Adds Altair plot component and some other fixes #2644

nissu99 · 2025-01-26T06:35:31Z

- rename make_space_altair to something like make_altair_space_component
- add make_altair_plot_component
- Add support for all spaces. This means including the hexgrid and network transformation of the x and y coordinates.
- Have a generic get_agent_data method

@quaquel @EwoutH I’ve added a new component, make_altair_plot_component. Could you kindly check if it aligns well with the project?

P.S working on space support.

for more information, see https://pre-commit.ci

github-actions · 2025-01-26T06:43:06Z

Performance benchmarks:

Model	Size	Init time [95% CI]	Run time [95% CI]
BoltzmannWealth	small	🔵 +1.3% [+0.4%, +2.2%]	🔵 -0.3% [-0.5%, -0.2%]
BoltzmannWealth	large	🔵 -1.2% [-1.8%, -0.6%]	🔵 -0.5% [-1.6%, +0.7%]
Schelling	small	🔵 -0.7% [-1.0%, -0.5%]	🔵 +0.8% [+0.6%, +0.9%]
Schelling	large	🔵 -0.7% [-1.1%, -0.3%]	🔵 -0.9% [-1.3%, -0.5%]
WolfSheep	small	🔵 +0.1% [-0.1%, +0.3%]	🔵 +1.1% [+0.9%, +1.2%]
WolfSheep	large	🔵 +0.9% [+0.3%, +1.5%]	🔵 +1.9% [+1.2%, +2.5%]
BoidFlockers	small	🔵 -0.7% [-1.3%, -0.2%]	🔵 +1.8% [+1.6%, +2.0%]
BoidFlockers	large	🔵 -0.1% [-0.5%, +0.4%]	🔵 +1.9% [+1.6%, +2.3%]

EwoutH · 2025-01-26T10:42:39Z

Thanks for the PR! A few of the automated checks broke, could you look into what caused that?

I will leave the review to others.

tpike3 · 2025-01-26T13:17:47Z

@nissu99 also please look at #2643 and #2641 As you are @sanika-n are working in the same space

As it looks like you two are looking over the entire altair implementation I would also recommend looking at #2642 discussion

It is always good to collaborate and think together you may address some larger visualization challenges.

"alt defined"

for more information, see https://pre-commit.ci

nissu99 · 2025-01-27T08:04:19Z

@tpike3 When I run the tests locally, they don't fail, but they are failing here.

tpike3 · 2025-01-27T11:02:15Z

@tpike3 When I run the tests locally, they don't fail, but they are failing here.

I am not sure why your tests are passing locally, however for tests to pass they need to updated. It is expecting None and now getting Chart, because you added post process.

__ ERROR collecting tests/test_components_matplotlib.py _____________
tests/test_components_matplotlib.py:[20](https://github.com/projectmesa/mesa/actions/runs/12984364297/job/36207123671?pr=2644#step:6:21): in <module>
    from mesa.visualization.mpl_space_drawing import (
mesa/visualization/__init__.py:13: in <module>
    from .components import make_plot_component, make_space_component
mesa/visualization/components/__init__.py:7: in <module>
    from .altair_components import (
mesa/visualization/components/altair_components.py:93: in <module>
    post_process: Callable[[alt.Chart], alt.Chart] | None = None,
E   AttributeError: 'NoneType' object has no attribute 'Chart'

"removed a line"

"updated toml"

"removed suppress"

for more information, see https://pre-commit.ci

"doc string "

for more information, see https://pre-commit.ci

"operator bug"

nissu99 · 2025-01-27T18:41:24Z

@tpike3 When I run the tests locally, they don't fail, but they are failing here.

I am not sure why your tests are passing locally, however for tests to pass they need to updated. It is expecting None and now getting Chart, because you added post process.

