Small changes

.gitignore

@@ -1,3 +1,5 @@
 test.*
 *.pt
-__pycache__/
+__pycache__/
+data/**
+data/image/**

S0_PrepareDataset.py

@@ -17,11 +17,13 @@ def download_file(url, pos):
 
             total_size = int(r.headers.get("content-length", 0))
             block_size = 8192
-            with tqdm(total=total_size,unit="B", unit_scale=True, position=pos, desc=f"#{pos} {local_filename}", ncols=100) as p_bar:
+            with tqdm(total=total_size,unit="B", unit_scale=True, position=pos, 
+                      desc=f"#{pos} {local_filename}", ncols=100, leave=True) as p_bar:
                 with open(local_filename, 'wb') as f:
                     for chunk in r.iter_content(chunk_size=block_size): 
-                        p_bar.update(len(chunk))
                         f.write(chunk)
+                        with tqdm.get_lock():
+                            p_bar.update(len(chunk))
     return local_filename
 
 IMAGE_DIR = "data/image"

S2_TimberDataset.py

@@ -9,7 +9,7 @@ import pandas as pd
 # 3264 x 2448
 
 DATA_DIR = "data/image/train"
-labels = ["Aspidosperma polyneuron", "Araucaria angustifolia", "Tabebuia sp.", "Cordia goeldiana", "Cordia sp.", "Hura crepitans", "Acrocarpus fraxinifolius", "Hymenaea sp.", "Peltogyne sp.", "Hymenolobium petraeum", "Myroxylon balsamum", "Dipteryx sp.", "Machaerium sp.", "Bowdichia sp.", "Mimosa scabrella", "Cedrelinga catenaeformis", "Goupia glabra", "Ocotea porosa", "Mezilaurus itauba", "Laurus nobilis", "Bertholethia excelsa", "Cariniana estrellensis", "Couratari sp.", "Carapa guianensis", "Cedrela fissilis", "Melia azedarach", "Swietenia macrophylla", "Brosimum paraense", "Bagassa guianensis", "Virola surinamensis", "Eucalyptus sp.", "Pinus sp.", "Podocarpus lambertii", "Grevilea robusta", "Balfourodendron riedelianum", "Euxylophora paraensis", "Micropholis venulosa", "Pouteria pachycarpa", "Manilkara huberi", "Erisma uncinatum", "Vochysia sp."]
+labels = os.listdir(DATA_DIR)
 label2id = {label:id for id, label in enumerate(labels)}
 
 def compile_image_df(data_dir:str, split_at = 0.9)-> pd.DataFrame: