forked from prl-mushr/mushr_rhc
-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
1 parent
59d9c7d
commit f560064
Showing
1 changed file
with
92 additions
and
21 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1,37 +1,108 @@ | ||
| from dis import dis | ||
| import numpy as np | ||
| import matplotlib.pyplot as plt | ||
| import seaborn as sns | ||
| import pandas as pd | ||
|
|
||
| # file_path = '/home/azureuser/hackathon_data/e2e_eval/GPTiros_e2e_8gpu_2022-02-17_v2/info.csv' | ||
| file_paths = ['/home/azureuser/hackathon_data_premium/e2e_eval/6L0p0/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval/12L0p0/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval/24L0p0/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval/6L0p01/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval/12L0p01/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval/24L0p01/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval/6L0p1/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval/12L0p1/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval/24L0p1/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval/6L0p5/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval/12L0p5/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval/24L0p5/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval/6L1p0/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval/12L1p0/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval/24L1p0/info.csv'] | ||
|
|
||
| for file_path in file_paths: | ||
| # file_paths = ['/home/azureuser/hackathon_data_premium/e2e_eval/6L0p0/info.csv', | ||
| # '/home/azureuser/hackathon_data_premium/e2e_eval/12L0p0/info.csv', | ||
| # '/home/azureuser/hackathon_data_premium/e2e_eval/24L0p0/info.csv', | ||
| # '/home/azureuser/hackathon_data_premium/e2e_eval/6L0p01/info.csv', | ||
| # '/home/azureuser/hackathon_data_premium/e2e_eval/12L0p01/info.csv', | ||
| # '/home/azureuser/hackathon_data_premium/e2e_eval/24L0p01/info.csv', | ||
| # '/home/azureuser/hackathon_data_premium/e2e_eval/6L0p1/info.csv', | ||
| # '/home/azureuser/hackathon_data_premium/e2e_eval/12L0p1/info.csv', | ||
| # '/home/azureuser/hackathon_data_premium/e2e_eval/24L0p1/info.csv', | ||
| # '/home/azureuser/hackathon_data_premium/e2e_eval/6L0p5/info.csv', | ||
| # '/home/azureuser/hackathon_data_premium/e2e_eval/12L0p5/info.csv', | ||
| # '/home/azureuser/hackathon_data_premium/e2e_eval/24L0p5/info.csv', | ||
| # '/home/azureuser/hackathon_data_premium/e2e_eval/6L1p0/info.csv', | ||
| # '/home/azureuser/hackathon_data_premium/e2e_eval/12L1p0/info.csv', | ||
| # '/home/azureuser/hackathon_data_premium/e2e_eval/24L1p0/info.csv'] | ||
|
|
||
| file_paths = ['/home/azureuser/hackathon_data_premium/e2e_eval_models1/3L0p0/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval_models1/6L0p0/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval_models1/12L0p0/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval_models1/24L0p0/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval_models1/3L0p01/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval_models1/6L0p01/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval_models1/12L0p01/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval_models1/24L0p01/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval_models1/3L0p1/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval_models1/6L0p1/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval_models1/12L0p1/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval_models1/24L0p1/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval_models1/3L0p5/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval_models1/6L0p5/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval_models1/12L0p5/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval_models1/24L0p5/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval_models1/3L1p0/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval_models1/6L1p0/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval_models1/12L1p0/info.csv', | ||
| '/home/azureuser/hackathon_data_premium/e2e_eval_models1/24L1p0/info.csv'] | ||
|
|
||
| all_vals_mean = np.zeros(shape=(4*5)) | ||
| all_vals_median = np.zeros(shape=(4*5)) | ||
|
|
||
| for i, file_path in enumerate(file_paths): | ||
|
|
||
| data = np.genfromtxt(file_path, delimiter=',') | ||
|
|
||
| distances = data[:,0] | ||
| times = data[:,1] | ||
|
|
||
| # distances = data[100:,0] | ||
| # times = data[100:,1] | ||
|
|
||
| # clear data that crashes immediately | ||
| min_time = 0.0 | ||
| distances = distances[times>min_time] | ||
| times = times[times>min_time] | ||
| min_time = 20.0 | ||
| condition = times>min_time | ||
| distances = distances[condition] | ||
| times = times[condition] | ||
|
|
||
| # clear data above episode max | ||
| max_dist = 1000 | ||
| condition = distances<max_dist | ||
| distances = distances[condition] | ||
| times = times[condition] | ||
|
|
||
| plt.hist(distances, bins=30, facecolor='green', alpha=0.75) | ||
| # plt.savefig('/home/azureuser/hackathon_data/e2e_eval/GPTiros_e2e_8gpu_2022-02-17_v2/fig.png') | ||
| # plt.savefig('/home/azureuser/hackathon_data/e2e_eval/model_test/fig.png') | ||
| plt.savefig('/home/azureuser/hackathon_data_premium/e2e_eval_models1/model_test/fig{}.png'.format(str(i))) | ||
| plt.show() | ||
| plt.clf() | ||
|
|
||
| if distances.shape[0]>0: | ||
| all_vals_mean[i] = np.mean(distances) | ||
| all_vals_median[i] = np.median(distances) | ||
| else: | ||
| all_vals_mean[i] = 0. | ||
| all_vals_median[i] = 0. | ||
|
|
||
| # print the stats: | ||
| print("SIZE: {} | AVG: {} | MED: {} | PATH: {}".format(distances.shape[0], np.mean(distances), np.median(distances), file_path)) | ||
|
|
||
| # 'Num tokens': ['540', '30K', '300K', '1.5M', '3M'] | ||
|
|
||
| # data = {'3L': all_vals_mean[::4], | ||
| # '6L': all_vals_mean[1::4], | ||
| # '12L': all_vals_mean[2::4], | ||
| # '24L': all_vals_mean[3::4], | ||
| # # 'Dataset fraction': [0.0, 0.01, 0.1, 0.5, 1.0], | ||
| # 'Dataset fraction': ['540', '30K', '300K', '1.5M', '3M']} | ||
|
|
||
| data = {'3L': all_vals_median[::4], | ||
| '6L': all_vals_median[1::4], | ||
| '12L': all_vals_median[2::4], | ||
| '24L': all_vals_median[3::4], | ||
| # 'Dataset fraction': [0.0, 0.01, 0.1, 0.5, 1.0], | ||
| 'Dataset fraction': ['540', '30K', '300K', '1.5M', '3M']} | ||
|
|
||
| # sns.lineplot(data=data, x="Dataset fraction", y=['6L', '12L', '24L']) | ||
|
|
||
| df = pd.DataFrame(data) | ||
| print(df) | ||
| dfm = df.melt('Dataset fraction', var_name='cols', value_name='Average meters traveled') | ||
| sns.catplot(x="Dataset fraction", y="Average meters traveled", hue='cols', data=dfm, kind='point') | ||
| plt.savefig('/home/azureuser/hackathon_data_premium/e2e_eval_models1/model_test/all_plots.png') |