__ ERROR collecting tests/test_components_matplotlib.py _____________
tests/test_components_matplotlib.py:[20](https://github.com/projectmesa/mesa/actions/runs/12984364297/job/36207123671?pr=2644#step:6:21): in <module>
    from mesa.visualization.mpl_space_drawing import (
mesa/visualization/__init__.py:13: in <module>
    from .components import make_plot_component, make_space_component
mesa/visualization/components/__init__.py:7: in <module>
    from .altair_components import (
mesa/visualization/components/altair_components.py:93: in <module>
    post_process: Callable[[alt.Chart], alt.Chart] | None = None,
E   AttributeError: 'NoneType' object has no attribute 'Chart'

Thank You, the issue got resolved by adding altair to the toml file.

nissu99 · 2025-01-30T18:40:03Z

@quaquel @tpike3 i was making a generic get_agent_data function ,should i remove all the old functions like _get_agent_data_new_discrete_space or keeping them will be right?

quaquel · 2025-01-30T19:20:59Z

Removing stuff that has become redundant is fine.

nissu99 · 2025-01-31T10:54:47Z

@quaquel i have added all the stuff,can you review it?

quaquel · 2025-01-31T11:14:23Z

I'll try to review it over the weekend, but I am rather bussy at the moment. This is a part of the code base I am not intimately familiar with and his been a while since I used Altair, so reviewing will take more time than parts of the code base I know inside out.

nissu99 · 2025-02-16T08:15:54Z

@nissu99 Thanks for making some changes already! Could you share the state of this PR and preferably reply to the review comments (of quaquel) individually?

@Sahil-Chhoker this would be a great first PR for you review, as it blocks some others.

Really sorry for the delay ,was travelling due to some family reasons,I am working on changes to make the api similar to that of matplotlib.

quaquel · 2025-02-16T08:18:11Z

@nissu99 can you give an indication of your timeline? We might otherwise merge some other altair PR first.

nissu99 · 2025-02-16T08:21:19Z

@nissu99 can you give an indication of your timeline? We might otherwise merge some other altair PR first.

As you had pointed out, I was seeing if a common overarching function could be made for both Matplotlib and Altair stuff.
other changes are not that time consuming.

nissu99 · 2025-02-16T08:23:15Z

@quaquel what do you say should I focus on that or just make other changes and will make a future PR for the Common function?

quaquel · 2025-02-16T08:27:24Z

This PR is already an improvement. The logical next step, which ideally would be part of this PR, is to structure the code exactly as done for matplotlib. This, for example, means having a single overarching method to get agent data and updating agent positions for e.g., networks, and hexgrids inside their respective draw method. The last step, and well beyond this PR is to see what we can harmonize across altair and matplotib.

for more information, see https://pre-commit.ci

nissu99 · 2025-02-16T10:40:32Z

This PR is already an improvement. The logical next step, which ideally would be part of this PR, is to structure the code exactly as done for matplotlib. This, for example, means having a single overarching method to get agent data and updating agent positions for e.g., networks, and hexgrids inside their respective draw method. The last step, and well beyond this PR is to see what we can harmonize across altair and matplotib.

made the changes , is this fine?

Sahil-Chhoker · 2025-02-16T12:23:20Z

@nissu99, is the grid drawing logic implemented yet, because I am unable to see the grid lines?

Sahil-Chhoker

@nissu99 I have done a basic overview of functioning of the altair drawing system, if you find similar problems please fix them. I will do a thorough review on a later date.

Sahil-Chhoker · 2025-02-16T12:09:39Z

mesa/visualization/components/altair_components.py

+            return _draw_legacy_grid(space, all_agent_data, agent_portrayal)
+        case HexGrid():
+            return _draw_hex_grid(space, all_agent_data, agent_portrayal)
+        case NetworkGrid():
+            return _draw_network_grid(space, all_agent_data, agent_portrayal)
+        case ContinuousSpace() | mesa.experimental.continuous_space.ContinuousSpace():
+            return _draw_continuous_space(space, all_agent_data, agent_portrayal)


if the agent_portrayal was already used to get the agent data, why is it passed into these functions separately?

In mpl_space_drawing we follow similar structure ,that is why passed it.

I don't think you have to follow that here, passing parameters that are not being used in the function is more confusing.

@nissu99 Can you make the changes here as well.

