, please cite:
|
|
306
|
+
|
|
307
|
+
```bibtex
|
|
308
|
+
@article{fu2025gnodevae,
|
|
309
|
+
title={GNODEVAE: a graph-based ODE-VAE enhances clustering for single-cell data},
|
|
310
|
+
author={Fu, Z. and Chen, C. and Wang, S. and others},
|
|
311
|
+
journal={BMC Genomics},
|
|
312
|
+
volume={26},
|
|
313
|
+
pages={767},
|
|
314
|
+
year={2025},
|
|
315
|
+
doi={10.1186/s12864-025-11946-7}
|
|
316
|
+
}
|
|
317
|
+
```
|
|
318
|
+
|
|
319
|
+
**Full Citation:**
|
|
320
|
+
Fu, Z., Chen, C., Wang, S. et al. GNODEVAE: a graph-based ODE-VAE enhances clustering for single-cell data. BMC Genomics 26, 767 (2025). https://doi.org/10.1186/s12864-025-11946-7
|
|
321
|
+
|
|
322
|
+
[](https://doi.org/10.5281/zenodo.15826042)
|
|
323
|
+
|
|
324
|
+
## License
|
|
325
|
+
|
|
326
|
+
See LICENSE file for details.
|
|
327
|
+
|
|
328
|
+
## Contact
|
|
329
|
+
|
|
330
|
+
For questions and feedback, please open an issue on the GitHub repository.
|
|
@@ -0,0 +1,23 @@
|
|
|
1
|
+
GNODEVAE/CLEAR.py,sha256=8MMUa2N_fta4dpYupCJIwIIKEDCWpFWCJTNwEXw5u88,6680
|
|
2
|
+
GNODEVAE/GODEVAE.py,sha256=K9r-gD18vD_KtzKdrhzAeONNPvig8-13R6r948s71Gw,15930
|
|
3
|
+
GNODEVAE/GODEVAE_agent.py,sha256=Er6tBKpdhj2cE4uUoikft7kflIwZpelpmoALDAnPIpo,14129
|
|
4
|
+
GNODEVAE/GODEVAE_env.py,sha256=E2GdQ8beChDMDMPVzm8afCECamsEl9mrotyKGnSaNq0,26827
|
|
5
|
+
GNODEVAE/GODEVAE_trainer.py,sha256=BcKHl6L5-C4yN02L27_GAyUl9AteNyPwpEWp0kC1upE,11092
|
|
6
|
+
GNODEVAE/SC3.py,sha256=mWknt91wbeGvdYoiiYpG39F07Bz8dNjxUbijBp9iw_o,8322
|
|
7
|
+
GNODEVAE/__init__.py,sha256=JvOj7J_-n5E25xYCWQqT_8ybuOeqTiN9FJUzcTk3ebc,196
|
|
8
|
+
GNODEVAE/agent.py,sha256=2n0Mf6sC3M7FVXoE9fi89V952_q05Km0e7DOlHvW0r0,17103
|
|
9
|
+
GNODEVAE/env.py,sha256=C7hWIZjcbNsEMnMw9fC8NTq6IA9iUuersMdyjw05gj4,14428
|
|
10
|
+
GNODEVAE/mixin.py,sha256=G9uD99jy1sjZnHWsQOZ4rFWocLDjRW0FlM6_TpBQ8ss,14990
|
|
11
|
+
GNODEVAE/module.py,sha256=W5bIkzuXvqENtv5SJ8nJ2gG_h7u8sbif5-sP0uuLmOw,34969
|
|
12
|
+
GNODEVAE/module_refined.py,sha256=Ii3ECuOMB8Bjou-gZXXJW_Y26vAYP6nyYdRT5Xc4Hu0,20158
|
|
13
|
+
GNODEVAE/partition_time.py,sha256=c_lO-8FhS_Pj7BBZMXP67TLF2_XvYDJRGgP5y7w87Yw,6280
|
|
14
|
+
GNODEVAE/scGCC.py,sha256=TluCDszhEcyZ1wXzuLOpo5dBzXD93-yAXD6DkayivRY,7832
|
|
15
|
+
GNODEVAE/scGNN.py,sha256=nCVi_JWFySN46BoX2WlrLDsXQ4BH3gQYdGAqlyEXFM4,6188
|
|
16
|
+
GNODEVAE/trainer.py,sha256=dAdUJA0CyjtILnSLpHoLYuZudwj_d2etJ0FHtHB_EOo,13452
|
|
17
|
+
GNODEVAE/utils.py,sha256=I1WrHpK7v3sMJQdKQTRkLCiYy6Q2VotwMvFVdk8XQAk,15677
|
|
18
|
+
GNODEVAE/utils_ODE.py,sha256=gStxCqXdHMq3UgBqFwoqU-xl2EuhYk8LlMRAxMdjcYo,4952
|
|
19
|
+
gnodevae-0.0.3.dist-info/licenses/LICENSE,sha256=eHiyf509g75dhcKRi3G4eVJWaltCAwYdQ3T2yMn7d1I,1064
|
|
20
|
+
gnodevae-0.0.3.dist-info/METADATA,sha256=NrvWnA2eRwW-Tb13bx94CYzAG6ApDup-DkMkexwV4CU,12885
|
|
21
|
+
gnodevae-0.0.3.dist-info/WHEEL,sha256=aeYiig01lYGDzBgS8HxWXOg3uV61G9ijOsup-k9o1sk,91
|
|
22
|
+
gnodevae-0.0.3.dist-info/top_level.txt,sha256=6mY_IDiMn4UY5Hiwz_hkvUOQ1To9zJzFOjL_cXTzMEE,9
|
|
23
|
+
gnodevae-0.0.3.dist-info/RECORD,,
|
|
@@ -0,0 +1,21 @@
|
|
|
1
|
+
MIT License
|
|
2
|
+
|
|
3
|
+
Copyright (c) 2025 Zeyu Fu
|
|
4
|
+
|
|
5
|
+
Permission is hereby granted, free of charge, to any person obtaining a copy
|
|
6
|
+
of this software and associated documentation files (the "Software"), to deal
|
|
7
|
+
in the Software without restriction, including without limitation the rights
|
|
8
|
+
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
|
9
|
+
copies of the Software, and to permit persons to whom the Software is
|
|
10
|
+
furnished to do so, subject to the following conditions:
|
|
11
|
+
|
|
12
|
+
The above copyright notice and this permission notice shall be included in all
|
|
13
|
+
copies or substantial portions of the Software.
|
|
14
|
+
|
|
15
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
|
16
|
+
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
|
17
|
+
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
|
18
|
+
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
|
19
|
+
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
|
20
|
+
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
|
|
21
|
+
SOFTWARE.
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
GNODEVAE
|