folivo/PINA
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update Tensorboard use

Browse Source
This commit is contained in:
Matteo Bertocchi
2025-03-11 12:01:06 +01:00
committed by Nicola Demo
parent b38b0894b1
commit 67a2b0796c
6 changed files with 58 additions and 137 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
tutorials/tutorial1/tutorial.ipynb vendored
View File

@@ -505,7 +505,7 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": 10, "execution_count": null,
"id": "fcac93e4", "id": "fcac93e4",
"metadata": {}, "metadata": {},
"outputs": [ "outputs": [
@@ -546,10 +546,8 @@
} }
], ],
"source": [ "source": [
"# Load the TensorBoard extension\n", "print('\\nTo load TensorBoard run load_ext tensorboard on your terminal')\n",
"%load_ext tensorboard\n", "print(\"To visualize the loss you can run tensorboard --logdir 'tutorial_logs' on your terminal\\n\")"
"# Show saved losses\n",
"%tensorboard --logdir 'tutorial_logs'"
] ]
}, },
{ {

8
tutorials/tutorial1/tutorial.py vendored
View File

@@ -261,13 +261,11 @@ plt.legend()
# The solution is overlapped with the actual one, and they are barely indistinguishable. We can also take a look at the loss using `TensorBoard`: # The solution is overlapped with the actual one, and they are barely indistinguishable. We can also take a look at the loss using `TensorBoard`:
# In[10]: # In[ ]:
# Load the TensorBoard extension print('\nTo load TensorBoard run load_ext tensorboard on your terminal')
get_ipython().run_line_magic('load_ext', 'tensorboard') print("To visualize the loss you can run tensorboard --logdir 'tutorial_logs' on your terminal\n")
# Show saved losses
get_ipython().run_line_magic('tensorboard', "--logdir 'tutorial_logs'")
# As we can see the loss has not reached a minimum, suggesting that we could train for longer! Alternatively, we can also take look at the loss using callbacks. Here we use `MetricTracker` from `pina.callback`: # As we can see the loss has not reached a minimum, suggesting that we could train for longer! Alternatively, we can also take look at the loss using callbacks. Here we use `MetricTracker` from `pina.callback`:

56
tutorials/tutorial2/tutorial.ipynb vendored
View File

File diff suppressed because one or more lines are too long

7
tutorials/tutorial2/tutorial.py vendored
View File

@@ -311,12 +311,11 @@ trainer_learn.train()
# Let us compare the training losses for the various types of training # Let us compare the training losses for the various types of training
# In[10]: # In[ ]:
# Load the TensorBoard extension print('To load TensorBoard run load_ext tensorboard on your terminal')
get_ipython().run_line_magic('load_ext', 'tensorboard') print("To visualize the loss you can run tensorboard --logdir 'tutorial_logs' on your terminal")
get_ipython().run_line_magic('tensorboard', "--logdir 'tutorial_logs'")
# ## What's next? # ## What's next?

99
tutorials/tutorial3/tutorial.ipynb vendored
View File

File diff suppressed because one or more lines are too long

13
tutorials/tutorial3/tutorial.py vendored
View File

@@ -194,12 +194,11 @@ trainer.train()
# Let's now plot the logging to see how the losses vary during training. For this, we will use `TensorBoard`. # Let's now plot the logging to see how the losses vary during training. For this, we will use `TensorBoard`.
# In[5]: # In[ ]:
# Load the TensorBoard extension print('\nTo load TensorBoard run load_ext tensorboard on your terminal')
get_ipython().run_line_magic('load_ext', 'tensorboard') print("To visualize the loss you can run tensorboard --logdir 'tutorial_logs' on your terminal\n")
get_ipython().run_line_magic('tensorboard', "--logdir 'tutorial_logs'")
# Notice that the loss on the boundaries of the spatial domain is exactly zero, as expected! After the training is completed one can now plot some results using the `matplotlib`. We plot the predicted output on the left side, the true solution at the center and the difference on the right side using the `plot_solution` function. # Notice that the loss on the boundaries of the spatial domain is exactly zero, as expected! After the training is completed one can now plot some results using the `matplotlib`. We plot the predicted output on the left side, the true solution at the center and the difference on the right side using the `plot_solution` function.
@@ -335,12 +334,12 @@ plt.figure(figsize=(12, 6))
plot_solution(solver=pinn, time=1) plot_solution(solver=pinn, time=1)
# We can see now that the results are way better! This is due to the fact that previously the network was not learning correctly the initial conditon, leading to a poor solution when time evolved. By imposing the initial condition the network is able to correctly solve the problem. We can also see using Tensorboard how the two losses decreased: # We can see now that the results are way better! This is due to the fact that previously the network was not learning correctly the initial conditon, leading to a poor solution when time evolved. By imposing the initial condition the network is able to correctly solve the problem. We can also see how the two losses decreased using Tensorboard.
# In[11]: # In[ ]:
get_ipython().run_line_magic('tensorboard', "--logdir 'tutorial_logs'") print("To visualize the loss you can run tensorboard --logdir 'tutorial_logs' on your terminal")
# ## What's next? # ## What's next?