@Sahil-Chhoker tests/test_solara_viz.py::test_call_space_drawer is failing due to absence of argument.
should i make changes to the tests?

for matplotlib, the reasonagent_portrayal is passed to the respective _draw functions is because the data from the agents is gathered in the respective draw methods. My suggestion is to do the same on the altair side. So, I suggest moving the gathering of agent data into the respective draw functions.

for matplotlib, the reasonagent_portrayal is passed to the respective _draw functions is because the data from the agents is gathered in the respective draw methods.

I was aware of that, but thought that the new agent data collector function was part of the new API redesign, since it separates the drawing of agents from drawing of grids, is that not the case?

Fair question. I am not sure yet. I am currently inclined to have some kind of adapter object in between the model and the drawing. However, implementing this will require wider changes throughout the code base. So, the more similar altair and matplotlib drawing are, the easier it will be to make the changes on both sides.

makes sense!

Nice idea! this way it will be easier if we add other visualization support.

Sahil-Chhoker · 2025-02-16T12:57:49Z

mesa/visualization/components/altair_components.py

+def _draw_discrete_grid(space, all_agent_data, agent_portrayal):
+    """Create Altair visualization for Discrete Grid."""
    invalid_tooltips = ["color", "size", "x", "y"]
+    x_y_type = "ordinal"

-    x_y_type = "ordinal" if not isinstance(space, ContinuousSpace) else "nominal"
+    encoding_dict = {
+        "x": alt.X("x", axis=None, type=x_y_type),
+        "y": alt.Y("y", axis=None, type=x_y_type),
+        "tooltip": [
+            alt.Tooltip(key, type=alt.utils.infer_vegalite_type_for_pandas([value]))
+            for key, value in all_agent_data[0].items()
+            if key not in invalid_tooltips
+        ],
+    }
+
+    has_color = "color" in all_agent_data[0]
+    if has_color:
+        encoding_dict["color"] = alt.Color("color", type="nominal")
+    has_size = "size" in all_agent_data[0]
+    if has_size:
+        encoding_dict["size"] = alt.Size("size", type="quantitative")
+
+    chart = (
+        alt.Chart(
+            alt.Data(values=all_agent_data), encoding=alt.Encoding(**encoding_dict)
+        )
+        .mark_point(filled=True)
+        .properties(width=280, height=280)
+    )
+
+    if not has_size:
+        length = min(space.width, space.height)
+        chart = chart.mark_point(size=30000 / length**2, filled=True)
+
+    return chart


The scale of grid automatically adjusts to the position of agents, is there no way to stop the resizing of the grid? Also no visualization of grid lines.

Some images for better understanding:

Notice the scale and positioning of the larger agent.

Also the coordinates of grids are reversed for matplotlib and Altair, not sure how it can affect the usage.

i think resizing is a default behaviour in altair , these things can be handled in the post_process function @quaquel what do you think?

Sahil-Chhoker · 2025-02-16T13:02:33Z

mesa/visualization/components/altair_components.py

+def _draw_hex_grid(space, all_agent_data, agent_portrayal):
+    """Create Altair visualization for Hex Grid."""
+    invalid_tooltips = ["color", "size", "x", "y", "q", "r"]
+    x_y_type = "quantitative"
+
+    # Parameters for hexagon grid
+    size = 1.0
+    x_spacing = math.sqrt(3) * size
+    y_spacing = 1.5 * size
+
+    # Calculate x, y coordinates from axial coordinates
+    for agent_data in all_agent_data:


Same problem as draw_discrete_grid here. Also no grid lines visualization.

nissu99 · 2025-02-16T14:42:56Z

@nissu99 I have done a basic overview of functioning of the altair drawing system, if you find similar problems please fix them. I will do a thorough review on a later date.

I have not yet implemented the grid visualization

nissu99 · 2025-02-16T15:13:03Z

@quaquel @Sahil-Chhoker I think we should keep the core API very concise, and changes like controlling grid visibility can be handled in the post-processing stage, what is your opinion?

Sahil-Chhoker · 2025-02-16T15:18:58Z

I also think the the same and I think it is followed everywhere in the visualization API, but the grid resizing every time the agents move may be confusing and not very elegant.

quaquel · 2025-02-16T15:24:28Z

Grid visibility is not trivial for networks or hexgrids. Hence, it is included on the matplotlib side and controlled via a boolean. As indicated in #2642, I am considering changing the API to give the user more control over this beyond just a boolean. So I am fine with leaving this out of altair for now.

