Merge pull request #48 from ATTPC/gwm17_dev

Fix the SpyralWriter, update docs

Author: gwm17

docs/user_guide/detector/index.md

@@ -75,6 +75,6 @@ The final step of the process is to write the simulated detector response to dis
 
 ### Use with Spyral
 
-For convience, attpc_engine includes a `SpyralWriter`, which will output a HDF5 file formatted to mimic [Spyral](https://attpc.github.io/Spyral), the AT-TPC group-supported analysis framework. `SpyralWriter` also includes the useful feature of splitting the output into multiple files based on file size, which can help take advantage of parallel analysis pipelines. Note that attpc_engine does not claim to provide values in the integral or amplitude fields of a Spyral point cloud that will match real data; `SpyralWriter` use the [theoretical response function](https://www.sciencedirect.com/science/article/pii/S0168900216309408?casa_token=DAWbhPvX49MAAAAA:Zen7nFs-pgG9wu__g0rrzus01B1pa7ZYspbz_KOnn-dZyhXNglEqtgywFKwYrvKKsIwEY10n49XV) of the GET electronics to make an attempt to approximate the amplitude and integral, but this should be regarded as dubious at best.
+For convience, attpc_engine includes a `SpyralWriter`, which will output a HDF5 file formatted to mimic [Spyral](https://attpc.github.io/Spyral), the AT-TPC group-supported analysis framework. `SpyralWriter` also includes the useful feature of splitting the output into multiple files based on number of events, which can help take advantage of parallel analysis pipelines. Note that attpc_engine does not claim to provide values in the integral or amplitude fields of a Spyral point cloud that will match real data; `SpyralWriter` use the [theoretical response function](https://www.sciencedirect.com/science/article/pii/S0168900216309408?casa_token=DAWbhPvX49MAAAAA:Zen7nFs-pgG9wu__g0rrzus01B1pa7ZYspbz_KOnn-dZyhXNglEqtgywFKwYrvKKsIwEY10n49XV) of the GET electronics to make an attempt to approximate the amplitude and integral, but this should be regarded as dubious at best.
 
 Given that attpc_engine outputs point clouds, the results should not be run through the PointcloudPhase, rather starting directly with the ClusterPhase. Note that because Spyral expects trace data, one will need to "fake" the trace datapath. The simplest way to do this is to point the trace path to the Pointcloud directory of the workspace. Just be sure not to run the PointcloudPhase in this case!

docs/user_guide/getting_started.md

@@ -164,13 +164,13 @@ electronics = ElectronicsParams(
     shaping_time=1000,
     micromegas_edge=10,
     windows_edge=560,
-    adc_threshold=40,
+    adc_threshold=10,
 )
 
 pads = PadParams()
 
 config = Config(detector, electronics, pads)
-writer = SpyralWriter(output_path, config, 1_000_000_000)
+writer = SpyralWriter(output_path, config, 5_000)
 
 def main():
     run_simulation(
@@ -216,18 +216,18 @@ electronics = ElectronicsParams(
     shaping_time=1000,
     micromegas_edge=10,
     windows_edge=560,
-    adc_threshold=40,
+    adc_threshold=10,
 )
 
 pads = PadParams()
 
 config = Config(detector, electronics, pads)
 ```
 
-Note that by not passing any arguments to `PadParams` we are using the default pad description that is bundled with the package. See the [detector](./detector/index.md) guide for more details. For the output, we create a SpyralWriter object. This will take in the simulation data and convert it to match the format expected by the Spyral analysis. Note the final argument of the Writer; this is the maximum size of an individual file in bytes (here we've specified 1GB). The writer will then split our output up into many files, which will help later when trying to analyze the data with a framework like Spyral.
+Note that by not passing any arguments to `PadParams` we are using the default pad description that is bundled with the package. See the [detector](./detector/index.md) guide for more details. For the output, we create a SpyralWriter object. This will take in the simulation data and convert it to match the format expected by the Spyral analysis. Note the final argument of the Writer; this is the maximum size of an individual file in events (here we've specified 5,000 events). The writer will then split our output up into many files, which will help later when trying to analyze the data with a framework like Spyral.
 
 ```python
-writer = SpyralWriter(output_path, config, 1_000_000_000)
+writer = SpyralWriter(output_path, config, 5_000)
 ```
 
 Then, just like in the kinematics script we set up a main function and set it to be run when the script is processed

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "attpc_engine"
-version = "0.5.0"
+version = "0.6.0"
 description = "AT-TPC Monte-Carlo simulation engine"
 authors = [
     {name = "Gordon McCann", email = "gordonmccann215@gmail.com"},

src/attpc_engine/detector/writer.py

@@ -229,6 +229,8 @@ def write(self, data: np.ndarray, config: Config, event_number: int) -> None:
             f"cloud_{event_number}", data=spyral_format
         )
 
+        dset.attrs["orig_run"] = self.run_number
+        dset.attrs["orig_event"] = event_number
         # No ic stuff from simulation
         dset.attrs["ic_amplitude"] = -1.0
         dset.attrs["ic_multiplicity"] = -1.0