Major update. Support for 15 LLMs, World Flora Online taxonomy validation, geolocation, 2 OCR methods, significant UI changes, stability improvements, consistent JSON parsing

api_cost/api_cost.yaml

@@ -7,7 +7,13 @@ GPT_4_32K:
 GPT_4:
   in: 0.03
   out: 0.06
-GPT_4_TURBO:
+# GPT_4_TURBO: ###############
+#   in: 0.01
+#   out: 0.03
+GPT_4_TURBO_0125:
+  in: 0.01
+  out: 0.03
+GPT_4_TURBO_1106:
   in: 0.01
   out: 0.03
 GPT_3_5_INSTRUCT:
@@ -24,7 +30,10 @@ AZURE_GPT_4_32K:
 AZURE_GPT_4:
   in: 0.03
   out: 0.06
-AZURE_GPT_4_TURBO:
+AZURE_GPT_4_TURBO_1106:
+  in: 0.01
+  out: 0.03
+AZURE_GPT_4_TURBO_0125:
   in: 0.01
   out: 0.03
 AZURE_GPT_3_5_INSTRUCT:

app.py

@@ -1561,7 +1561,6 @@ def content_project_settings(col):
             st.session_state.config['leafmachine']['project']['dir_output'] = st.text_input("Output directory", st.session_state.config['leafmachine']['project'].get('dir_output', ''))
         
 
-# @st.cache_data
 def content_llm_cost():
     st.write("---")
     st.header('LLM Cost Calculator')
@@ -1596,7 +1595,6 @@ def content_llm_cost():
         n_img = st.number_input("Number of Images", min_value=0, value=1000, step=100)
 
     # Function to find the model's Input and Output values
-    @st.cache_data
     def find_model_values(model, all_dfs):
         for df in all_dfs:
             if model in df.keys():

vouchervision/tool_wikipedia.py

@@ -2,7 +2,8 @@ import itertools, wikipediaapi, requests, re, json
 from langchain_community.tools import WikipediaQueryRun
 from langchain_community.utilities import WikipediaAPIWrapper
 # from langchain_community.tools.wikidata.tool import WikidataAPIWrapper, WikidataQueryRun
-
+import cProfile
+import pstats
 
 class WikipediaLinks():
 
@@ -370,13 +371,13 @@ class WikipediaLinks():
 
         self.info_packet['WIKI_TAXA']['DATA'].update(self.get_taxonbar_data(page.title))
 
-        for back in page.backlinks:
-            back = self.sanitize(back) 
-            if ':' not in back:
-                link = self.sanitize(self.get_wikipedia_url(back))
-                if link not in links:
-                    links.append(link)
-                    self.info_packet['WIKI_TAXA']['LINKS'][back] = link
+        # for back in page.backlinks:
+        #     back = self.sanitize(back) 
+        #     if ':' not in back:
+        #         link = self.sanitize(self.get_wikipedia_url(back))
+        #         if link not in links:
+        #             links.append(link)
+        #             self.info_packet['WIKI_TAXA']['LINKS'][back] = link
 
 
     def extract_info_geo(self, page, opt=None):
@@ -577,5 +578,16 @@ if __name__ == '__main__':
     "minimumElevationInMeters": "",
     "maximumElevationInMeters": ""
     }
-    Wiki = WikipediaLinks()
-    info_packet= Wiki.gather_wikipedia_results(test_output)
+    do_print_profiler = True
+    if do_print_profiler:
+        profiler = cProfile.Profile()
+        profiler.enable()
+    
+    Wiki = WikipediaLinks('D:/D_Desktop/usda_pdf/test.json')
+    info_packet= Wiki.gather_wikipedia_results(test_output)
+
+    if do_print_profiler:
+        profiler.disable()
+        stats = pstats.Stats(profiler).sort_stats('cumulative')
+        stats.print_stats(50)
+