I agree with @Sahil-Chhoker, the view limits should be fixed and not change every step.

nissu99 · 2025-02-16T15:59:13Z

@Sahil-Chhoker can you share the example you are using?

Sahil-Chhoker · 2025-02-16T16:11:44Z

Don't mind the ugly and commented code, I keep changing this example as per need, and I really recommend the Command Center from #2674 for better debugging.

import random
from mesa.experimental.cell_space.grid import HexGrid, OrthogonalMooreGrid
from mesa.experimental.continuous_space import ContinuousSpace, ContinuousSpaceAgent
from mesa import Model
import math
from mesa.experimental.cell_space import CellAgent
from mesa.visualization import SolaraViz, make_space_component
from mesa.experimental.cell_space.property_layer import PropertyLayer
import numpy as np
from portrayal_components import PropertyLayerPortrayal, AgentPortrayalStyle, PropertyLayerStyle

import shutil
from agents import myAgent
 
class myModel(Model): 
    def __init__(self, seed=None, width=5,  height=5 , n_agents=4):
        super().__init__(seed=seed) 
        self.width = width
        self.height = height
        self.num_agents = n_agents
        self.grid = OrthogonalMooreGrid( 
            (self.width, self.height), capacity=math.inf, torus=False, random=self.random
        )
        self.x = 1
        self.test_layer = PropertyLayer("test", (self.width, self.height), default_value=1, dtype=int)
        self.test_layer.data = np.random.randint(2, 6, (self.width, self.height))

        for i in range(self.width * self.height):
            cell = self.grid.all_cells.cells[i]
            if cell.coordinate == (1, 1):
                myAgent(self, cell=cell, id=0)
            # else:
            #     myAgent(self, cell=cell, id=self.x)
            #     self.x += 1
            # myAgent(self, cell=self.grid.all_cells.cells[i], id=self.x)
            # self.x += 1
    
        # self.grid.add_property_layer(self.test_layer)
        # myAgent(self, cell=self.grid.all_cells.cells[17], id=self.x)
        # self.x += 1
        # myAgent(self, cell=self.grid.all_cells.cells[5], id=self.x)
        # self.x += 1
        # myAgent(self, cell=self.grid.all_cells.cells[7], id=self.x)
        # self.x += 1
        # myAgent(self, cell=self.grid.all_cells.cells[1], id=self.x)
        # self.x += 1
        # myAgent(self, cell=self.grid.all_cells.cells[1], id=self.x)
        # self.x += 1
        # myAgent(self, cell=self.grid.all_cells.cells[1], id=self.x)
        # self.x += 1
        # myAgent(self, cell=self.grid.all_cells.cells[1], id=self.x)

        self.running = True

    def step(self):
        self.agents.shuffle_do("step")


class ContModel(Model):
    def __init__(self, seed=None):
        super().__init__(seed=seed)
        self.grid = ContinuousSpace([[0, 1], [0, 1]], torus=True, random=self.random)
        self.test_layer = PropertyLayer("test2", [1, 1], default_value=0.0, dtype=float)
        # self.grid.add_property_layer(self.test_layer)
        self.k_agent = ContinuousSpaceAgent(self.grid, self)
        self.k_agent.position = [0.5, 0.5]
        self.a_agent = ContinuousSpaceAgent(self.grid, self)
        self.a_agent.position = [0.55, 0.55]
    
    def step(self):
        self.k_agent.position = [random.random(), random.random()]
        

def agent_portrayal(agent):
    portrayal = AgentPortrayalStyle(
        color='tab:orange',
        marker='^',
        size=150 if agent.id == 0 else 50,
        zorder=1
    )
    return dict(portrayal)

property_layer_portrayal = PropertyLayerPortrayal()
property_layer_portrayal.add_layer(name="test",color="blue", alpha=0.5, colorbar=True, vmin=0, vmax=6)
property_layer_portrayal = dict(property_layer_portrayal)


