### Install pytntp from GitHub

Source: https://github.com/ben-hudson/pytntp/blob/main/README.md

Install the pytntp library directly from its GitHub repository using pip.

```bash
pip install git+https://github.com/ben-hudson/pytntp
```

--------------------------------

### Visualize Network Congestion and Flow

Source: https://github.com/ben-hudson/pytntp/blob/main/docs/getting-started.md

Visualizes the network, coloring edges by congestion (vc_ratio) and traffic flow (Volume). Centroid connectors (link_type == 3) are excluded to improve map clarity. Requires matplotlib and osmnx.

```python
import matplotlib.pyplot as plt
import osmnx as ox

road_edges = [(u, v, k) for u, v, k, d in network.edges(keys=True, data=True) if d.get("link_type") != 3]
roads = network.edge_subgraph(road_edges)

fig, (ax_vc, ax_flow) = plt.subplots(1, 2, figsize=(16, 8))
ox.plot.plot_graph(roads, ax=ax_vc, edge_color=tntp.quantile_edge_colors(roads, "vc_ratio", "RdYlGn_r"), show=False)
ox.plot.plot_graph(roads, ax=ax_flow, edge_color=tntp.quantile_edge_colors(roads, "Volume", "magma"), show=False)
```

--------------------------------

### Convert to NetworkX Graph

Source: https://github.com/ben-hudson/pytntp/blob/main/docs/getting-started.md

Converts node and network dataframes into a NetworkX graph, compatible with osmnx. Edge attributes from the network dataframe are preserved. Calculates the volume-to-capacity ratio for each edge.

```python
network = tntp.convert_to_networkx(node_df, net_df)

for u, v, k, data in network.edges(keys=True, data=True):
    data["vc_ratio"] = data["Volume"] / data["capacity"]
```

--------------------------------

### Read TNTP Network Files

Source: https://github.com/ben-hudson/pytntp/blob/main/docs/getting-started.md

Loads network data including flow, node, and net files from a specified URL. It renames 'From' and 'To' columns for consistency and merges flow data into the network edges. Ensure the root URL and file names are correct.

```python
import tntp
from urllib.parse import urljoin

name = "ChicagoSketch"
# Note: the file prefix is "ChicagoSketch" but the directory is "Chicago-Sketch".
root = "https://raw.githubusercontent.com/bstabler/TransportationNetworks/refs/heads/master/Chicago-Sketch/"

flow_df = tntp.read_flow_file(urljoin(root, f"{name}_flow.tntp")).rename(
    columns={"From": "init_node", "To": "term_node"}
)
node_df = tntp.read_node_file(
    urljoin(root, f"{name}_node.tntp"), index_col="node", x_col="X", y_col="Y", crs="EPSG:26771"
)
net_df = tntp.read_net_file(urljoin(root, f"{name}_net.tntp"), crs="EPSG:26771").merge(
    flow_df, on=["init_node", "term_node"]
)
```

--------------------------------

### Read TNTP files and convert to NetworkX graph

Source: https://github.com/ben-hudson/pytntp/blob/main/docs/index.md

Load node and network data from .tntp files using specified URLs and convert them into a NetworkX graph. Ensure the correct CRS is provided for accurate geospatial representation.

```python
import tntp
from urllib.parse import urljoin

root = "https://raw.githubusercontent.com/bstabler/TransportationNetworks/refs/heads/master/SiouxFalls/"
node_df = tntp.read_node_file(
    urljoin(root, "SiouxFalls_node.tntp"), index_col="Node", x_col="X", y_col="Y", crs="EPSG:4326"
)
net_df = tntp.read_net_file(urljoin(root, "SiouxFalls_net.tntp"), crs="EPSG:4326")
network = tntp.convert_to_networkx(node_df, net_df)
```

--------------------------------

### Load and Split TNTP Network Data

Source: https://github.com/ben-hudson/pytntp/blob/main/docs/zone-splitting.md

Loads network, node, demand, and flow data from specified URLs and then splits the centroid nodes using `tntp.split_zone_nodes`. The network data is merged with flow data before splitting to ensure volume and cost information is preserved.

```python
import geopandas as gpd
import tntp
from urllib.parse import urljoin

root = "https://raw.githubusercontent.com/bstabler/TransportationNetworks/refs/heads/master/Anaheim/"
node_df = gpd.read_file(urljoin(root, "anaheim_nodes.geojson")).set_index("id").rename_axis("node")
net_df = tntp.read_net_file(urljoin(root, "Anaheim_net.tntp"), crs=node_df.crs)
demand_df = tntp.read_demand_file(urljoin(root, "Anaheim_trips.tntp"))
flow_df = tntp.read_flow_file(urljoin(root, "Anaheim_flow.tntp")).rename(
    columns={"From": "init_node", "To": "term_node"}
)

# Merge flow into net BEFORE splitting so Volume/Cost ride along through the centroid
# rename. pd.merge drops attrs, so preserve FIRST THRU NODE manually.
attrs = net_df.attrs
net_df = net_df.merge(flow_df, on=["init_node", "term_node"])
net_df.attrs = attrs

node_df, net_df, demand_df = tntp.split_zone_nodes(node_df, net_df, demand_df)
network = tntp.convert_to_networkx(node_df, net_df)
```

--------------------------------

### Filter Road Edges Using `parent_node`

Source: https://github.com/ben-hudson/pytntp/blob/main/docs/zone-splitting.md

Identifies actual road edges after zone splitting by filtering edges where both endpoints descend from a through node (i.e., their `parent_node` ID is greater than or equal to the `FIRST THRU NODE`). This creates a subgraph containing only the road network.

```python
first_thru_node = int(net_df.attrs["FIRST THRU NODE"])
through_nodes = set(node_df.index[node_df["parent_node"] >= first_thru_node])
road_edges = [(u, v, k) for u, v, k in network.edges(keys=True) if u in through_nodes and v in through_nodes]
roads = network.edge_subgraph(road_edges)
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.