Spaces:
Sleeping
Sleeping
Abby Reynolds commited on
Commit ·
1476642
1
Parent(s): 42db9ab
added api and display for analyzing department for specific artworks
Browse files- .gitignore +2 -1
- main.py +21 -1
- met_api.py +61 -0
- requirements.txt +3 -1
.gitignore
CHANGED
|
@@ -1 +1,2 @@
|
|
| 1 |
-
venv
|
|
|
|
|
|
| 1 |
+
venv
|
| 2 |
+
final_project
|
main.py
CHANGED
|
@@ -1,5 +1,6 @@
|
|
| 1 |
import streamlit as st
|
| 2 |
-
|
|
|
|
| 3 |
# Page setup
|
| 4 |
st.set_page_config(page_title="The METrics", layout="centered")
|
| 5 |
|
|
@@ -62,6 +63,25 @@ st.markdown("""
|
|
| 62 |
</div>
|
| 63 |
""", unsafe_allow_html=True)
|
| 64 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 65 |
st.write("")
|
| 66 |
|
| 67 |
# Footer
|
|
|
|
| 1 |
import streamlit as st
|
| 2 |
+
import met_api
|
| 3 |
+
import pandas as pd
|
| 4 |
# Page setup
|
| 5 |
st.set_page_config(page_title="The METrics", layout="centered")
|
| 6 |
|
|
|
|
| 63 |
</div>
|
| 64 |
""", unsafe_allow_html=True)
|
| 65 |
|
| 66 |
+
# Department analytic
|
| 67 |
+
st.write("")
|
| 68 |
+
st.subheader("Departments for your search")
|
| 69 |
+
|
| 70 |
+
query = st.text_input("Search term (for images only)", value="cats", key="dept_query")
|
| 71 |
+
max_ids = st.slider("How many results to analyze", 20, 400, 150, 10, key="dept_max")
|
| 72 |
+
|
| 73 |
+
if st.button("Run department analytic", key="dept_run"):
|
| 74 |
+
with st.spinner("Fetching and tallying departments…"):
|
| 75 |
+
rows = met_api.department_counts(q=query, max_ids=max_ids)
|
| 76 |
+
|
| 77 |
+
if not rows:
|
| 78 |
+
st.info("No results found (or the API call failed). Try another term.")
|
| 79 |
+
else:
|
| 80 |
+
df = pd.DataFrame(rows, columns=["Department", "Count"])
|
| 81 |
+
# show a quick bar chart + table
|
| 82 |
+
st.bar_chart(df.set_index("Department"))
|
| 83 |
+
st.dataframe(df, use_container_width=True)
|
| 84 |
+
|
| 85 |
st.write("")
|
| 86 |
|
| 87 |
# Footer
|
met_api.py
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import requests
|
| 2 |
+
import random
|
| 3 |
+
|
| 4 |
+
URL = "https://collectionapi.metmuseum.org/public/collection/v1/"
|
| 5 |
+
|
| 6 |
+
'''
|
| 7 |
+
Need to add error handling
|
| 8 |
+
'''
|
| 9 |
+
|
| 10 |
+
# gets information on a single object
|
| 11 |
+
def get_object(objectID):
|
| 12 |
+
return requests.get(f"{URL}/objects/{objectID}").json()
|
| 13 |
+
|
| 14 |
+
# gets all the objects that have some sort of image
|
| 15 |
+
def get_objectsWithImages():
|
| 16 |
+
response = requests.get(f"{URL}search?hasImages=true&q=*").json()
|
| 17 |
+
total = response["total"]
|
| 18 |
+
objectIDs = response["objectIDs"]
|
| 19 |
+
return total, objectIDs
|
| 20 |
+
|
| 21 |
+
# gets the urls for random objects with images
|
| 22 |
+
def get_images(totalObjects, objectIDs, limit):
|
| 23 |
+
images = []
|
| 24 |
+
# grabbing extra in case a primary image is blank (Works best on small limits)
|
| 25 |
+
rand_indexes = random.sample(range(totalObjects), limit + 20)
|
| 26 |
+
for i in rand_indexes:
|
| 27 |
+
obj = get_object(objectIDs[i])
|
| 28 |
+
if obj["primaryImage"]:
|
| 29 |
+
images.append((obj["primaryImage"], obj["title"]))
|
| 30 |
+
if len(images) == limit:
|
| 31 |
+
break
|
| 32 |
+
|
| 33 |
+
return images
|
| 34 |
+
|
| 35 |
+
def department_counts(q="*", max_ids=200):
|
| 36 |
+
"""
|
| 37 |
+
Analytic: return a list of (department, count) for search results.
|
| 38 |
+
- q: search query (default '*' = anything)
|
| 39 |
+
- max_ids: cap how many object IDs to inspect (keeps it fast)
|
| 40 |
+
|
| 41 |
+
Uses the Met search endpoint (images only), then tallies the 'department'
|
| 42 |
+
field from each object's metadata.
|
| 43 |
+
"""
|
| 44 |
+
try:
|
| 45 |
+
resp = requests.get(f"{URL}search", params={"q": q, "hasImages": True}, timeout=15)
|
| 46 |
+
resp.raise_for_status()
|
| 47 |
+
ids = (resp.json().get("objectIDs") or [])[:max_ids]
|
| 48 |
+
except Exception:
|
| 49 |
+
return []
|
| 50 |
+
|
| 51 |
+
counts = {}
|
| 52 |
+
for oid in ids:
|
| 53 |
+
try:
|
| 54 |
+
obj = get_object(oid) # uses your existing helper
|
| 55 |
+
dep = obj.get("department") or "(unknown)"
|
| 56 |
+
counts[dep] = counts.get(dep, 0) + 1
|
| 57 |
+
except Exception:
|
| 58 |
+
continue
|
| 59 |
+
|
| 60 |
+
# return sorted (department, count) pairs, highest first
|
| 61 |
+
return sorted(counts.items(), key=lambda x: x[1], reverse=True)
|
requirements.txt
CHANGED
|
@@ -1 +1,3 @@
|
|
| 1 |
-
streamlit==1.48.1
|
|
|
|
|
|
|
|
|
| 1 |
+
streamlit==1.48.1
|
| 2 |
+
pandas
|
| 3 |
+
requests
|