def post_process(ax):
    ax.set_aspect("equal")
    ax.get_figure().set_size_inches(8, 8)


model_params = {
    "width": {
        "type": "SliderInt",
        "value": 50,
        "label": "Width",
        "min": 5,
        "max": 60,
        "step": 1,
    },
    "height": {
        "type": "SliderInt",
        "value": 50,
        "label": "Height",
        "min": 5,
        "max": 60,
        "step": 1,
    },
}



model = myModel()

space_component = make_space_component(
    agent_portrayal=agent_portrayal, backend="altair", draw_grid=True
)

# space_component = make_space_component(
#     agent_portrayal=agent_portrayal, draw_grid=True, post_process=post_process, propertylayer_portrayal=property_layer_portrayal
# )

page = SolaraViz(
    model,
    components=[space_component],
    model_params=model_params,
    name="Test Model",
    additional_imports={
        "myAgent": myAgent,
    }
)
page

nissu99 · 2025-02-17T14:58:30Z

Don't mind the ugly and commented code, I keep changing this example as per need, and I really recommend the Command Center from #2674 for better debugging.

import random
from mesa.experimental.cell_space.grid import HexGrid, OrthogonalMooreGrid
from mesa.experimental.continuous_space import ContinuousSpace, ContinuousSpaceAgent
from mesa import Model
import math
from mesa.experimental.cell_space import CellAgent
from mesa.visualization import SolaraViz, make_space_component
from mesa.experimental.cell_space.property_layer import PropertyLayer
import numpy as np
from portrayal_components import PropertyLayerPortrayal, AgentPortrayalStyle, PropertyLayerStyle

import shutil
from agents import myAgent
 
class myModel(Model): 
    def __init__(self, seed=None, width=5,  height=5 , n_agents=4):
        super().__init__(seed=seed) 
        self.width = width
        self.height = height
        self.num_agents = n_agents
        self.grid = OrthogonalMooreGrid( 
            (self.width, self.height), capacity=math.inf, torus=False, random=self.random
        )
        self.x = 1
        self.test_layer = PropertyLayer("test", (self.width, self.height), default_value=1, dtype=int)
        self.test_layer.data = np.random.randint(2, 6, (self.width, self.height))

        for i in range(self.width * self.height):
            cell = self.grid.all_cells.cells[i]
            if cell.coordinate == (1, 1):
                myAgent(self, cell=cell, id=0)
            # else:
            #     myAgent(self, cell=cell, id=self.x)
            #     self.x += 1
            # myAgent(self, cell=self.grid.all_cells.cells[i], id=self.x)
            # self.x += 1
    
        # self.grid.add_property_layer(self.test_layer)
        # myAgent(self, cell=self.grid.all_cells.cells[17], id=self.x)
        # self.x += 1
        # myAgent(self, cell=self.grid.all_cells.cells[5], id=self.x)
        # self.x += 1
        # myAgent(self, cell=self.grid.all_cells.cells[7], id=self.x)
        # self.x += 1
        # myAgent(self, cell=self.grid.all_cells.cells[1], id=self.x)
        # self.x += 1
        # myAgent(self, cell=self.grid.all_cells.cells[1], id=self.x)
        # self.x += 1
        # myAgent(self, cell=self.grid.all_cells.cells[1], id=self.x)
        # self.x += 1
        # myAgent(self, cell=self.grid.all_cells.cells[1], id=self.x)

        self.running = True

    def step(self):
        self.agents.shuffle_do("step")


class ContModel(Model):
    def __init__(self, seed=None):
        super().__init__(seed=seed)
        self.grid = ContinuousSpace([[0, 1], [0, 1]], torus=True, random=self.random)
        self.test_layer = PropertyLayer("test2", [1, 1], default_value=0.0, dtype=float)
        # self.grid.add_property_layer(self.test_layer)
        self.k_agent = ContinuousSpaceAgent(self.grid, self)
        self.k_agent.position = [0.5, 0.5]
        self.a_agent = ContinuousSpaceAgent(self.grid, self)
        self.a_agent.position = [0.55, 0.55]
    
    def step(self):
        self.k_agent.position = [random.random(), random.random()]
        

def agent_portrayal(agent):
    portrayal = AgentPortrayalStyle(
        color='tab:orange',
        marker='^',
        size=150 if agent.id == 0 else 50,
        zorder=1
    )
    return dict(portrayal)

property_layer_portrayal = PropertyLayerPortrayal()
property_layer_portrayal.add_layer(name="test",color="blue", alpha=0.5, colorbar=True, vmin=0, vmax=6)
property_layer_portrayal = dict(property_layer_portrayal)


def post_process(ax):
    ax.set_aspect("equal")
    ax.get_figure().set_size_inches(8, 8)


model_params = {
    "width": {
        "type": "SliderInt",
        "value": 50,
        "label": "Width",
        "min": 5,
        "max": 60,
        "step": 1,
    },
    "height": {
        "type": "SliderInt",
        "value": 50,
        "label": "Height",
        "min": 5,
        "max": 60,
        "step": 1,
    },
}



model = myModel()

space_component = make_space_component(
    agent_portrayal=agent_portrayal, backend="altair", draw_grid=True
)

# space_component = make_space_component(
#     agent_portrayal=agent_portrayal, draw_grid=True, post_process=post_process, propertylayer_portrayal=property_layer_portrayal
# )

page = SolaraViz(
    model,
    components=[space_component],
    model_params=model_params,
    name="Test Model",
    additional_imports={
        "myAgent": myAgent,
    }
)
page

you are using some custom imports here ,nevermind Thank You!

for more information, see https://pre-commit.ci

nissu99 · 2025-02-17T15:09:13Z

@Sahil-Chhoker can you check the changes?

Sahil-Chhoker · 2025-02-17T15:14:31Z

@nissu99 Before I do an through review, can you please tell me the scope of this PR and what it aims to add into the project in detail except the ones listed in the description. And if all the functionalities are there or not.

nissu99 · 2025-02-17T15:20:20Z

This PR added Altair Plot support ,worked on a generic data extracting function , created different functions for drawing different types of grids similar to that in matplot _space and tried to make overall api as similar as that of mpl_space api.

Sahil-Chhoker · 2025-02-18T12:54:33Z

@nissu99 very sorry for misleading you but the generic agent data collection needs to be changed to be like the one used in matplotlib.

nissu99 · 2025-02-18T14:06:24Z

@nissu99 very sorry for misleading you but the generic agent data collection needs to be changed to be like the one used in matplotlib.

No worries ,this is how we learn.

for more information, see https://pre-commit.ci

Sahil-Chhoker

I have done the review of basic rectangular and hex grids, first I would love to get them finished up. Then I also noticed that the implementation of the new Network is also missing. I will do a review again once you make the changes, please take a look at the review and don't hesitate to ask for implementations as well if unable to get them.

Sahil-Chhoker · 2025-02-18T14:02:35Z

mesa/visualization/components/altair_components.py

+def _draw_discrete_grid(space, all_agent_data, agent_portrayal):
+    """Create Altair visualization for Discrete Grid."""
    invalid_tooltips = ["color", "size", "x", "y"]
+    x_y_type = "ordinal"

-    x_y_type = "ordinal" if not isinstance(space, ContinuousSpace) else "nominal"
+    encoding_dict = {
+        "x": alt.X("x", axis=None, type=x_y_type),
+        "y": alt.Y("y", axis=None, type=x_y_type),
+        "tooltip": [
+            alt.Tooltip(key, type=alt.utils.infer_vegalite_type_for_pandas([value]))
+            for key, value in all_agent_data[0].items()
+            if key not in invalid_tooltips
+        ],
+    }
+
+    has_color = "color" in all_agent_data[0]
+    if has_color:
+        encoding_dict["color"] = alt.Color("color", type="nominal")
+    has_size = "size" in all_agent_data[0]
+    if has_size:
+        encoding_dict["size"] = alt.Size("size", type="quantitative")
+
+    chart = (
+        alt.Chart(
+            alt.Data(values=all_agent_data), encoding=alt.Encoding(**encoding_dict)
+        )
+        .mark_point(filled=True)
+        .properties(width=280, height=280)
+    )
+
+    if not has_size:
+        length = min(space.width, space.height)
+        chart = chart.mark_point(size=30000 / length**2, filled=True)
+
+    chart = chart.encode(
+        x=alt.X(
+            "x", axis=None, type=x_y_type, scale=alt.Scale(domain=(0, space.width - 1))
+        ),
+        y=alt.Y(
+            "y", axis=None, type=x_y_type, scale=alt.Scale(domain=(0, space.height - 1))
+        ),
+    )
+
+    return chart


I don't think this still works as expected, did you test this code using the example? For me it looks something like this even on placing 25 agents on 5x5 grid at every position:

Sahil-Chhoker · 2025-02-18T14:16:45Z

mesa/visualization/components/altair_components.py

        case Grid():
-            all_agent_data = _get_agent_data_new_discrete_space(space, agent_portrayal)
+            return _draw_discrete_grid(space, all_agent_data)
        case _Grid():
-            all_agent_data = _get_agent_data_old__discrete_space(space, agent_portrayal)
-        case ContinuousSpace():
-            all_agent_data = _get_agent_data_continuous_space(space, agent_portrayal)
+            return _draw_legacy_grid(space, all_agent_data)
+        case HexGrid():
+            return _draw_hex_grid(space, all_agent_data)


This order will never draw the HexGrid, because the hex grid inherits form the Grid. I think reversing their order will fix it.

Sahil-Chhoker · 2025-02-18T14:21:14Z

mesa/visualization/components/altair_components.py

+        case Grid():
+            # New DiscreteSpace or experimental cell space
+            for cell in space.all_cells:
+                for agent in cell.agents:
+                    data = agent_portrayal(agent)
+                    data.update({"x": cell.coordinate[0], "y": cell.coordinate[1]})
+                    all_agent_data.append(data)

-    Returns:
-        list of dicts
+        case _Grid():
+            # Legacy Grid
+            for content, (x, y) in space.coord_iter():
+                if not content:
+                    continue
+                agents = [content] if not hasattr(content, "__iter__") else content
+                for agent in agents:
+                    data = agent_portrayal(agent)
+                    data.update({"x": x, "y": y})
+                    all_agent_data.append(data)
+
+        case HexGrid():


Again this order won't work, because the HexGrid inherits from Grid.

Sahil-Chhoker · 2025-02-18T14:22:11Z

mesa/visualization/components/altair_components.py

+
+        case HexGrid():
+            # Hex-based grid
+            for content, (q, r) in space.coord_iter():


coord_iter doesn't exist for HexGrid

Sahil-Chhoker · 2025-02-18T14:26:54Z

mesa/visualization/components/altair_components.py

+def _draw_hex_grid(space, all_agent_data, agent_portrayal):
+    """Create Altair visualization for Hex Grid."""
+    invalid_tooltips = ["color", "size", "x", "y", "q", "r"]
+    x_y_type = "quantitative"
+


Hex grid does not look like this:

Review the implementation and visualization from what's done in the matplotlib part.

nissu99 · 2025-02-18T14:45:21Z

@Sahil-Chhoker Sorry for the inconvenience. I will test more carefully this time.

nissu99 and others added 3 commits January 26, 2025 11:22

renamed make_space_altair

566b72f

added make_altair_plot_component support

e1f947c

[pre-commit.ci] auto fixes from pre-commit.com hooks

9e3a625

for more information, see https://pre-commit.ci

nissu99 changed the title ~~ADDS ALTAIR PLOT COMPONENT AND SOME OTHER FIXES #2435~~ ADDS ALTAIR PLOT COMPONENT AND SOME OTHER FIXES Jan 26, 2025

EwoutH added the visualisation label Jan 26, 2025

EwoutH requested review from Corvince and quaquel January 26, 2025 10:41

tpike3 mentioned this pull request Jan 26, 2025

added property_layer with altair #2643

Open

nissu99 and others added 3 commits January 27, 2025 13:15

Update altair_components.py

a88d816

"alt defined"

[pre-commit.ci] auto fixes from pre-commit.com hooks

c579f81

for more information, see https://pre-commit.ci

Merge branch 'main' into main

489c0fd

nissu99 and others added 7 commits January 27, 2025 23:30

Update altair_components.py

95a607b

"removed a line"

Update pyproject.toml

8e1779b

"updated toml"

Update altair_components.py

2b69a62

"removed suppress"

[pre-commit.ci] auto fixes from pre-commit.com hooks

b5613f1

for more information, see https://pre-commit.ci

Update altair_components.py

0d1b8be

"doc string "

[pre-commit.ci] auto fixes from pre-commit.com hooks

ed8ca71

for more information, see https://pre-commit.ci

Update altair_components.py

da8fd53

"operator bug"

nissu99 and others added 2 commits January 31, 2025 11:49

Merge branch 'main' into main

24210f4

blank

4a38eef

nissu99 and others added 2 commits February 16, 2025 16:05

changes regarding updating agent position and docs

0a37307

[pre-commit.ci] auto fixes from pre-commit.com hooks

4d1bd06

for more information, see https://pre-commit.ci

Sahil-Chhoker reviewed Feb 16, 2025

View reviewed changes

nissu99 and others added 2 commits February 17, 2025 20:32

resize fix

d6e96b0

[pre-commit.ci] auto fixes from pre-commit.com hooks

54daa24

for more information, see https://pre-commit.ci

removed unused parameter

3d53d52

Corvince changed the title ~~ADDS ALTAIR PLOT COMPONENT AND SOME OTHER FIXES~~ Adds Altair plot component and some other fixes Feb 18, 2025

nissu99 and others added 2 commits February 18, 2025 19:55

removed generic_data_collection function

329bd24

[pre-commit.ci] auto fixes from pre-commit.com hooks

a29c557

for more information, see https://pre-commit.ci

Sahil-Chhoker reviewed Feb 18, 2025

View reviewed changes

Adds Altair plot component and some other fixes #2644

Are you sure you want to change the base?

Adds Altair plot component and some other fixes #2644

Conversation

nissu99 commented Jan 26, 2025 • edited Loading

github-actions bot commented Jan 26, 2025

EwoutH commented Jan 26, 2025

tpike3 commented Jan 26, 2025 • edited Loading

nissu99 commented Jan 27, 2025 • edited Loading

tpike3 commented Jan 27, 2025

nissu99 commented Jan 27, 2025

nissu99 commented Jan 30, 2025 • edited Loading

quaquel commented Jan 30, 2025

nissu99 commented Jan 31, 2025

quaquel commented Jan 31, 2025

nissu99 commented Feb 16, 2025

quaquel commented Feb 16, 2025

nissu99 commented Feb 16, 2025

nissu99 commented Feb 16, 2025

quaquel commented Feb 16, 2025

nissu99 commented Feb 16, 2025 • edited Loading

Sahil-Chhoker commented Feb 16, 2025

Sahil-Chhoker left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

nissu99 Feb 18, 2025 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

nissu99 commented Feb 16, 2025

nissu99 commented Feb 16, 2025

Sahil-Chhoker commented Feb 16, 2025

quaquel commented Feb 16, 2025

nissu99 commented Feb 16, 2025

Sahil-Chhoker commented Feb 16, 2025

nissu99 commented Feb 17, 2025

nissu99 commented Feb 17, 2025

Sahil-Chhoker commented Feb 17, 2025

nissu99 commented Feb 17, 2025 • edited Loading

Sahil-Chhoker commented Feb 18, 2025

nissu99 commented Feb 18, 2025

Sahil-Chhoker left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

nissu99 commented Feb 18, 2025

nissu99 commented Jan 26, 2025 •

edited

Loading

tpike3 commented Jan 26, 2025 •

edited

Loading

nissu99 commented Jan 27, 2025 •

edited

Loading

nissu99 commented Jan 30, 2025 •

edited

Loading

nissu99 commented Feb 16, 2025 •

edited

Loading

nissu99 Feb 18, 2025 •

edited

Loading

nissu99 commented Feb 17, 2025 •

edited

Loading